The State of the Smartphone Market in Q3 2016

You’ve probably seen headlines over the last couple of weeks about Apple accounting for over 100% of the profits in the smartphone business in the last quarter. I’m never a fan of that particular metric because it excludes all the smartphone vendors that don’t publicly report their financials, including several of the largest. Instead, I prefer to drill down a bit and look at what’s really happening beneath the surface, both in terms of shipments and in terms of the financials associated with those shipments.

Shipments – the big three steady, lots of movement below

When it comes to shipments, we’re seeing a fairly clear pattern emerging – the big three globally have been steady for some time now but there’s lots of movement below that. The big three are Samsung, Apple, and Huawei (in that order) and they’ve been in those positions for the last six quarters. Depending on who you believe, Samsung may have briefly ceded its number one spot to Apple two years ago when the iPhone 6 launched, but it’s safe to say Samsung has been consistently in that spot for a very long time. Apple has also been the consistent number two for a long time, while Huawei only ascended consistently to the number three spot last year. There are big gaps between these three, though, with Samsung typically tens of millions ahead of Apple in terms of shipments (except for fourth quarters) and Apple, in turn, 10-20 million ahead of Huawei except in the fourth quarter, when Apple sells far more than in the other quarters.

However, what’s become really interesting to watch over the last couple of years is all the turmoil in the next few spots in the smartphone rankings. The two biggest vendors to look at are Oppo and Vivo, both from China, which have come seemingly out of nowhere to take the fourth and fifth spots, pushing others further down the ladder. Xiaomi, meanwhile, which has flirted with the number three and even the number two spots, has fallen down to fifth or worse. Perhaps more remarkably, of the top 10 vendors by shipments, seven are Chinese, with only Samsung, Apple, and LG the exceptions. Three countries now make up the entire top 10. But it’s also worth noting that three of the largest vendors – Oppo, Vivo, and OnePlus – are all owned by the same company, BBK Electronics.

The chart below shows how rankings have changed over the past few years between the largest vendors.

Financial reporting and margins

Of course, shipments are only one metric and say nothing directly about finances. Yes, there’s something of a correlation between scale and profitability but it’s far from linear. Much depends on which segments these companies target, how differentiated their offerings are, and the selling prices of their devices. Those targeting the premium segment tend to have far higher margins – especially at scale – than those targeting the mid-market or low end. Though Apple is exclusively focused on the premium market and Samsung sells its Galaxy S and Note offerings into that market too, much of the volume among these top ten is very much low-end stuff. As a result, margins for even the largest players are likely to be low, especially since they’re mostly selling relatively undifferentiated Android phones.

Sadly, we don’t have financial reporting for many of these companies, a number of which are privately held or subsidiaries of larger companies (or both). However, margins for these mobile businesses and, in Apple’s case, its overall business are shown in the chart below:


As you can see, there’s been a fairly clear dichotomy between the two makers that sell premium smartphones at scale and the rest of the market. As I wrote about previously, Samsung saw a big dip in margins in its mobile business this past quarter due to the Note7 recall but, in general, is well ahead of most of the other vendors. However, it’s also well behind Apple’s overall margins, which are relatively representative (though perhaps a little low) of its iPhone business. This past quarter, HTC, LG, and Lenovo were all in the red and several of them have been so for some time. Sony bounced back into the black recently, having spent most of its time as an independent company following Ericsson’s exit from the business generating losses. It is finally now turning things around thanks to its focus on the premium segment, which has resulted in lower shipments but higher margins. LG, who had generated some positive momentum in 2015, has slipped further behind again and is back in the red. HTC continues to be only marginally in the market, with very low shipments and significant losses over the past couple of years.

A preview of Q4

I’ve focused here mostly on the historical picture but it’s worth thinking briefly about next quarter, too. We might just see Apple pip Samsung to the number one spot this quarter, as Apple rebounds from a year of shrinking sales and Samsung sees a continued lull following the Note7 fiasco. Behind them, the next three spots are likely to remain unchanged and I suspect the broad financial picture will remain fairly consistent as well. The big question for some of these vendors is how long can they remain in the market with, in some cases, substantial losses. As the upstart Chinese vendors continue to take share of the Android market, some of the more established vendors will definitely need to consider following Sony’s example and refocus on the premium market, or get out entirely. It’s going to be almost impossible for them to compete in a race to the bottom.

Painting a Clearer Picture of the Global Smartphone Market

Oftentimes, our peers in the analyst community put out press releases on category and vendor market share numbers and the true insight of the market gets lots in a set of very generic statistics. I understand why they do this. They make them fuzzy on purpose so people inquire about their data and pay them for the broader cuts. My frustration with this approach is it almost always leads the press astray with a headline because the media rarely understands how to ask the deeper questions. So a headline like “Nobody is buying an Apple Watch” gets published when it is total nonsense. I cleared this up in this piece where I outlined why we are revising some parts of our model for wearables going forward. That was just one example (of many) that happens when market share numbers get reported. In an ever increasing effort to more thoroughly educate our readers, I’ll make this point. Statistics are not insights. The insight comes from how those statistics are interpreted and the deep truths of what is happening in the market and why are communicated. Prepare yourself for a flood of similar headlines around the smartphone market.

I’d like to make a few generic smartphone market points then focus the rest of the article on what is happening in China, since that is where I foresee most of the disingenuous data will come from.

Globally, there are only six brand names worth keeping an eye on right now. They are, in no particular order: Samsung, Apple, Huawei, Xiaomi, Oppo, and Vivo (I am leaving LeEco out for the moment until I see where they are in six months). Oppo and Vivo are subsidiaries of a larger Chinese company called BBK, who also has a financial interest in OnePlus. Only Samsung, Apple, and Huawei sell more than 100m smartphones per year and it remains to be seen if any other brand can cross the 100m annual threshold.

This chart, which I have adapted from Morgan Stanley research, paints what I think is the best look at understanding a critical point about the smartphone market — pricing.


As you can see, the chart paints a picture of who makes money and who barely does, even if they ship large volumes of smartphones each quarter. This also tells you where their focus on the market is and thus, where their growth can come from.

Apple sells annually more than any other brand in the premium tier of the market. They have a near monopoly on the high-end and that is extremely unlikely to change anytime soon, if ever. Only Samsung should even be considered in the conversation for high-end smartphone sales. For them, that equates to around 60-70 million high-end phones each year (potentially declining) to Apple’s 180-190m high-end phones sold annually (potentially increasing). For Apple, in nearly all the big markets where they compete, they have nearly saturated the number of customers who can afford north of $500. So, their cycles are now largely replacement driven or driven by switchers. Either way, it’s a slower growth curve than when they were adding tons of new users. This is why analyzing Apple’s growth prospects now largely depends on understanding replcement cycles, payment plans, and other dynamics of replacement markets.

Interestingly, it appears global smartphone shipments may be up in 2016 which is contra to what most were forecasting early on. I attribute this to a much more predictable upgrade cycle in the developed markets than what was anticipated in early 2016. This, however, is leading many to believe 2017 may be more of a down year and most consensus data we see has people shifting the dynamics that were leading to negativity toward 2016 to 2017. Then again, look for accelerated growth overall for the category in 2018.

When it comes to a big key market like the US, we are seeing what looks like a bigger upgrade cycle this fall than originally thought. Which means it is possible next year’s cycle is not as strong. For Apple, the idea of a 2017 super cycle may be more myth than reality. Right now, China is the most interesting market to talk about as some new developments are starting to take shape.

What will get missed in a lot of the China market share forecasts is the biggest factor changing the shape of market share is tier 3-6 (the lower income tiers of China) are starting to contribute to the growth of China as a whole. When you look at brands I mentioned before like Oppo and Vivo, which look to having impressive gains, they are acquiring those gains given their strategy to attack the tier 3-6 parts of China with offline stores and nicely designed phones and specs in the $200-$300 range which is largely considered mid-range nowadays.

The rise of Oppo and Vivo are showcase examples of a major theme about how developing a strategy for the next 1-2 billion consumers who are yet to get smartphones in the least developed parts of the world is a completely different strategy than how you go after the current most profitable billion plus consumers with more mature and refined interests and greater disposable income. Only a handful of companies can do this and globally I think Huawei is building this strategy as they look to Africa and other continents with hundreds of millions of people yet to get a smartphone. This dynamic will likely boost the market share of certain vendors but it will be important to know what price point they are shipping in volume as it will be in the lower tier. Apple is clearly not going after this audience and it is unknown how Samsung will go after them. Both of these companies seem to be hunkering down to go after customers who don’t want to compromise and that self-identification only comes with time as consumers themselves learn what they want.

An interesting example of this is in India where it is clear Apple is not the favorite (due to their high prices) to see a big share gain any time soon. However, in a recent study my firm conducted in India, we asked which brand consumers felt was the market leader. Overwhelmingly, 61% of consumers in India said Apple was the clear market leader citing a range of reasons, mostly around software quality, design, etc. They acknowledge Apple is the best and said Apple is the phone they WANT to buy. Yet, they still are not going to, for a variety of reasons our research indicates Apple can solve.

This is a prime example of my thesis that, as portions of consumers in immature markets mature, they will slowly gravitate toward Apple. Even if the growth happens incrementally, I’m convinced Apple will still grow their unique consumer base, which could hit a billion users in the not too distant future.

There will always be tiers of smartphones but I maintain the strongest fuel for growth is brand. We can bet on those who have strong global brands. Companies like Xiaomi, Oppo, and Vivo are attempting to make a global brand but they will live or die on their success to become a brand consumers trust and love.

There is no question China is a very big market where we will see many companies attempt to create a global brand. Right now, Samsung, Apple, and Huawei remain the most fundamentally sustainable global brands, with a variety of different market nuances helping each independently. The main story to watch is if Vivo, Oppo, Xiaomi, and perhaps LeEco can break free from the cutthroat China market and become global brands. This is the hill they have to climb and it is going to be more costly than I believe they anticipate.

One last point on pricing. I maintain that, if you start in the low-end/low mid-range, it will become nearly impossible to ever go upstream too far. Once you sell a cheap phone, I think a “cheap” stigma will always be on the minds of consumers and will be very hard for those brands to compete in segments with Apple and Samsung. This is still a thesis but we have yet to see evidence it is not true in computing segments. I’m happy to be proven wrong and will point it out as a key case study if it happens.

I leave you with my estimates for Q4. While the line looks bad for Samsung, our research indicates it is likely they can recover from this as most of their customers will remain loyal. It could impact their growth prospects, however, which is a key dynamic to keep an eye on.


The Smartphone Market in Q1 2016

Now that Lenovo has reported its Q1 2016 financial results, we have a reasonably full picture of how the global smartphone market performed in the quarter. While many of the largest vendors report their performance directly as public companies, some of the medium-sized and smaller vendors are privately held, so we need to rely on third party data from analyst firms such as IDC, Strategy Analytics, and Gartner to fill in some of the gaps. All of this leaves us with some interesting insights into the changing dynamics of the smartphone market.

Small legacy vendors continue to decline

Some of the longest-standing players in the market continue to see serious declines over time in their smartphone shipments. Among this group are BlackBerry, HTC, and Sony, as well as the former Nokia smartphone business now owned by Microsoft. This group, along with Motorola (now part of Lenovo), has at different times made up the top five smartphone vendors in the world but each now ships well under five million smartphones per quarter, as shown in the chart below:

Small legacy vendors

That’s quite a fall from grace for each of this group and it’s a useful reminder of how much fortunes can change in this market despite very significant share and volumes. At its peak, BlackBerry shipped over 40 million phones a year yet, over the past four quarters, it shipped under 4 million. Nokia led the global smartphone market for a number of years but Microsoft’s smartphone business shipped just 21 million devices over the last four quarters and just 6.8 million in the last two. Sony had looked like it was doing better than most of its medium-sized Android competitors for some time, seeing growth in shipments from 2011 to 2014, but over the last year, things have turned sharply downward as it refocused its business on premium, higher-margin devices.

Dynamics are shifting among the larger players

Among the larger players, dynamics and market shares have shifted fairly dramatically over the last few years, though there’s been relative stability among the top three, as the chart below shows:

Large vendors

Samsung and Apple continue to occupy the top two spots, with little sign this is likely to change. Though Apple very briefly matched Samsung’s shipments in its big iPhone 6 quarter at the end of 2014, the gap between the two has widened again since, as Apple’s growth has fallen off and Samsung’s has recovered slightly. Meanwhile, Huawei has left the rest of the pack behind as the number three vendor, leaving a handful of others clustered together vying for fourth place. Interestingly, two new Chinese vendors – Oppo and Vivo – have rocketed into top spots over the last year or so, with Oppo claiming fourth in Q1 2016 for the first time according to several analyst firms and Vivo taking sixth. Meanwhile, Lenovo (including Motorola) went through a period of very strong growth driven both by Lenovo’s domestic success and the acquisition of Motorola, only to see its domestic business fall off a cliff over the past year, with the Motorola business unable to make up the difference. As such, Lenovo dropped from fourth place in late 2015 to ninth this past quarter. Xiaomi, too, which had been a darling of the tech press for some time and put out some ambitious forecasts in the past, has seen its star fade somewhat over the past year as its shipment growth has plateaued.

A flat market overall

I’ve written a couple of pieces in this space over the last month or two about the flattening and declining US smartphone market but it’s worth taking a step back and looking at the global picture too. Total global smartphone sales

As you can see, there was a period of very rapid growth for most of this chart, rising from a little over 100 million at the end of 2011 to just over 400 million in the fourth quarter of last year. However, you can also see a dramatic slowdown over the last six quarters or so, as growth largely evaporated from the market. Q1 2016 hit around 340 million units, as did Q1 2015 a year earlier, so there was essentially zero growth over that year. Given the market had grown fairly consistently by 70-100 million each year until that point, that’s a major change. Obviously, the main reason is the increasing saturation of mature smartphone markets and the ensuing slowing of growth there. Even big growth markets like China have seen this deceleration over the last year or so and, as I’ve discussed previously, total shipments in the US are now shrinking. This is likely the shape of things to come over the upcoming years, which will make growth more challenging for individual vendors unless they’re able to win significant share or tap into the remaining growth in emerging markets.

Margins continue to be terrible for most

Third party analyst firms don’t publish financial figures for the major vendors in the same way they do shipments, so we’re left with more gaps in the data here. But from those who do report publicly, we can still draw some fairly strong conclusions. What is clear is, though Apple and Samsung are consistently profitable in their smartphone businesses, hardly anyone else is. Apple doesn’t report its margins for iPhones directly but consensus suggest it is in the 30% range, while Samsung has dropped from its peak in the high teens but managed a 14% operating margin on its mobile business in Q1. The chart below shows numbers for these and other vendors, with Apple’s overall operating margin but divisional operating margins for the others:

Smartphone margins

As you can see, all the others are fairly consistently unprofitable and, in some cases, severely so. As shipments have declined for some of the smaller vendors in particular, we’ve seen margins decline with them. Even in the better times, few of these companies managed margins above 10%, with HTC a rare exception thanks to its focus on premium smartphones. HTC’s margins have consistently worsened (and fallen off the bottom of the chart in Q1 2016), while Sony’s have also been very poor for several recent quarters. LG had appeared to be doing better but has dropped into the red recently, while Lenovo continues to push back the date for its future profitability in its mobile business.

A grim market for most

Between the disappearing growth in the overall market and the stiffening competition among the medium and small vendors, especially in Android, this has become a grim market for most players. Only Apple and Samsung among the publicly traded companies are generating meaningful margins from smartphones, while most of the others are losing money and share to rapidly growing Chinese vendors. Even some of the Chinese vendors who had done well in the past, such as Lenovo and Xiaomi, have struggled to keep up and are suffering as a result. We’re going to continue to see new names pop up in the top vendor charts, while familiar names that had seemed dominant continue to fall off. It should be fascinating to watch.

Peak Smartphone in the US?

I wrote previously about the slowing smartphone upgrade rate in the US. Today, I want to tackle the other side of the smartphone sales coin — slowing growth in the US smartphone base. Together, these two factors will make life more challenging for companies trying to sell smartphones in the US in the coming years.

Each quarter, growth slows

I track figures for the major US wireless operators and, according to those numbers, the smartphone base has been growing less each quarter. If you look at it on a quarterly growth basis, the numbers jump around quite a bit because the market is fairly cyclical, but you can still discern a downward trend:

Quarterly smartphone base growth

If you look at the numbers on a year on year basis, the trend is even clearer:

Year on year smartphone growth

Each quarter for at least the last six quarters shown, the five major operators I track have added fewer smartphone subscribers year on year than the previous quarter. From a total growth of almost 28 million back in Q4 2014, they added just 18.8 million in Q1 2016.

Increasing saturation in postpaid

The obvious reason for the slowdown in growth is the increasing saturation of the postpaid base with smartphones. Sprint and T-Mobile have both crossed the 90% mark, while AT&T and Verizon are both in the 80s already in terms of penetration of their postpaid phone base. The aggregate penetration for these four carriers is shown in the chart below:

Postpaid base penetration

As you can see, we’re approaching 90% across the market and the slope of the curve has flattened over the past three years as the market approaches a state of saturation. Note, however, this is just the postpaid base and penetration rates on the prepaid side (where devices aren’t subsidized) is quite a bit lower. Hence, the slowdown in growth on the prepaid side has been less dramatic and prepaid smartphone sales have held up better in recent quarters than on the postpaid side.

Trends vary by carrier

As well as the postpaid/prepaid differences, there are also fairly different trends by carrier, as the chart below shows:

Postpaid smartphone base growth by carrier

At both Verizon and AT&T, the rate of growth has declined dramatically over the last two years, with a halving of growth at Verizon and an even more significant decline at AT&T, which added just 1.1 million postpaid smartphone customers over the past year, versus 5.2 million in a similar period in late 2013 and early 2014. Meanwhile, Sprint has returned to overall phone subscriber growth and, as a result, boosted its smartphone base growth to the extent it’s now adding more subscribers each year than AT&T, despite its much smaller base of overall customers. Then we have T-Mobile, which has been consistently growing its base by around 4 million a year for the past two years, almost matching Verizon’s total.

More headwinds for device vendors

Again though, the key point here is clear: as the upgrade cycle lengthens and as the growth in the smartphone base also slows, there will be two major headwinds for device vendors looking to sell smartphones in the US market. That market has already arguably peaked, with its biggest single quarter in Q4 2014 and declines over recent quarters, but we’re likely only two years or less away from an end to meaningful growth in the US smartphone base. That will bring big changes to the number of smartphones sold and this trend is likely to be repeated in many other mature markets with similar levels of smartphone adoption.

FBI v. Apple: “Unduly Burdensome”

You might well be asking yourself, if the FBI withdrew its challenge against Apple, then why are we still talking about FBI v. Apple? Well, the San Bernadino case is over, but there are many, many more cases still pending. The ACLU published an interactive map of locations where the FBI is currently using the All Writs Act to demand assistance from Tech companies. You can view it HERE. So yeah, this matter is far from over.

Author’s Note: For an outline of how I think FBI v. Apple will play out, please see HERE.

As I’ve written before, there are two big issues that the FBI must overcome before it can get a court to order Apple to undo its own encryption. The first issue is CALEA — a statute that appears, on its face, to prohibit the FBI from asking for the very thing they’re currently asking for. I discussed CALEA HERE.

Today I focus on the second big stumbling block facing the FBI. If the FBI is going to use the All Writs Act to order Apple to assist them in breaking their own encryption, the FBI must first demonstrate that the requested assistance is not “unduly burdensome”.

One Phone, One Time

The FBI has steadfastly insisted that this case is just about one phone, and that the action they are requesting of Apple would occur only once. That is a crock of manure.

So important is this issue, that I was going to devote an entire article to it. However, since the current case is over, let me just point out that the director of the FBI — during testimony — under oath — before Congress — said:

“(O)f course” the FBI would use the ruling from this case to “return to the courts in future cases to demand that Apple and other private companies assist . . . in unlocking secure devices.”

So much for ‘one phone, one time’.

But is it really all that important? Is the claim that this is just about ‘one phone, one time’ really that big of a deal?

Oh yeah.

In their pleadings, the FBI said that Apple “desperately needs” this case to be about more than just one phone.

Apple desperately wants—desperately needs—this case not to be “about one isolated iPhone.” ~ FBI Pleadings

Someone was desperate, all right, but it wasn’t Apple. The FBI knows that if they are forced to acknowledge that Apple is going to have to comply with similar requests over and over again, then they will also be forced to acknowledge that the burdens that Apple could be expected to endure will expand exponentially.

It was the FBI who desperately wanted — desperately needed — this case to be about one isolated iPhone. That is why they continue to defend their position even when the facts, logic, common sense and the testimony of their own director make their position indefensible. Zealous advocacy is to be commended. Purposefully misstating a material fact is to be condemned.

In just one week, the FBI’s gone from “just one phone” to sharing with the entire law enforcement community. Remember this for the next one. ~ Jonathan Ździarski (@JZdziarski)


This case was never about getting information from the phone. It was always about setting a legal precedent that would allow the FBI to force tech companies to build back doors to the FBI’s specifications.

“Ah,” you say, “you can’t prove that. You’re starting to sound like a conspiracy nut.”

Oh yeah? You know who else sounds like a conspiracy nut? Richard Clarke, former national security advisor and head of counter terrorism.

[The FBI] is not as interested in solving the problem as they are in getting a legal precedent,” Clarke said. “Every expert I know believes the NSA could crack this phone. They want the precedent that government could compel a device manufacturer to let the government in.

The FBI director is exaggerating the need for this, trying to build it up as an emotional case … It’s Jim Comey. And the Attorney General is letting him get away with it. ~ Richard Clarke

All the evidence is consistent with the “crackpot conspiracy theory” that the FBI has been systematically trying to compel firms to backdoor their own encryption. At this point, I would venture to say that you have to be a crackpot NOT to believe these theories.

Unduly Burdensome

Duty To Assist Law Enforcement

Even now, I don’t think people realize what this case is all about. Apple did nothing wrong here. They were just being asked to help law enforcement out. Our legal system allows that, but only to a very limited degree.

[pullquote]The government does not hold the general power to enlist private third parties as investigative agents[/pullquote]

While the government can, in some circumstances, require third parties to support law enforcement investigations — for example, by requiring them to produce relevant evidence or give truthful testimony — the government does not hold the general power to enlist private third parties as investigative agents.

Some typical examples of what citizens can be asked to do:

— Produce existing business records;
— Freeze assets and accounts;
— Turn over security footage.

Let’s examine that last example. The FBI can go to a store and ask them for their security footage, but that’s about it. They can’t ask the store owner to stay up all night filming a suspect, and they can’t ask the store owner to install additional cameras in his lunch room, bathroom and boardroom. All of that is way, way, way beyond the call of duty.

What the FBI is asking of Apple is way beyond the call of duty too.

Legal Buzzwords

The Courts have placed severe limitations on what law enforcement can and cannot ask third parties to do. Here are the kind of buzzwords that are seen when reading through the applicable case law:

Must not be “in any way burdensome”; “meager assistance”; “minimal effort”; “no costs will be incurred”; “require minimal uses of company resources”; “no disruption to its operations”; the absence of any conceivable “adverse effect”; “normal course of their business”; “must not adversely affect the basic interests of the third party or impose an undue burden”.

Perhaps now you begin to see how little law enforcement can demand of us, and how free we are to refuse those demands.

Public Utilities vs. Private Entities

Most of the cases cited by the FBI in support of their position concerned highly regulated public utilities, not private companies. And the Courts have gone out of their way to note that much more can be demanded of public utilities — such as phone companies — than of private parties.

Even in those cases where public utilities were required to assist law enforcement, the types of burdens the Courts imposed were not at all as onerous as the one being requested of Apple.

For example, the FBI claimed that the Courts compelled the Mountain Bell telephone company to do programming, so it would certainly be nothing new for the Courts to compel Apple to do the same.

But in 1979, when the Mountain Bell case occurred, “programming consisted of a technician — a single technician mind you — using a ‘teltypewriter’ and the entire process “t[ook] less than one minute” ~ Apple Pleadings

Here is an image of the type of device that law enforcement asked Mountain Bell to “program” in 1979:

Pasted Graphic 2

So yeah, not the same as asking a company to create a tailored operating system.


In the current case, we’re talking about messing around with someone’s proprietary intellectual property. To my knowledge that has never occurred under the All Writs Act before. Ever.

Let me repeat that: The Courts have never required a third-party to alter — more less degrade — their proprietary property in order to aid law enforcement.

Screwing around with someone’s proprietary property is not like asking them for security footage. It’s more akin to asking an author to rewrite portions of their book and then put that book up for sale under the author’s name. Similarly, what the FBI wants Apple to do is to rewrite portions of their security software and then put it up for sale under Apple’s imprimatur.


In fact, the requested action by the FBI is unprecedented at every level. Never before has law enforcement asked that such a burden be imposed under the All Writs Act.

For A Living

Proprietary? Unprecedented? The FBI shrugs these off, blithely responding ‘fiddily dee dee, writing a little software is not a burden for Apple. After all, they write software for a living.’

(I)t is not an unreasonable burden for a company that writes software code as part of its regular business.

(T)his case requires Apple to provide modified software, modifying an operating system—writing software code—is not an unreasonable burden for a company that writes software code as part of its regular business.

Oh yeah? As Apple pointed out in their pleadings, following the government’s thinking to its logical conclusion leads to absurd results.

(I)t would not be unreasonably burdensome to demand that Boeing build a custom jet for the government because Boeing builds planes as part of its regular business or to demand that a pharmaceutical company make drugs for executions after it has made the intentional decision not to. ~ Apple Pleadings

Just because Apple is in the business of building encryption software does not mean Apple is in the business of tearing down their encryption software any more than Boeing is in the business of building planes that are specially designed to crash.


The compromised operating system that the government demands would require significant resources and effort to develop. Although it is difficult to estimate, because it has never been done before, the design, creation, validation, and deployment of the software likely would necessitate six to ten Apple engineers and employees dedicating a very substantial portion of their time for a minimum of two weeks, and likely as many as four weeks. Members of the team would include engineers from Apple’s core operating system group, a quality assurance engineer, a project manager, and either a document writer or a tool writer. ~ Apple Pleadings

The costs for Apple to accede to the government’s request are extraordinary and unprecedented…

…but no one cares. Apple makes billions of dollars, so no one has any sympathy for them.

There are, of course, other, more long term, and more damaging, costs such as forcing Apple to violate their existing representations, and the harm that would be caused to Apple’s reputation, their global brand and their bottom line.

Again, hardly anyone outside of Apple cares about those costs either. “Whoop de doo,” they say. “Price of doing business.”

Except, of course, that’s dead wrong. Hurting your brand and your reputation and your product are NOT the price of doing business. On the contrary, they’re the penalty paid for doing your business very, very badly. And Law enforcement simply does not — at least not without statutory authority — have the power to command you to run your business badly.

It’s important to understand that the costs described above are just the beginning, not the end, of the burdens that the FBI’s request would place upon Apple. The truly oppressive costs would come in forms that few have adequately considered.


Standard forensic practice would require the code of any forensic tool Apple produces to be preserved for at least as long as it might be needed as evidence in court. We’re talking about years, and, with appeals, perhaps decades. So just forget about the idea of creating, then destroying, a software skeleton key as quickly as you made it. That’s out.

And, of course, the Apple engineers would have to testify at trial about the back door that they had created, otherwise, under the fourth amendment, the evidence would be inadmissible.

Essentially, the encryption breaking portion of Apple would become a permanent arm of the government’s forensic team.

This, of course, is materially different from merely asking a store owner to provide law enforcement with a copy of their security footage.


Apple would have to maintain an in-house team of engineers dedicated to hacking its own users and affirmatively undermining the company’s promised security measures. The engineers involved in this effort might be some of the very same engineers responsible for designing and building the security features in the first place. Can you imagine what an awkward position that would place them in? Everyone else at Apple would view them as the as saboteurs. They would be treated like pariahs and their job would make their lives a living hell.

A House Divided

It is extremely difficult to write bug-free code.

There are two ways to write error-free programs; only the third one works. ~ Alan Perlis

The suggestion that Apple would be able to program a back door without risking a major screwup is laughable. I mean, have you even MET programing?

QUESTION: Joe’s code has 20 bugs. If Joe fixes 2 bugs per hour for 8 hours, how many bugs does Joe’s code have now?


[pullquote]Improving security would be costly and dangerous[/pullquote]

Software bugs can interact with existing code in complex ways, creating unanticipated new paths for bypassing iPhone security and exploiting the phone. Purposefully creating vulnerabilities likely creates even more vulnerabilities and those can be pretty dangerous. That means every design choice Apple makes to improve device security entails, not only the foreseeable front-end costs of implementing it, but the unpredictable back-end costs of degrading that improved security. And that’s especially true in this situation, where Apple would have to create the code entirely by itself, and without the possibility of any outside security audit.

[pullquote]The smart thing would be to stop improving your encryption[/pullquote]

Have you considered the contradictory incentives that would create? What is the point of making your encryption better if you know that you are simultaneously required to break that encryption? You’re just making your life — and life of your co-workers — harder. Instead of simply asking whether new security measures are cost-effective to implement from a user’s perspective, engineers would need to evaluate whether they could justify the additional cost of being required to attack those measures too.

The only way to avoid unnecessary costs, unnecessary work, unnecessary danger, and unnecessary conflicts with co-workers would be to stop improving the encryption.


The effect on morale, for both the engineers and the company overall, would be devastating.

What Apple engineer is going to want to destroy the company’s encryption and make things worse for their customers? Is that even ethical? There’s already been talk of Apple engineers refusing to comply with such an order or resigning their positions.

And how is Apple supposed to keep up overall morale when employees all know that one part of the company is actively sabotaging the other, and all in order to make their product worse and to make their customers less safe?


As if it’s not enough that the government is forcing Apple to create govtOS, they’re also making Apple responsible for safeguarding it too.

Apple is being forced to make a nuclear weapon, then either take responsibility for guarding that weapon or destroy it and rebuild it later. ~ Jonathan Ździarski on Twitter

The FBI frames this burden as a favor since they’re “allowing” Apple to decide for itself whether they wish to share the requested software skeleton key or keep it in the safety of their own secure headquarters.

The Court’s Order is modest. It applies to a single iPhone, and it allows Apple to decide the least burdensome means of complying. ~ FBI Pleadings

Oh, thanks a bunch, FBI.

When the FBI says that it is “allowing” Apple to decide the least burdensome means of complying with their request, what they’re really saying is that they’re foisting the responsibility of solving this impossible task onto Apple. To paraphrase Pyrrhus (he, of the pyrrhic victory), if the FBI does Apple another such favor, Apple is ruined.

Apple v FBI debate remind anyone of Jurassic Park? “We want you to create mutant dinosaurs, but only for safe captivity on this one island.” ~ Jon Fortt on Twitter

When asked, during a Congressional hearing, about whether it would be difficult to safeguard govtOS, FBI Director Comey testified that he had “a lot of faith” that Apple could protect the code from falling into the wrong hands. How oh so very convenient for Comey and the FBI and how oh so very inconvenient for Apple — who has to do all the work and endure all the risk as well.

Once you’ve created code that’s potentially compromising, it’s like a bacteriological weapon. You’re always afraid of it getting out of the lab. ~ Michael Chertoff, co-author of the Patriot Act, US Secretary for Homeland Security under George W. Bush

The government says, ‘Hey, security is no big deal’.

(T)here is no reason to think that the code Apple writes in compliance with the Order will ever leave Apple’s possession. ~ FBI Pleadings

There is, in fact, EVERY reason to think that the requested code will leave Apple’s possession.

Apple would end up being responsible for the Hope diamond of security keys. ((Now that I think about it, the value of the Hope Diamond pales in comparison to the value of breaching Apple’s encryption.))

The code would be a major prize and actors would go to almost any length — including kidnapping — to obtain it.

It makes Apple employees targets of foreign governments, kidnappings, hacking, surveillance, blackmail, etc. ~ Jonathan Ździarski on Twitter

And who would these attackers be? The baddest of the bad. Hackers, cybercriminals, authoritarian governments such as China and Russia. Some of the best minds — with some of the worst intentions — would bend their efforts toward obtaining this newly created skeleton key.

[I]t may simply be impossible to keep the program from falling into the wrong hands. ~ NSA expert Will Ackerly

And since Apple would have to maintain each key, and since Apple would have to create and re-create the key thousands upon thousands of times a year, Apple’s burden would be constant and never-ending.

We strongly believe the only way to guarantee that such a powerful tool isn’t abused and doesn’t fall into the wrong hands is to never create it. ~ Apple Pleadings

A Skeleton Key

The government is also very mistaken in their claim that the crippled iOS it wants Apple to build can only be used on one iPhone.

The technical experts have warned us that it is impossible to intentionally introduce flaws into secure products—often called backdoors—that only law enforcement can exploit to the exclusion of terrorists and cyber criminals. ~ Congressman John Conyers

Once GovtOS is created, personalizing it to a new device becomes a simple process. If Apple were forced to create GovtOS for installation on the device at issue in this case, it would likely take only minutes for Apple, or a malicious actor with sufficient access, to perform the necessary engineering work to install it on another device of the same model. ~ Apple Pleadings

A signed firmware update that is not truly limited to a single device, even one created for legitimate forensic purposes, becomes like a ‘skeleton key’ for the entire class of devices. Moreover, the more often this tool is used, the greater the risk it will be stolen or otherwise disclosed. ~ Apple Pleadings

Apple wouldn’t be creating a single key to open a single lock. They would be creating a skeleton key that would be capable of opening a billion locks.

Just in case you didn’t get the irony, FBI now has a backdoor that isn’t restricted to a single device, like they insisted Apple could make. ~ Jonathan Ździarski (@JZdziarski)

A Footprint

As if it’s not bad enough that the requested skeleton key is likely to be stolen, security experts say that the mere act of creating the software would put at risk the privacy and integrity of the data stored on millions of iPhones worldwide.

[u]sing the software even once could give authorities or outsiders new clues to how Apple’s security features work, potentially exposing vulnerabilities that could be exploited in the future. ~ Brandon Bailey

[T]here is no way to make a backdoor that works only for this single phone—the process of creating the backdoor establishes a blueprint and workflow for compromising all iPhones. ~ Kalev Leetaru, Forbes Contributor

The danger is not whether the FBI submits one request or a thousand, it’s forcing Apple to create the tool. ~ Bruce Schneier, security technologist at Harvard University’s Berkman Center for Internet and Society

We’re not just talking iPhones here. Security experts fear that if Apple is forced to create a “key” to access one of the San Bernardino terrorists’ iPhones, then that technology will leave a “footprint” that cannot be erased. And that ‘footprint’ could provide a hacker with a path for attacking not just Apple, but the encryption of others as well.

My definition of an expert in any field is a person who knows enough about what’s really going on to be scared. ~ P. J. Plauger

We’ve come a long, long way from law enforcement merely asking a store owner for some security footage, right?

Does Not Exist

Another aspect of this case that hasn’t been receiving enough attention is the fact that the FBI is asking Apple to create something that does not exist. In all the government’s past requests for citizen assistance, never — NEVER — has the government asked someone to create something that didn’t already exist.

[pullquote]How exactly do you compel creativity and ingenuity?[/pullquote]

And we’re not talking about building a new chair or making a new dress either. We’re talking about writing code — something that’s not easy to create. Setting aside the First Amendment issues — which deserve an article of their own — how exactly do you compel creativity and ingenuity?

Apple is being asked to use their very best engineers — you know, the ones who were supposed to be making their encryption harder to break — to use their creative juices in an effort to break and degrade their encryption.

FBI is not only ordering Apple to perform surgery, they’re ordering them to invent a new medical procedure, and with no medical training. ~ Jonathan Ździarski on Twitter

Actually, it’s even more sinister than the above analogy implies. The FBI is ordering Apple to invent a new medical procedure that would undo prior surgical repairs and do their patient harm.

Apple Has A Compelling Interest Not To Comply


“Ah, so what,” says the government, “This is all Apple’s fault anyway. They brought this on themselves.”

This burden, which is not unreasonable, is the direct result of Apple’s deliberate marketing decision to engineer its products so that the government cannot search them, even with a warrant. ~ FBI Pleadings

Let’s just set aside for the moment that what Apple is doing is one hundred percent legal (and, according to CALEA, what the government is attempting to do is one hundred percent illegal). Apple’s refusal to dilute their encryption is a “marketing decision”…

…in roughly the same sense that not serving burgers garnished with sewage is a “marketing decision”. ~ Julian Sanchez on Twitter


Have you given any thought to which engineers Apple would have to use in order to break their own encryption? You should.

Apple has maybe 5 employees capable of writing the software. Doing this means not fixing some other vital bug. ~ Rob Graham ❄️ on Twitter

The FBI’s request would not just turn Apple’s best minds toward the task of breaking Apple’s encryption, it would also divert those self-same minds away from the all important task of making Apple’s encryption better.


If Apple can be forced to use their automatic updates to remove security features, it creates an incentive for customers to not update their devices. It’s in Apple’s best interests that customers update their operating systems as soon as possible and customers also benefit, not just from the features provided in updates but by security enhancements as well. The disincentive created by the FBI’s intrusion into Apple’s software update procedure would make the operating system open to even more security vulnerabilities.


Apple is not just petulantly refusing to honor the FBI’s request out of childish spite or due to a lack of patriotic fervor. Apple has a compelling interest in safeguarding the data protection systems that ensure the security of hundreds of millions of customers who depend upon, and store their most confidential data on, their iPhones. An order compelling Apple to create software that defeats those safeguards undeniably threatens those systems and adversely affects both Apple’s interests and the interests of iPhone users around the globe. The protections that the government asks Apple to compromise are the most security-critical software component of the iPhone—any vulnerability or back door, whether introduced intentionally or unintentionally, can represent a risk to all users of Apple devices simultaneously.

Apple is being asked to build a cruise ship that will flood just one customer’s compartment, without making the ship any less seaworthy.

In essence, the FBI is demanding that Apple re-write its own software code, degrade the security of their customers, and create potentially catastrophic risks to the security of users’ Apple devices. How is that not going to be construed as unduly burdensome?


In every other case where the Courts have compelled a company to assist law enforcement, they justified it by pointing out that the request did not require the company to do anything other than what it was already doing in its normal course of business anyway. For example, law enforcement can request security footage from a store because the store, in the normal course of business, installed a security camera and took, and kept, security footage. The burden of providing a copy to law enforcement is minimal.

That is not the case here. Not only is the FBI asking Apple to do something they would not do in the normal course of business, they’re asking Apple undo what they normally do.


Apple is being asked to take an action that is not only costly, not only NOT in the normal course of business, not only dangerous, but something that is antithetical to their business and something that is plainly “offensive to it.” N.Y. Tel. Co., 434 U.S. at 174.34

Apple is not required to sabotage its own products. On the contrary:

(Apple is) free to choose to promote its customers’ interest in privacy over the competing interest of law enforcement. ~ Magistrate Judge Ornstein

Buzzwords Redux

Let’s look at that list of legal buzzwords again:

Must not be “in any way burdensome”; “meager assistance”; “minimal effort”; “no costs will be incurred”; “require minimal uses of company resources”; “no disruption to its operations”; the absence of any conceivable “adverse effect”; “normal course of their business”; “must not adversely affect the basic interests of the third party or impose an undue burden”.

After re-reading the above, do you really think there is any reasonable way to construe the government’s request of Apple as anything but burdensome?

Third Party

One final time, I feel I need to re-remind everyone that Apple is not the bad guy here. Apple did nothing wrong. They didn’t break any laws. This is about law enforcement asking a third party — who is not engaged in any wrongful conduct — to not just take an action, but to take an action they don’t want to take and one that would be harmful to them.

The government is asking Apple to do them a favor, but what a favor. The FBI asking Apple to trash it’s own encryption is like your neighbor asking you to burn down your house so he can stay warm.

[pullquote]Consider the effect on small businesses[/pullquote]

And have you considered the effect the FBI’s request would have on companies not named Apple?

The government is desperately trying to maintain the ludicrous fiction that this is about one phone, one time, because it doesn’t want the Court to think about the very real world consequences of what would happen when their requests were made to not just Apple, but to all companies, everywhere, all the time. Unlike Apple, few companies have the resources necessary to comply with such requests and even fewer have the resources necessary to resist such requests. ((Two examples are Hush Mail (a Canadian company) and Lavibit. Both companies were devastated, and essentially forced to close their doors, simply because they could not afford to fend off government requests for their client’s data.))

The government’s requests would chill innovation and deter companies from entering the important field of encryption.


An FBI that asks Apple to break their encryption for the greater good is like a cannibal that asks a chef to teach him how to cook so he may better serve mankind. Sounds noble, but it’s just going to get us all in hot water.

The FBI is attempting to compel Apple to reengineer a product design solely to defeat the product’s purpose. Asking Apple to create that which does not already exist and which Apple does not want to create and which will harm the company now, and going forward, is the very definition of burdensome.

What the government is demanding of Apple is simply above and beyond what can be, and should be, demanded of a good citizen.

Samsung Gear VR – Training Wheels for Virtual Reality

One of the most iconic pictures of nerdom was taken at the last Mobile World Congress when thousands of people in the audience were captured wearing the Samsung Gear VR headsets. Although this scene has been lampooned and ridiculed as a geek-only experience, the fact is the Samsung Gear VR headset has become the training wheels of VR and giving people who use it a taste of what VR is all about and its potential.

I know that Google’s Cardboard blazed the trail in low-end VR experiences and, if you want to get technical about it, Cardboard got the consumer VR ball rolling. However, Samsung’s integration of a 75 Hertz optics, powered by the Oculus Store and the automatic upload of specific content has delivered a powerful approach to giving anyone who uses it a pretty good taste of what VR is about. If they have an analytical bone in their body, they can imagine how VR is a game changer for entertainment, business and education.

For that, I applaud Samsung’s forward thinking on this and for the foresight to kick start consumers interest in VR. But I also see it as laying the groundwork for what will be the best delivery method for VR. If you want a great VR experience today, you need to buy the Oculus or HTC Vive headset powered by a PC with a powerful graphics card. And it is a tethered experience.

But that approach is cumbersome at many levels. First, having to tie it to a PC to handle the processing and rendering to me is kind of a boat anchor. Add the fact it being tethered limits mobility, especially in games where the virtual experience begs for a lot of movement.

I believe that, for VR to really take off, especially beyond its use in gaming and vertical markets, the headsets have to be untethered and all of the processing power and intelligence needs to be in the VR goggles themselves. To an extent, that is what Samsung does with the Gear VR since it is drawing all of its power from the Galaxy smartphone. But at moment, smartphones are too underpowered to deliver the same kind of experiences one gets with higher end models like the Oculus Rift and HTC’s Vive.

The good news is, in talking with all of the major folks in the VR component and device business, they all understand that, for VR to really gain broad adoption, it has to be through an untethered headset. More importantly, at least one key semiconductor player has already optimized one of their low power processors for use in a VR headset where all of the power and intelligence is in the goggles themselves. And other major semiconductor companies are also working on similar processors.

The way I see the VR market developing is it will get its first boost through the gaming sector. VR is a game changer for this $90 billion dollar industry and, while the VR technology is OK today, it will only get better and greater gaming apps will emerge and bring even more people into the world of gamers. It will also gain serious traction in vertical markets.

In past columns, I have written about VR’s impact on travel, cruise lines and real estate. In the case of the cruise lines, the idea is to put prospective customers in the cabin they might want to book or to virtually explore the ship they are interested in cruising on. VR goggles could also be used to deliver VR-based armchair travel experiences. As for real estate, they want to be able to let a person interested in a house actually walk through it via a VR experience regardless of where they are in the world and use it to shorten the buying cycle.

The ad world sees it as a whole new storytelling medium. Last summer, the folks from Patron Tequila came to my office to show me a VR ad. It started by allowing me to walk the fields in Mexico where they cut the agave plants. It then transported me to their distilling facility to see how Tequila is made. Finally, they highlighted all three of their tequilas, discussing their flavor profiles, all in about 2 minutes. But the experience of viewing this in VR made the ad so memorable, I can still envision the whole thing today as if I had just seen it.

We know sports teams and broadcasters are looking at ways to use VR to make it possible to feel as if you are in the front row at a game. Also, the entertainment industry is starting to embrace VR for future storytelling content.

For the moment, both the gaming and vertical industries will tolerate a tethered VR experience. However, I am convinced these industries would prefer an untethered solution if it was available. I suspect over the next 2-3 years, gaming and vertical markets will be the prime users of VR solutions but I have to admit the Samsung Gear could be a sleeper that drives demand with a large group of consumers sooner rather than later.

Samsung’s Gear VR headset is selling well and reports by those consumers using it like what it offers. The Oculus store gets new VR apps each week and is broadening its reach through apps like Netflix and Hulu. In fact, I watch all of my Netflix shows on the Gear VR now.

The other unknown when it comes to VR adoption in the consumer market could be Apple. They have been conspicuously silent about VR but we all suspect they are working on their own VR headset and expect it to be untethered, tied to an iPhone and connected to Apple’s ecosystem of apps and services.

Should Apple enter the VR market with an iPhone-based solution, they could accelerate the adoption of VR in the consumers space. I doubt it would come to market this year but it would not surprise me if they offered their own solution in 2017.

The market for VR will be interesting to watch since it really is a game changer in terms of the way we can play games, visit places as if in person and interact with a VR world for play, work and education. It is too early to tell how big the VR market can be, but the signs point to VR being a disruptive technology that eventually could touch many markets around the world.

I believe that, for it to take off, the headsets must be untethered so the users could have more mobility and freedom to interact with these VR apps and services in unrestricted ways. In that form, and with lower prices, VR adoption could move faster than any of us can dream.

Is the iPhone Coke, New Coke, Pepsi or Just Sugar Water?

On January 10, 2016, long time subscriber and frequent commentator, Obarthelemy wrote:

User Experience is in the eye of the beholder. Until I see double-blind tests about it, I flatly deny that Apple’s is superior/premium…. ~ Obarthelemy

This really got me thinking. How would the Apple iPhone, Phones that ran Android, Windows Phone, etc. fare against one another in a double-blind test?

Like Carl Sagan, I’m a huge believer in the scientific method:

…the scientific method was the best method ever invented for arriving at the truth of things. ~ Carl Sagan

However, before we discuss whether the iPhone or Android Smartphones ((Smartphones that run the Android operating system.)) or any other smartphone would win in a double-blind test, we should first take a step back and ask ourselves whether a double-blind test is the best way to judge consumer preferences for smartphones — or consumer preferences for ANY product, for that matter.

Fortunately, we don’t have to guess. That question was asked and answered in the 1980s by the marketing campaign known as the Pepsi Challenge.

The Pepsi Challenge

The Pepsi Challenge (see 30 second commercial) marketing campaign of the 1980s was supposed to be a scientific inquiry; a double-blind experiment.

In a world overwhelmed with soda options, how could you really know which soda you liked best? It made sense to put prejudice and branding aside, wear a metaphorical blindfold and focus purely on the flavor of the various options. ~ Pepsi paradox: Why people prefer Coke even though Pepsi wins in taste tests

Here’s the thing: The Pepsi Challenge wasn’t just a marketing gimmick. It really is true that in blind taste tests people preferred the taste of Pepsi over Coke.

In fact, the Pepsi Challenge marketing campaign was so successful that Coke began a series of its own internal taste tests aimed at developing a superior product. The result was New Coke — a sweeter cola reformulated to be better than Pepsi and better than the classic formulation of Coke in blind taste tests.

The reaction to New Coke was not at all what the Coca-Cola company had expected. Regular Pepsi drinkers were underwhelmed. Regular Coke drinkers hated it.

The board of Coca-Cola then reversed itself, re-introduced the old formula under the brand name “Classic Coke”, and sold both New Coke and Classic Coke side-by-side. Over time, New Coke all but disappeared with Classic Coke, once again, taking its place as the company’s flagship product.

Today, despite the double-blind taste tests that showed that Pepsi was preferred over ‘Classic’ Coke, and New Coke was preferred over both Pepsi and Classic Coke, people buy far more Coke than Pepsi, and almost no one at all is interested in buying New Coke. ((According to industry statistics compiled by Beverage Digest, Coke owns 17 percent of the American market for carbonated soft drinks. The next most popular choice is Diet Coke with 9.4 percent. Pepsi languishes in third place at 8.9 percent.))

What’s the heck is going on here? If New Coke beats Pepsi in taste tests, then why is it less popular than Pepsi? If Pepsi wins taste tests against Coke, then why does Coke still dominate the soda market?

Hypothesis #1: Marketing Is All That Matters

Some industry observers contended that Coke’s ultimate success over Pepsi was proof that superior marketing wins out even over a superior product. Marketing, therefore, was all that really mattered and consumer companies should invest lots of money in advertising. But that explanation doesn’t really hold water.

If marketing is all that matters, then why didn’t Pepsi — which supposedly had the superior product — just improve its marketing? For that matter, if marketing is more important than product, why doesn’t every company just improve their marketing?

Half the money I spend on advertising is wasted, and the trouble is I don’t know which half. ~ A maxim of obscure origins, put in famous mouths

And if marketing were all that, then why wasn’t the Coca-Cola company able to sell New Coke? They devoted far more advertising dollars to New Coke than they had ever used to promote the former version of Coke, but New Coke — which was specifically formulated to beat both Coke and Pepsi in taste tests — went exactly nowhere.

When a man says there’s nothing that marketing can’t do, you know that the man has nothing to do with marketing.

Hypothesis #2: Sweet Sips

A second theory was that people preferred Pepsi in bind taste tests because people prefer sweeter tastes when sipping. And there is a factual foundation for this assertion. Even in blind taste tests of wine, people almost invariably preferred sweeter varieties. And no one is seriously contending that sweeter wines are always superior to other varieties of wine.

However, if people prefer sweet tastes when sipping, then why do the taste tests reverse themselves when the sodas being tested are labeled as Coke and Pepsi? If sweetness was what mattered most in taste tests, then Pepsi should win out over Coke. And New Coke — which is even sweeter than Pepsi — should win out over both Pepsi and Coke. But this is not what happens. When taste tests with labeled sodas are conducted, the verdicts are reversed. Coke beats Pepsi and both Coke and Pepsi beat New Coke.

Again, what the heck is going on here?

The Brain Overrules The Taste Buds

When Read Montague of Baylor College Medicine performed a version of the Pepsi Challenge with subjects hooked up to an fMRI machine, he found something interesting. In blind taste tests, most people preferred Pepsi, and Pepsi was associated with a higher level of activity in an area of the brain known as the ventral putamen, which helps us evaluate different flavors. By contrast, in a non-blind test, Coke was more popular and was also associated with increased activity in the medial prefrontal cortex — the part of our brain associated with higher-thinking functions. In other words, the higher-thinking functions of the brain were overruling the decision of the taste buds.

You might be saying, “See! This is exactly why we need double-blind studies. People are letting their irrational feelings for a brand interfere with their taste buds and when it comes to choosing flavors that we like, the taste buds — not our irrational brand preferences — should win out. Double-blind tests are the answer.”


Double-blind tests are the answer all right, but they’re the answer to the wrong question.

There are no right answers to wrong questions. ~ Ursula K. Le Guin

Let’s re-review Carl Sagan’s quote on the scientific method:

…the scientific method (is) the best method ever invented for arriving at the truth of things. ~ Carl Sagan

The truth of “things”, yes. But people are not “things”. Consumer preferences are not “things”.

Double-blind tests are designed to eliminate pre-existing biases and the power of suggestion. That’s ideal for scientific inquiries, but it’s totally inappropriate for studying consumer preferences. In fact, it’s worse than useless because double-blind tests eliminate the very thing we’re looking for. When determining consumer preferences, biases and the power of suggestion are not noise to be eliminated — they’re the signal we wish to identify. When studying consumer preferences, we don’t eliminate biases to get to reality. Our biases ARE reality.

The Human Rowboat

The philosopher J. S. Mill once observed:

(T)here are two kinds of wisdom in the world:

1) Scientific; and
2) The Wisdom of Ages.

The first kind of wisdom changes every day. The second kind of wisdom changes not at all. The first kind of wisdom consists in what we know about the world and how it works. The second is what we’ve collectively learned about human nature through the experience of individuals across thousands of years of history. The second kind of knowledge is unsystematic, consists in psychological rather than empirical facts, and is present in more or less equal amounts in every historical period. ((As an aside, my style of writing has been deeply influenced by this idea that there are two kinds of wisdom. I write about technology, but I pepper my articles with quotes filled with the wisdom of ages. It’s always surprising to me how relevant the thoughts of Socrates, Nietzsche, Benjamin Franklin, Groucho Marx, and others who lived yesteryear are to the technology problems of today.))

The scientific method is good for discovering facts, but personal preferences are not facts to be discovered, they are feelings to be uncovered. We are creatures of both logic and emotion. To assume that human beings are only logical is — well — it’s illogical. And very, very counterproductive.

The brain and the heart are like the oars of a rowboat. When you use only one to the exclusion of the other, you end up going around in circles. ~ Dr. Mardy

The above is just a simile, but I think it’s a great one. When it comes to understanding consumer preferences in technology, many otherwise very intelligent people go around and around in circles because they put all their weight behind technology, and neglect — or refuse to acknowledge — the human half of the equation.

The history of technology is the history of understanding things and misunderstanding people.

The Intersection

The best technology products — and this is important, because it’s so widely misunderstood — do not consist only of the best technology. The technology must also cater to the way human beings think and work.

You’ve got to start with the customer experience and work back toward the technology, not the other way around. ~ Steve Jobs

The broader one’s understanding of the human experience, the better design we will have. ~ Steve Jobs

This is why Apple goes on, and on, and on about standing at the intersection of technology and psychology.

I think really great products come from melding two points of view: the technology point of view and the customer point of view. You need both. ~ Steve Jobs

Dr. Land at Polaroid said, “I want Polaroid to stand at the intersection of art and science,” and I’ve never forgotten that. ~ Steve Jobs

Apple has the opportunity to set a new example of how great an American corporation can be, sort of an intersection between science and aesthetics.

We want to stand at the intersection of computers and humanism. ~ Steve Jobs

The reason Apple resonates with people is that there’s a deep current of humanity in our innovation. ~ Steve Jobs

People Don’t Buy A Product, They Buy An Experience

When we make purchases, we use all of our senses, along with our accumulated memories, feelings, knowledge, etc. We apply the lessons we learned yesterday to the purchases of today. We don’t buy products in a vacuum, we buy them within the context of our lives.

When people drink soda in a blind taste test, they prefer Pepsi. When people drink soda in a cup with the soda’s logo on it, they prefer Coke. Why? Because people don’t buy a product, they buy an experience. And for most, drinking Coke is a better experience than drinking Pepsi.

It was actually Pepsi, not Coke, that tricked us with their marketing by convincing us that a blind taste test represented an accurate way to measure the desirability of a soda. The Pepsi challenge wasn’t scientific at all — it was a gimmick because it measured the wrong thing. Products and services aren’t judged by how good they are, they’re judged by how good they make us feel.

We never desire strongly, what we (only) desire rationally. ~ Francois De La Rochefoucauld

New Coke succeeded in labs and Pepsi succeeded in blind taste tests because they appealed to a single sense. Classic Coke succeeded in the marketplace because it appealed to our overall sense of well being.

People don`t ask for facts in making up their minds. They would rather have one good, soul-satisfying emotion than a dozen facts. ~ Leavitt

The Apple Experience

Coke is not successful because of their ingredients any more than a great Chef is successful because of his or her ingredients. It’s how the ingredients are put together and how they are presented that make a great meal.

— Great Chefs sell an experience.
— Coke sells an experience.
— Apple sells an experience.

You can’t use double-blind tests to determine which is the best soda and you can’t use double-blind tests to determine which is the best smartphone either.

The primary reason why so many industry analysts misunderstand Apple is confusion surrounding what Apple actually sells. Apple doesn’t sell phones, tablets, laptops, and desktops, and… smart watches. Apple sells experiences.

Apple is a counterintuitive company because they are an experienced technology company that tries to arrange technology so we don’t have to experience it.

Apple has always been, and I hope it will always be, one of the premiere bridges between mere mortals and this very difficult technology. We may have the fastest PCs, which we do, we may have the most sophisticated machines, which we do. But the most important thing is that Apple is the bridge. ~ Steve Jobs

Apple is a company of solutions wrapped in experiences. ~ Lou Miranda

The reason why Apple can consistently collect between 90-95% of all the smartphone profits is because, while they’re competitors are selling technology, Apple is selling an experience.

Apple has no competition who sell what their customers are buying. ~ Horace Dediu (@asymco)


User Experience is in the eye of the beholder. Until I see double-blind tests about it, I flatly deny that Apple’s is superior/premium…. ~ Obarthelemy

Dude, that ship has sailed.

Double blind tests are not the standard by which you judge the taste of soda and they’re not the standard by which you judge the user experience of a smartphone either. In the marketplace, profit garnered from sales — not double-blind tests — is the only measure that matters. And Coca-Cola and Apple Co., have most all the profits.

What I love about the consumer market that I always hated about the enterprise market is that we come up with a product, we try to tell everybody about it, and every person votes for themselves. They go “yes” or “no.” And if enough of them say yes, we get to come to work tomorrow. ~ Steve Jobs

In the free market, you get to vote “yes” with your dollars, but you don’t get to veto the votes of others.

Indeed, a major source of objection to a free economy is precisely that it… gives people what they want instead of what a particular group thinks they ought to want. ~ Milton Friedman

The verdict of the market is final and inviolate. ((Absent the use of force, the verdict of the market is final and inviolate))

Facts do not cease to exist because they are ignored. ~ Aldous Huxley

What Is Apple Afraid Of?

Apple, the company, just had one of its best quarters ever. AAPL, the stock, just had one of its worst quarters ever. What the hell is going on here?

It’s hard to say who gets criticized the most, the successful person, or the failure but it’s mighty close. ~ Joe Moore

Apple And AAPL

The first thing we have to understand is Apple, the company, is a very different thing from AAPL, the stock. It’s true that severe stock fluctuations can have an ill effect on a company but on balance, the stock follows the company, not the other way around.

Bad Advice

Whenever Apple’s stock (or any company’s stock) suffers a precipitous drop, the know-it-alls ((Bloggers just like me.)) come crawling out of the woodwork with advice on what they should do. As Harry McCracken pointed out in his January 2014 article entitled “Apple Must…”: A Brief History of People Instructing the Company to Do Things, this advice has been uniformly awful.

Most of the “advice” people have been giving Tim Cook today on how to run Apple would eventually bankrupt the company. ~ Neil Cybart on Twitter

Advice to Apple usually falls on one of two sides of the same coin. Apple should:

— Stop doing what made them successful; and
— Start doing what made their competitors unsuccessful.

You guys keep asking Apple to do what everyone else does yet somehow think the result would be different. ~ Ben Thompson (@monkbent)

What Is Apple Afraid Of?

A good way to sharpen our focus on the actual threats to Apple is to ask ourselves: What is Apple afraid of? What is the worst case scenario Apple is trying to avoid? What worries keep Apple’s executives up into the wee hours of the night?

Benedict Evans suggests Apple’s worst fear is developers leave the Apple platforms.

For Apple, I’d suggest the fear is that the developers leave. This is what happened in the 90s and it was a key part of the company’s near-death experience.

Once developers start leaving you’re in a vicious circle that’s very hard to reverse (this is where Windows Phone is now). ~ Benedict Evans

Most industry observers thought the Apple of 2007 was going to suffer the same fate as the Apple of 1984: Introduce a great new product (Macintosh/iPhone), keep prices extraordinarily high, allow a competitor to undercut their price and gain market share, (Microsoft Windows/Google Android), inevitably lose their developers, then their platform, then just plain lose everything.

Apple has done a great job of keeping their developers by attracting the most profitable customers to its platform. The developers go where the money is. So even though Apple only owns about 14% of the smartphone market, they control almost all of the customers developers cherish most.

Out Of The Woods

But that doesn’t mean Apple is out of the woods. They need to find new ways of insuring their current developers stay with their platforms and new developers are attracted to their platforms. I would suggest Apple is currently taking a three-pronged approach to this problem.

Customer Sats
First, Apple needs to maintain its premium customers since they’re the ones most willing to pay for the developer’s wares. This is why Tim Cook absolutely OBSESSES over “Customer Sats” (customer satisfaction).

It’s a virtuous cycle. Apple creates great products that attract great customers that attract great developers. Apple’s customers are happy, which makes Apple’s developers happy, which makes Apple happy, and so on and so forth.

Accelerate Refresh Rates
Second, Apple is attempting to accelerate the iPhone’s refresh cycle. You buy a phone from Apple, you pay it off over two years, but every year you can upgrade it. This will cause a certain segment of iPhone buyers to routinely upgrade their phones every year instead of every two years. And the the incentives for buyers to stick with the iPhone will be enormous.

Third, Apple needs to find a way to continue to grow its iPhone platform.

Wall Street’s greatest fear is Apple has nowhere to go and nowhere to grow. They have a point. There’s only so many people who are willing to pay $600-$700 for a smartphone. ((Although the number of people willing to buy an iPhone is always greater than the naysayers believe. The cry that Apple was going to soon run out of customers has been made each and every year since the iPhone’s inception.)) At some point, Apple will no longer be able to go up. What then?

When companies run into their market ceiling, the traditional business strategy is to go down. That is, expand their product portfolio, introduce cheaper versions of the existing products, gain more market share. There are two problems with this strategy.

First, lowering prices would hurt Apple’s brand. Apple attracts premium buyers because they sell a premium product. Lowering prices might increase sales, but it would dilute Apple’s cachet among its existing customers.

Second, Apple already controls most of the profits in the smartphone sector. For example, Apple garners between 90 to 95 percent of the profits from smartphone hardware sales. What is the point of lowering prices and lowering margins in order to gain another point of profit share? It’s counterproductive.

Expanding The Platform

Instead of going down, Apple is, instead, spreading their platform out. Apple may be running out of customers willing to spend money on an iPhone but that doesn’t mean iPhone customers are running out of money.

[pullquote]Apple may be running out of customers willing to spend money on an iPhone but that doesn’t mean iPhone customers are running out of money.[/pullquote]

Apple is spreading their platform so those who own and love their iPhone can buy more and more Apple products. That’s why Apple works very, very hard to maintain the quality of their notebook and desktop computers and why Apple is introducing new platform products, such as the Apple Watch and the iPad Pro. These various additions to the overall Apple platform give happy buyers more opportunities to buy, give happy developers more opportunities to sell, and give happy Apple more developers developing.

Is It Enough?

Will this be enough to guarantee Apple’s continued future success? Absolutely not. This is a holding action, but it’s a holding action that’s netting Apple as much as $18 billion a quarter. Someday Apple will have to create new and better products. But is this a surprise to anyone? Every company, everywhere, throughout all time, has had to do the same.

New products cannot be rushed. Microsoft came out with a smartphone in 2001, but Apple — even if they had been the Apple of today and not the nearly bankrupt Apple of yesterday — could not have come out with the iPhone in 2001. The technology was simply not ready yet.

These waves of technology, you can see them way before they happen, and you just have to choose wisely which ones you’re going to surf. If you choose unwisely, then you can waste a lot of energy, but if you choose wisely it actually unfolds fairly slowly. It takes years. ~ Steve Jobs

Nor should Apple rush new products to market. That’s not smart and that’s not their raison d’être.

Our north star is to make the best product. Our objective isn’t to make this design for this kind of price point, or for this arbitrary schedule, or line up other things or have X number of phones, it’s to build the best. ~ Tim Cook

The smartphone market is eight years old. Eight. That’s it. The pace of technology is certainly accelerating, but the smartphone market has hardly seen its best days. It took the desktop/notebook market forty to fifty years to mature. I suspect the smartphone still has a few good years ahead of it too.

The US Wireless Industry’s Pivotal Quarter

The US wireless industry is going through an interesting period. T-Mobile has captured much of the attention with its Un-Carrier strategy, while Sprint has also been reinventing itself under its new CEO over the past year. Meanwhile, AT&T and Verizon are shifting their focus to new areas beyond the traditional phone business. Even with all this ongoing change in the industry however, Q4 2015 promises to be a pivotal quarter, for several reasons.

The biggest quarter for switching

The fourth quarter of the year has often been the biggest one for subscribers switching between carriers. Last year, this trend was especially notable, as some 9.4 million subscribers chose to leave one of the big four carriers’ postpaid services during the quarter, the highest number of switchers for years. The chart below shows the number of people leaving the big four carriers’ postpaid services in each quarter (“Losses”), the total number of new postpaid customers joining these carriers each quarter (“Gross adds”), as well as the net number of new postpaid customers these carriers added (“Net adds”) and the net new phone customers they added (“Phone net adds”):

Postpaid adds and losses

As you can see, Q4 last year was an anomalous quarter, with higher numbers in three of these four categories. Interestingly,, there wasn’t actually much growth in the number of phone customers across the industry – much of this activity was simply people switching between the carriers and people adding tablets and other non-phone devices to their plans. But the switching activity was intense, with churn rates for all four carriers higher than the year before. What drove all this switching? Well, two primary things: more aggressive marketing and promotions and the launch of the iPhone 6, which we’ll come back to. The aggressive marketing was driven primarily by T-Mobile and Sprint, both of which had major promotions towards the end of the year designed to win new customers from the other carriers. Verizon and AT&T fought back, of course, but that was largely a defensive maneuver, without a lot of aggressive promotions of their own.

Smartphone sales were also through the roof last year

As well as being a huge switching quarter, Q4 2014 was also the largest quarter ever for smartphone sales in the US. The big four carriers between them sold over 40 million smartphones in one quarter, far higher than any previous quarter, and around 10 million more than both the prior quarter and the same quarter a year ago, which was already that year’s high point:

Smartphone sales

What drove this enormous quarter for smartphone sales? The launch of the iPhone 6. That launch pulled in a lot of upgrade activity that would normally have happened in the first couple of quarters of the following year into the fourth quarter. But it also acted as a vehicle for a lot of switching between carriers. Sprint and T-Mobile in particular made a lot of hay over their great pricing and promotions on the new iPhones, as both carriers are very keen to increase the share of iPhones – and the kinds of customers who buy iPhones – in their base. Yes, all the carriers also sold plenty of other phones in the quarter, but the iPhone’s share rose significantly above even normal Q4 levels.

Questions for Q4 2015

All of that brings us back to the quarter we’re currently in. What does all this activity in the fourth quarter last year portend for this year’s big quarter? Well, we already know Android smartphone vendors seem to be suffering even worse this year than last year in terms of sales. We also know sales of the new iPhones are looking very healthy on a global basis, but it’s too early to know for certain how they’re faring in the US specifically. My hunch is, between iPhone sales around the same level or slightly down from last year and Android sales well down on last year, we’ll actually see the first meaningful year-on-year drop in smartphone sales in the US market. That’s a big shift, given that smartphone sales have grown by 3-5 million year on year in most recent quarters, and it may be the first of a number of quarters of slower smartphone growth in the US market. Apple should do fine, especially since its global growth is now driven by many other markets around the world, not the least of which is China. But it may well be a tough holiday quarter for Android OEMs.

The other thing that’s becoming clear now we’re roughly halfway through the quarter is that carrier marketing is going to be at least as full-on as last year, with T-Mobile announcing free lower-quality video streaming on many of its plans and Sprint announcing this past week it will halve any of the other three carriers’ service pricing. The Sprint promotion is something of a repeat of last year’s main offering, but this time around it includes T-Mobile, which should make it at least as effective as last time. Verizon and AT&T haven’t announced big new initiatives yet and it’s quite possible they won’t, but it is likely they’ll allow their telemarketing teams and in-store reps to present competitive offers to customers threatening to leave and to potential new customers. Overall switching will likely be a little lower than last year without the massive iPhone 6 driver, but I suspect we’ll still see a really significant quarter for switching. The big question is just which of the carriers make out best from all the switching – AT&T has struggled with its phone net adds recently while Sprint has just barely broken into the black in terms of quarterly growth. Verizon has largely held its own and T-Mobile has been king of phone growth for some time now, but even its growth has been slowing. This quarter could be a real boon for some carriers, and could put a real dent in other carriers’ numbers – it’s too soon to tell which will be which, but it’s certainly going to be an important quarter either way.

RIP $500 and Above Android Phones

Today, Samsung launched the Galaxy Note 5 and the Galaxy 6 + Edge. Neither device will change their declining fortunes in the smartphone landscape. What Samsung is up against is something many struggle to internalize. Most understand Samsung’s issues as being eaten alive from the middle price tier of Android. There is truth to this and I’ll address that. However, it is their losses in the high-end of the market most did not see coming and it is going to get worse as subsidies disappear in the US. I’d like to start by addressing this dynamic, then the change in premium Android pricing.

Change in US Subsidy Structure: The US market is largely a replacement market. The initial belief was the change in subsidy structure would hurt Apple because, once consumers could see the full price of the iPhone, they would choose a less expensive device. The fault in this thinking is a consumer was not going to go from an iPhone to a sub $200 Android handset. Furthermore, the structural shift to installment plans is the icing on the cake for Apple and the nail in the coffin for Samsung.

Carriers knew early on moving to installment plans would not hurt the premium tier. They also had a hunch it would help Apple, but it was just a hunch. US consumers like payment plans and they like putting things on credit. The massive amount of US consumer credit card debt is all you need to look at to understand this point. Nearly all research in this area shows US consumers have no problem paying a monthly fee to get something they otherwise could not afford to get it right now. I recognize the unhealthiness of this, but it is reality.

Let’s look at this through the lens of pricing. Take the cost of an iPhone or Samsung phone and divide it up over 12 or 24 months. You get an average of $25-$30 depending on the SKU. Now, let’s use some logic that says a competitor wants to undercut Apple on price to drive sales. Say Motorola wants to launch an amazingly competitively spec device to Apple and charge $350 for it. Their assumption is a device nearly half the cost of the iPhone will be attractive. And, yes, looking at a price of $350 vs. $649 or even $749 looks attractive. However, most consumers are not going to drop $350 right there to pay full price. Therefore, they will still likely utilize a payment plan to get this $350 Motorola phone. Take $350 and divide it up over $24 and you get $14. The consumer is now looking at $14 vs. $25 per month. Not quite the same differential as $300. What is happening is many consumers opt to get the device they want to begin with when looking at the monthly plan dynamics over the full price of the device.

Samsung seems to be using some of this logic, without realizing the real math, by offering the Note Edge 6 + for more than the Note 5 and the iPhone. The problem is a few dollars isn’t going to make any difference and one can argue they hurt themselves by not being on par with the Note 5 or 6 Plus price-wise. While Samsung has certainly benefitted from payment plans, this may all be about to change. Android OEMs who want into this country understand this dynamic. What is still being settled, in their mind, is what the true price differential needs to be to entice a consumer to buy the phone outright and not opt for a payment plan. Let’s say an Android OEM offers roughly the same specs as a premium Samsung phone but for a price of $250. Would that do it? Maybe $199? Somewhere in that range could make a difference and, believe it or not, there are a few handset companies out there who can make money in those prices ranges. As the costs of components come down as well, we could see premium Android devices get under $199. This would make life miserable for Samsung and I believe things are heading in this direction for Android in the US over the next few years.

All the while, the change in the subsidy structure and the move to installment plans could be the driver that puts Apple over 50% market share in the US. I have even had folks at carriers tell me 60% is not out of the question for Apple. Lastly, a data point. From my sell-through tracking model, Apple captured 85% of the sales of smartphones costing more than $600 in Q2 2015. (Yes somebody on the earnings call suggested 90%, but my data suggests 85%. Either way it is high.)

Overall, premium Android is shifting. Look at devices like the OnePlus or Xiaomi’s phones or Micromax in India. All over the world, extremely competitive phones, specification-wise, are popping up in the $300 range and defining the new Android premium price and spec range. Even in this country, a brand like Alcatel sees growth signs from this strategy with the Idol 3 — an incredible phone for $250. I actually use this device as my primary Android tester phone.

These devices are changing the shape of Android premium phones and, remarkably, there is still a roadmap to keep specs high and drive the price even lower. R.I.P. $500 and above Android phones.

Preliminary Q2 2015 Global Smartphone Market and Observations

I thought I’d give our subscribers a quick view of what the landscape is shaping up for with regards to Q2 global smartphone shipments.

The top 5 looks like this:
#1 Samsung
#2 Apple
#3 Huawei
#4 Xiaomi
#5 Lenovo

Screen Shot 2015-07-15 at 10.36.21 AM

For Samsung, we believe shipments will be in the low 70m range, 73-74m to be exact. Our Apple estimates are for 53m units sold. Huawei announced 50m smartphones sold in the first half of 2015 which, by doing the math on first quarter sales, means 32m smartphones shipped. Xiaomi announced 34m smartphones shipped in first half of 2015 for first quarter shipments of 20m.

For philosophical reasons, I do not lump Lenovo and Motorolla sales together. If we were to combine the two, Lenovo would be #4 and Xiaomi #5.

Folks love to talk about Xiaomi but it is clear their initial target of 100m smartphones sold in 2015 is unlikely. This is why they recently have publicly changed to a range of 80-100m smartphones as a goal. In all honesty, even 80m may be a stretch at this point. I’ve been telling people to keep an eye on Huawei as a more interesting Chinese player right now than Xiaomi. While I think Xiaomi has the better business model for the long haul, Huawei has quite a bit of assets and large amounts of capital behind them. Huawei is not just a smartphone manufacturer but also owns HiSilicon which is a semiconductor company supplying networking equipment as well as connected devices. And Huawei has an infrastructure business selling global telecom equipment, among other categories. They have many parts of the business to help fuel their R&D. They are well positioned in several categories from a business and revenue standpoint. That is translating into continued aggressive designs in smartphone hardware and globally their sales are increasing faster than Xiaomi’s. Huawei is, for now, the one to watch and perhaps this battle between Xiaomi and Huawei will get more interesting in 2016.

The markets where Huawei is starting to pick up steam is in Brazil (LATAM as a whole), SE Asia, India, and parts of Africa. In many markets, Huawei is now starting to take share from Samsung and compete strongly with local vendors as their branding and quality perception is increasing. As I said, I still think they are the Chinese company to watch right now with global smartphone sales.

Lastly, we still hear tech news sites insinuate these Chinese brands are coming to the US market in force. This is simply not true. The only area where companies like ZTE have any share is with pre-paid, very low-end smartphones. These are areas the carriers don’t really care much about as they make such little money off these customers. No Chinese brand has any relevant share of the post-paid market and I see no signs of that changing. Pre-paid is growing but very, very slowly and I emphasize these are not profitable customers. We remain pessimistic, given what we see about the US smartphone dynamics, that any Chinese branded handset becomes relevant in the US for the time being.

India: Where the Future of the Mobile Internet is On Display

Every market I study is fascinating in its own right. What keeps surprising me is the seeing how different things are in the markets I study. It would easier if all the markets were roughly the same, with the same brands, products, business models, et., all paralleling each other. But this would also be boring. I’ve outlined what is happening in China with mobile devices and the mobile internet on several occasions. Now I want to dive into India.

Mobile Only

I model global installed base of devices and platforms by country. There are several insightful points to understand about the India smart device landscape. The first is PC penetration is exceptionally low for such a large country. PC penetration (stat includes household, individual ownership, business, and Internet Cafe) in China is over 400m. In terms of pure volume, China and the US have the most mature PC audience respectively. In India, PC penetration is 10%. That breaks out to just over 100m PCs in a country of over 1 billion people. The PC in India is largely a work/business/productivity device and reflective of the number of tech workers or healthy individuals in the region. Yet, the number of smartphones in active use in India is over 250 million. Which makes India the country with the largest number of mobile/smartphone only internet users.

Screen Shot 2015-07-09 at 2.24.10 PM

India is shaping up to be the feature market showing us what a mobile-only world looks like.

One of the key stats we uncovered that got me looking into India as a template to understand mobile only markets was the share of mobile commerce, as a percent, of e-commerce vs other countries. India has the highest share of m-commerce transactions as a percent of total e-commerce than any country. Mobile share of total e-commerce transactions in India is 43%. The next closest is China where m-commerce makes up 32% of e-commerce transactions. Every other country is distant at 20% or below share of mobile commerce.

This incredible emphasis on mobility, where the mobile internet is truly the only consumer internet in India has led companies like Flipkart (the leading e-commerce company in India) to shut down their desktop website and only offer their service through a mobile app. Myntra, a Flipkart property, is also going in mobile and shutting down their desktop website. While these are several examples of companies who had a PC focus and shifted to mobile only, dozens of upstarts in India are starting with mobile only businesses and never bothering to develop a desktop website.

This is again because the internet in India is primarily mobile and with regard to the consumer-centric mobile internet it is absolutely mobile only. Because of this, we will see mobile only innovation coming out of India which we can use as case studies for other markets. I’m watching for new services in banking and finance, payments and micropayments, e-commerce/m-commerce, hyper-local services like fresh food delivery, and more. These areas are all growing out of a mobile-first environment where, even though I’m watching these services in other countries, many in Western markets particularly will still embrace the desktop web. My thesis is, the ones who focus mobile only will bring out innovations not seen yet by those who still embrace both mobile and desktop experiences. In all the areas above, from mobile banking, mobile cash/payments, mobile bill paying, India ranks among the highest in terms of mobile share of these activities.

The other thing that makes India interesting is the proportion of Android to iOS devices is dramatically in favor of Android and we see no signs of that changing for the foreseeable future. Normally, when I’d analyze a market for things like value of user (ASP), average order value with commerce, and other metrics, we would isolate iOS from Android. The basis for this is the understanding, generally speaking, that Apple customers are more valuable than Android users in metrics that matter. However in India, while the super wealthy do have iPhones and are worth more than their Android counterparts, there are simply not enough of them to truly matter. There are a number of economic reasons for this and one of the more fascinating dynamics for Apple between China and India. It is the perceived luxury which drives Apple in China, where that same strategy does not work in India. To help you visualize this gap, here is a snippet from my global country by country installed base model by mobile OS. This is India, iOS vs. Android number of smartphones in use.

Screen Shot 2015-07-09 at 3.30.06 PM

Apple will be challenged in India. Even though there are aggressive sales of the iPhone 5s happening, I’m hearing from users on the ground the legacy experience is not competitive with modern devices and leaves a sour taste in consumers’ mouths. India will be a challenge for Apple.

However, the point remains, what we are seeing happen in India in developing as a truly mobile only country is fascinating. While every mobile only example may not work in other markets, it is helpful to understand what is happening in the region and learn from any relevant innovations.

Discuss in the Tech.pinions Forums: What can I tell you about tech in India?

Forum Topics: Microsoft Smartphones, iPhones in China, Slice Apple Watch Data

One of the ways I’m using our new forums is to share some shorter posts and solicit more of a discussion around certain topics. So I’ll frequently post a short post, or top of mind thought to the forum. To let everyone know there is new content there I’ll create a post like this one and link to some of the newer topics I’ve posted in the forums. Here are the recent posts with some commentary by me to get the discussion going.

How Long Will Microsoft Keep Making Smartphones?
With the recent news of layoffs and restructuring what is the future of Windows Phone and does Microsoft need it to be successful.

Latest iPhone In China Charts via Baidu
Long time readers will know I have access to data from Baidu (China’s leading search engine) that helps me track iPhones in China as a percentage of iOS devices accessing Baidu. I’ve updated these charts up through June and added some commentary around the most recent trends.

Thoughts on Slice Apple Watch Data
I shared some of my thoughts on the Slice Apple Watch sales data that is circulating the news outlets. There is already a great discussion happening on this forum thread.

If you haven’t already be sure to check out or new subscriber only forums. I firmly believe our site has the smartest readers and commenters of any site out there so I deeply appreciate your engagement and we created forums just for this community engagement. There are also a few subscriber initiated forum topics on switching from Android to iOS or Vice Versa and the Future of the Apple Watch. I encourage you to check those out as well.

The China Smartphone Market Report

Every so often for subscribers I do a deep dive on the China smartphone market. I’ll do another one on India in the upcoming weeks. We have some recent data on the China smartphone market I think paints a fascinating picture of what is happening as it relates to several players, and Apple in particular.

China is like no other market I study when it comes to not just mobile but the mobile internet. While we may be able to glean insights from the mobile internet trends in China and apply them to other areas, I’m not convinced we can apply the hardware trends — namely, the size of the market for customers willing to pay extreme premiums to own iPhones.

Overall View of the Market

I track global sales of smartphones by vendor, but also break them out by region. Here is a view of the market share paths of the major vendors since early 2013.

Screen Shot 2015-06-22 at 12.45.17 PM

As you can see, the story of China’s smartphone growth was largely about Android, but also being driven by local manufacturers. In a market which grew from an average of 60 million units a quarter to over 100m units a quarter, there was plenty of share to go around. China was cutting their teeth on mobile and local vendors and low-cost handsets were driving the growth. All along, those of us who study the market knew Apple had a high aspirational brand but had yet to see the masses move. One thing these vendor share charts don’t show is the secondary market or the grey market. The market where iPhones are bought, not through official channels, but through unofficial ones. Remarkably, amidst the rapid smartphone growth of local vendors, Chinese consumers were buying iPhones in such significant numbers where, for a period of time in 2012, illegitimate iPhone sales made up 9-10% of all smartphones sold per quarter in China including through legitimate channels. Which, if legitimate plus grey market shipments were counted, would have put Apple in the top 5 of smartphone vendors by sales volume in China going back to 2012. Our research revealed that, during the time period, grey market iPhones sales were nearly double the number of iPhones sold through legitimate channels. This is why the device usage data from Baidu/Umeng had such high usage rates of iPhone 4 and 4s during that time frame. Such high penetration can only make sense when we add both grey market and legitimate market sales.

Screen Shot 2015-06-22 at 12.58.37 PM

As you see from the chart over time, the huge base of iPhone 4 and 4s owners began to diversify and Apple began bringing new customers into their ecosystem as they cut more deals with local carriers with China Mobile being the driver of Apple’s growth push in China.

Android remains the dominant OS in China’s market of now over 600 million smartphone users. It wasn’t until the launch of the iPhone 6 and 6 Plus that Apple began to pull ahead as the majority leader in market share for the premium smartphone market in China. By vendor, Samsung and Apple were the two primary drivers of the premium smartphone market with Huawei and a few offerings from Xiaomi in the mix. Thanks to the iPhone 6 and 6 Plus, Apple’s share of the premium market is now over 60%. Furthermore, for the past two quarters, the iPhone has been the best selling smartphone brand in mainland China.

Screen Shot 2015-06-22 at 1.55.02 PM

With the premium market sealed up by Apple, the real battle moves to the middle-income parts of the market. Tim Cook made an interesting statement last quarterly earnings call when he stated the iPhone was beginning to penetrate the mid-tier parts of the Chinese smartphone market. This correlates with some recent survey work we did in China where over 50% of those considered in the middle-income bracket expected to spend more money on their next smartphone. This is where the battle will lie for vendors in China. With the most developed parts of China already at 90% smartphone penetration, any meaningful growth by any vendor in mainland China will come from taking customers away from someone else. For the time being, this favors Apple but there are over 50 smartphone brands who sell into mainland China. However, only four brands rank among the most owned across all income classes–Apple, Samsung, Huawei, and Xiaomi.

Consolidation is likely coming in the China smartphone market as it will be difficult for all 50+ brands to continue to do well in the mid and premium price tiers. Any smartphone vendors looking for pure growth must look below the $100 device range and many are not positioned well to succeed there. While Apple is among those well positioned to grab new users from Android brands, Meizu and Oppo are brands to watch as well. Xiaomi was hot for a while but has now seen two quarters of back to back declines and we will see how 2015 pans out for Xiaomi in their home country.

Lastly, another data point that makes China interesting from a growth standpoint for well-positioned brands/vendors is that Chinese consumers replace their smartphones on faster cycles than any other market. The average US consumer refreshes their smartphone every 26 months while the average Chinese consumer refreshes every 22 months.

This is a picture of the market as it is today, but there are new dynamics coming to the Chinese smartphone market which may shake things up in an interesting way. In a similar vein as the US, the Chinese government is mandating all credit cards and retail store point of sale kiosks move to dual-purpose cards. These are cards not unlike the chip and pin cards coming to the US market but using different technology and including the secure element. These kiosks coming to point of sale will include both the RFID dual-purpose chip and an NFC chip which means there will be significant infrastructure for contactless mobile payments in many developed areas and the larger Chinese retailers by the end of 2015. I hear nearly every Android OEM is going full steam with NFC in China as well as OEMs and mobile payment companies starting to put the pieces together to compete for mobile wallets and mobile payments. Of course, if Apple can secure the right deals with UnionPay and others, they will also be right in the mix as the Chinese market ripens for mobile contactless payments via smartphones. Everyone I talk to at the component and hardware level who focuses on China believes another competitive cycle is coming from OEMs to capitalize on this new dynamic of mobile payments.

What’s perhaps most interesting about the smartphone hardware market in China is how it is moving more toward a premium market. Thanks to Apple, ASPs in China are actually growing not declining. Remember most believed the way to be successful in China was to make low-cost/cheaper handsets, but the market is actually moving up and this is true across income brackets. I’ve plotted estimates for ASP in China compared to that of the US to show this.

Screen Shot 2015-06-22 at 6.15.51 PM

All these dynamics make China quite unique. It is becoming the largest premium smartphone market and I don’t see that changing anytime soon. It is the largest smartphone market with over 600m smartphones in active use, and the vast majority of those 600m smartphone users are looking to spend more on their next handset and simultaneously are investing more in the Chinese mobile internet and contributing to local economy. You can see why this market is so important to many competitors due to its size but for Apple it’s crucial to their growth. This is why we are seeing and will continue to see significant investments and customizations with Apple’s hardware, software, and services specific to the China market. To compete in China, hardware, software and services core competencies are becoming essential. Which is why my sense is there are only a handful you can tough it out.

Soon I’ll do a deep dive on the Indian smartphone market. India in terms of potential user size will rival China, but as you will see dramatically different market dynamics exist.

Understanding Android to iPhone Switching

There are several dynamics I want to explore about Android-to-iPhone switchers. I’ve seen enough data to suggest double digit percentages of Apple’s iPhone sales the past few quarters came from consumers switching from Android. The bull case for Apple’s growth is driven by share gains from Android, not new/first time smartphone owners. Apple’s larger screen offerings, the iPhone 6 and 6 Plus, are the clear catalysts driving Android switching. I’ve been trying to gather research and insights with a goal to understanding more deeply this switching dynamic. For now, I’ve concluded a few things.

Upside Outside the US

From data models I’ve made and thanks to discussions with local analysts and carriers, it seems those switching from Android to the iPhone are largely coming from outside the United States. Research suggests two markets in particular are where the bulk of Android switching is happening–China and parts of Europe. In both these regions, the installed base of premium Android phones, ones from Samsung, HTC, and LG, are much higher than in the US. In China, Apple has always led the premium market in terms of sales but, until recently, this volume was relatively low. Premium Android in China is in the $300-$400 range and that price tier represented the largest volume of sales during the Chinese smartphone boom and is driven by local Chinese OEMs. In both these markets “premium” was more defined by specs than price. The Asian consumer viewed larger screen devices as more premium than smaller phones. This point underscores the significance of Apple making larger screen iPhones and why they are now selling in significant quantities in China.

Europe, also has a much larger Android installed base of upper mid-tier and premium smartphones. In both markets, the larger installed base of Android and market demand for larger screen smartphones creates the fertile ground for Apple to attempt to gain share. Where there is clear evidence this is happening is in China, with over 38% of Chinese saying they switched from an Android phone to an iPhone in the previous quarter. According to Kantar, a similar dynamic happened in Europe last quarter with 32% of iPhone buyers switching from Android. Similar data from our surveys shows the US switching rate was in the low 20% range. However, Apple is up against a dynamic which may slow sentiment to switch some of our behavioral research is uncovering.


In some recent research, I set out to understand what percentage of Android users in the US self-identify as Android users. Meaning, they have come to the conclusion Android, not iOS, is for them. I did a consumer insight panel and found 38% of Android users I interviewed specifically used the phrase (or something along the lines of) “I’m an Android user” as we were discussing preferences and decisions to keep buying Android phones vs the iPhone. This group was keenly aware of their self-identification as an Android user. Within this group, the sentiment was stronger to self-identify as an “Android user” for men than women. Another dynamic I uncovered in these interviews was 34% of those I spoke with were staying with Android because it did all the things they needed and felt no need to switch. For this group, Android was “good enough”. In nearly every discussion with Android users, the Android operating system was their entry into the smartphone world and they have built up a familiarity with it. To many, the idea of switching seemed overwhelming given how many years they had experience and invested in the Android platform. These interviews validated a great deal of larger survey work we did. It highlighted the first smartphone OS a consumer used tended to be the one they stuck with.

Through conversations from the carrier ecosystem, I learned the carriers themselves are weary of this familiarity angle and are hesitant to push consumers to switch operating systems, if they are undecided or on the fence, fearing it could lead to higher returns. Whether this is true or not it is a dynamic to keep in mind. In the US, Europe, and key parts of China, the smartphone market is mature. Mature markets have historically not seen platform share churn in significant quantities.

The case for switching from Android to iOS has indeed gotten stronger. However, those switching from Android are not the majority of iPhone buyers today. This dynamic is one worth continuing to study as it is extremely relevant to the global mobile discussion around smartphones. The majority of new customers for Apple already have a smartphone and it is most likely an Android phone. Apple’s iPhone installed base growth hinges on their ability to get consumers to switch platforms. While there are some share gains to be had in the US, it is the switching dynamic outside the US to keep an eye on.

Xiaomi Phones: Headed for U.S., but Some Wait to Come

Xiaomi’s Hugh Barra is ready to bring the hot growth of its Chinese phones to the the U.S. But, in a presentation at Re/code’s Coder conference, Barra made it clear there is a whole lot of change that needs to be made before it can sell anything more than a handful of accessories in the US market. Unlike other losses such as Huawei found in the market, he’s not going make a lot of mistakes heading for the top.

The problem isn’t the phones themselves. They are beautiful, matching leaders such as Apple and Samsung. And their pricing is phenomenal, selling high quality phones at $315 with no service charges. Barra says the prices are primarily the result of direct sales to consumers, excellent cost from key suppliers such as Qualcomm, and keeping the product in market for about 18 months allowing for falling component parts.

The U.S. market, however, will be difficult for Xiaomi to be a big player. Even though some carriers say they are getting tired of the duopoly of Apple and Samsung, they are actually increasing the market for a duopoly with their payment plans. Even phones being sold handled by retailers with carrier sales, whether it’s at Amazon or Best Buy.

It is unlikely carriers will front the advertising bill for Xiaomi like they do for other devices. Which means they would also need to create their own advertising, which is expensive and would eat into margins (ask Samsung how much that can cost). Primarily however, Xiaomi would need to change their phones for the U.S. market and they will have a tough time with retailers.

Another thing in the way of Xiaomi are areas where they seem to have more focus than the US for the time being. The company is, of course, already a major player in China (fighting mainly with Apple, Samsung, and Huawei) but its offerings in a rapidly growing market like India seems more attractive than pushing for the U.S. “There are countless things that need to be done,” Barra says. “And I don’t want to try to live in both San Francisco and Bangalore.”

Xiaomi will also have software challenges in the U.S. There are two approaches in the bulk of U.S. — Apple and full Google Android. The Xiaomi phones sold in China avoid the Google services, which are not allowed there, by using the Android Open Source Project and proprietary local services. In Hong Kong, India, and elsewhere, Xiaomi are sold with AOSP helped with a collection of Google standard services. Will that provide success in the U.S.? Time will tell.

The company is trying a very slow and simple retailing effort in the U.S. It has opened some Mi Stores, but currently offers only three products: a tracking band, a battery charging brick, and wireless headset. They are very attractive and at very competitive prices, but that is pretty limited competition. The products are not made by Xiaomi, but are only made by approval and co-branded.

Barra once ran Android’s Nexus phones, not a great experience. But he does seem to have learned a great deal in the U.S. Watch for Xiaomi to take its time and not move hard into the U.S. for quite some time.

Questioning China and India’s Smartphone Growth

There is an interesting narrative forming regarding certain markets, ones nowhere near saturated with smartphones, that they are slowing down in YoY shipments. Fresh off Q1 2015 numbers, the Wall St. Journal published an article on how China’s smartphone market growth is slowing. There was a particular statement that caught my attention:

Experts say the slowdown is largely driven by the disappearance of China’s first-time buyers. About three-quarters of China’s mobile phones in use are smartphones, and they make up 90% of cellphone sales, said Tom Kang, research director with market-research firm Counterpoint, meaning just about everybody in China who wants a smartphone already has one. “China is now a replacement market,” Mr. Kang said.

China (greater PRC) is not 90% saturated with smartphones. The actual smartphone penetration in PRC is 46-48% approximately. I discussed this stat with the Counterpoint folks and they said that they were misquoted. What their statistics state is 90% of quarterly sales in urban China are smartphones. Now, while this is a simplified understanding of urban sales in China, we typically break China out into tiers. Tiers 1-3 are more developed areas like Beijing, Guangdong, Shanghai, etc. Other regions make up the less developed and more rural parts of China. What Counterpoint is saying is the areas of Tier 1-3 are saturated with smartphones and 90% of sales in those regions are a replacement market. This is likely true. But their view, as well as IDC’s, that expectations are for smartphone growth to be flat in 2015 is an interesting narrative.

Significantly, reports are that overall in 2014 China smartphone shipments actually declined. It may have been close to decline or slightly positive but the main viewpoint stands out — China is not adding new smartphone users at the rate it once was.

Another observation is the same thing is happening in India. India is similarly nowhere near saturated; however, the more developed parts of India are rapidly saturating.

To view this, I use this chart:

Screen Shot 2015-05-11 at 10.51.35 AM

As you can see, there is significant headroom for growth in China and India, both areas I spend considerable time studying the local market with the same focus I study the US market. What is interesting is why markets with such headroom for smartphone growth are slowing. As I described earlier, when we segment China into developed and undeveloped tiers, we do the same in India. The tier 1-3 segments of India are cities like Delhi, Bangalore, and Mumbai. The difference between the developed parts of China and the developed parts of India are there are significantly more people in developed China than in developed India. If we just use the current computer installed base to measure this, we see it in a few numbers. In China, PC penetration is nearly 40%. In India, it is less than 10% as a percent of total population. Similarly, smartphone penetration is 46-48% in China. In India, it is closer to 20% penetration of total population. The key observation is in developed China (consumers with more disposable income, higher wages, etc.), penetration is more than two times what it is in India.

Here is another way to look at this from an economic/GDP standpoint:

screen shot 2015-02-25 at 12.20.47 pm

Many of our readers know I break out the mobile market into two segments. The approximately two billion existing smartphone owners and the next billion plus. My friend Benedict Evans, a VC at a16z is fond of saying, “The next billion are not anything like the first”. I fully agree.

The next two billion smartphone owners have needs that are very different. They have distinctive underlying economics in terms of how much money they can spend on a phone, a data plan, and even the cost to charge their phone. This last point gets misunderstood. In very rural and village parts of developing areas, people buy car batteries to power their homes. While it is true power is not free in the US, it is a small part of many people’s budget. Whereas, in these markets, the cost of power could often be more than half their disposable income.

As I observe the notion that smartphone sales are slowing in very large markets like China and India, it seems it has more to do with infrastructure issues like cost of data, and even power, than it does desire. Just to help our readers see this from a different perspective, a company called Digicel is doing some interesting things in heavily rural and village areas. I encourage you to read this article on what they are doing in Papa New Guinea to solve some problems many of us take for granted.

The Growing US Smartphone Base

In my job, I have to keep track of many different data sets. Among them are the numbers regularly put out by big analyst firms as well as specialists like Comscore and Kantar Worldpanel. Comscore in particular puts out one of the few data sets which is intended to quantify the base of smartphones in the US on a regular basis, but it has always seemed to be a little off. As well as these various data sets, I also track the US wireless operators closely and that provides a different way of looking at the US smartphone base — which ends up being quite a bit bigger than Comscore suggests.

Comscore’s figures

Comscore regularly reports the number of smartphones in use, the total number of mobile users, and the penetration of the total market by smartphones. The featurephone and smartphone numbers from these reports are shown in the chart below:

Screenshot 2015-05-08 17.07.52

The trend here is absolutely correct: the overall number of phones isn’t growing that much (though it is growing more than Comscore suggests), but smartphones make up an ever-increasing proportion of the total. Comscore has total phones at around 243 million as of the end of March.

Figures derived from the wireless operators

Meanwhile, the US operators, in their financial reporting, provide a rather different picture. They don’t all explicitly report the number of smartphones they serve, but they do provide enough information we can get very close to an accurate figure. I track the four largest operators in particular: AT&T, Sprint, T-Mobile, Tracfone, and Verizon Wireless. My estimate for the total number of smartphones and feature phones on these operators is shown in the chart below:

Screenshot 2015-05-08 17.12.12

As you can see, the broad pattern is similar, with limited growth in total, and smartphones growing steadily as a proportion of the total. But the total number is quite a bit higher than Comscore’s number, as you can see when we set the two smartphone data sets side by side:

Screenshot 2015-05-08 17.02.17

The difference is around 40 million smartphones, a fairly significant discrepancy between the two. And, although these five operators account for the vast majority of US mobile subscribers, they don’t account for all of them – we need to add a few million more for the various smaller carriers in the market.

Why so different?

What’s behind the difference in these two data sets? A look at Comscore’s methodology helps to illuminate the situation. Here’s the methodology statement from Comscore’s latest press release:

MobiLens data is derived from an intelligent online survey of a nationally representative sample of mobile subscribers age 13 and older. Data on mobile phone usage refers to a respondent’s primary mobile phone and does not include data related to a respondent’s secondary device.

First, Comscore’s data is survey-based. There’s nothing wrong with surveys per se – they’re often the only way to discover certain patterns. But it’s always a proxy for the real thing, which is of course what the wireless operators report. The wireless carrier data doesn’t give us breakdowns between operating systems, but it’s a great way to arrive at an overall market size. There can always be flaws in survey methodologies, including leaving out certain populations and thus under-representing them. Second, Comscore’s survey only includes those over 12 in the US population. Given that kids are getting phones younger and younger, that certainly doesn’t capture the whole US mobile population. The US Census Bureau indicates there are 20 million kids aged 10-14, of which roughly 60%, or 13-14 million, are likely 10-12 years old. Not all of these have smartphones, obviously, but some proportion does (and some even younger kids do too), so this may be part of the problem. Third, the Comscore survey only asks about users’ primary phone, which means it undercounts all those devices which are secondary devices, whether that’s a work device, a cheap smartphone for use in the car, or some other phone used in addition to the main one. Given how many people use both work and personal devices, it’s entirely possible this also significantly under-represents the overall phone (and smartphone) population in the US.

Does this matter?

The question, of course, is whether this matters? Should we just throw out the Comscore data as defective? Or can it still be useful? The shortfall is certainly an issue. If you were relying on Comscore’s data to forecast the total iPhone market in the US, for example, you’d end up with about 80 million users, while there would be just under 100 million Android users. But we already know that Comscore is under-counting the total smartphone market by around 35-40 million, so each of these numbers might be 15-20 million higher in practice. That’s important if you’re trying to get an accurate gauge. The next question is whether the split between operating systems is accurate, or whether that’s defective too. That’s harder to ascertain through other sources and Kantar and other data seems to bear out similar trends in broad terms, so I’m inclined to believe it’s fairly accurate. But I take it with at least a pinch of salt on the basis that I know the overall data is flawed. I think the key, ultimately, is not to rely on any one data point, but rather to find as many data points as possible that either corroborate or contradict one another, then synthesize and aggregate them to arrive at the best possible picture of reality. That’s an ongoing challenge and it will never produce perfect results, but it’s the only way to do this properly. Relying on a single data source, especially one with significant flaws, is always a dangerous business.

Windows Phone and Prepaid in the US

Alternative title: Cricket is saving Windows Phone in the US

I’ve written quite a bit about Windows Phone previously, but I have a slightly different angle this time around — the impact prepaid is having on Windows Phone in the US and, in particular, the impact AT&T subsidiary Cricket is having on Windows Phone sales in the US.

I’m going to draw on a couple of different sources here:

  • Comscore’s regular monthly updates on smartphone installed base market share in the US. I’m using these primarily to measure the size of the installed base for Windows Phone in the US. The latest data as I write is for February 2015.
  • AdDuplex’s data on the Windows Phone installed base, based on its ad analytics software. Its latest numbers are for March 2015. I’ve used these numbers before for other purposes and Nokia in the past has occasionally pointed at them as well, which leads me to believe they’re generally pretty accurate.

Windows Phone in the US is growing

The headline from the recent Comscore data as it relates to Windows Phone is that the platform is growing, at least a little bit. The chart below shows the percentage market share number Comscore reports, as well as the installed base of devices this implies, based on Comscore’s overall smartphone market sizing:

Screenshot 2015-04-26 08.50.21

What you can see is that market share is somewhat static, perhaps rising a tiny bit over time. But because the overall smartphone market is growing, the installed base this represents is also growing over time, from around 5.5 million a year ago to 6.6 million in January 2015 (the slight dip recorded in February likely isn’t meaningful). Given that Windows Phone is in a bit of a lull at the moment, since there hasn’t been a new flagship in quite some time (and isn’t likely to be another until Windows 10 launches later this year), where is this growth coming from?

As elsewhere, the low end is driving growth

The answer is pretty simple: as in other markets, from a device perspective, the growth is all coming from the low end. Here, we switch to the AdDuplex data on the Windows Phone installed base. AdDuplex only reports the top 10 devices in every market, so we don’t have a complete picture of the base, but we do know what’s selling best:

AdDuplex Windows Phone US by device

As you can see, what drove growth through the second half of 2013 and much of 2014 was the Lumia 521 and, to a lesser extent, the Lumia 520. Over the last six months or so, it’s the Lumia 635 that’s driving growth, with the Lumia 630 doing fairly well, too. The 500 and 600 ranges have been the lowest priced devices in the Lumia range, but they’re vastly outselling anything else right now both in the US and elsewhere.

The role of prepaid, and especially Cricket

So far, so predictable. But what’s particularly interesting in the US is the role of prepaid and especially the role of a single provider, Cricket. One of the nice things about the AdDuplex data set is the company breaks out wireless providers in the base, and one company that’s appeared to come out of nowhere over the last nine months is Cricket.

The history here is a little complicated, because today’s Cricket is actually the combination of two separate businesses: Leap Wireless, a CDMA-based prepaid operator which AT&T acquired, and Aio Wireless, a prepaid subsidiary which AT&T merged with Leap when the acquisition closed. Aio had a couple of Windows Phone devices before the transition, and the new Cricket launched its first Windows Phone handset, the Lumia 1320 phablet, in June 2014, followed in July by the Lumia 630, with the 530 and 635 landing later in the year. But, during that time, Cricket has gone from zero to over 20% of the base. The chart below shows that meteoric rise over a period of roughly nine months:

AdDuplex Cricket data

What gets really interesting is if you translate that share into a number in the installed base, based on the Comscore numbers. It’s a bit of a tricky calculation, because AdDuplex technically only breaks out Windows 8 devices by operator and there’s likely some number of Windows Phone 7 devices in the US base still, albeit a small one. But based on the combination of AdDuplex and Comscore data, it’s likely Cricket has between 1 and 1.5 million Windows Phone devices in its base, which is a fairly significant chunk of Cricket’s overall base, perhaps as much as 25%. Secondly, remember the overall Windows Phone growth numbers we looked at earlier? There was net growth of about 1 million Windows Phone handsets in the base during that same nine month period or, in other words, the difference between those that left the platform and joined it was 1 million. Given Cricket likely added around 1 million during that period, it’s possible it accounted for the vast majority of that net growth.

Windows Phone is disproportionately prepaid

Cricket isn’t the only prepaid brand where Windows Phone is big, though. It’s hard to get a full postpaid/prepaid breakdown from the AdDuplex numbers because, for the major carrier brands, the two aren’t separated. But even if we just focus on MetroPCS, Cricket and a phone only available on AT&T’s GoPhone prepaid service, these three add up to around 40% of the total base in the US. Add in a few percentage points for other prepaid sales on the major carriers and we could be getting close to half the Windows Phone base on prepaid, compared with about 25% of the total US phone base on prepaid. In this sense, Windows Phone is the anti-iPhone, with the iPhone well underrepresented in the prepaid market, just as Windows Phone is well over-represented.

The big question for me around Windows Phone remains whether this disproportionate emphasis on low end and prepaid phones will end at some point. The launch of Windows 10 in the coming months provides the best near term opportunity for Microsoft to re-engage with the premium end of the market, so these numbers will be well worth watching over the next year or so.

All Eyes on Android

Few topics come up more in my work with the major technology industry players than Android. From the outside looking in, everything may look rosy in Android Land. But get inside the heart of the industry where long term business decisions about Android are being made, you’ll see a different picture.

The Promise of Android

When Android first hit the market, there was a great deal of excitement because it offered an alternative to Microsoft many hardware companies were dying for. It wasn’t just the lure of a platform other than Microsoft’s but the idea of differentiation. The PC era was not, and is still not, kind to many PC OEMs revenue wise. Windows provides very little sustainable differentiation for hardware OEMs. This is the curse of being a modular company. When you ship someone else’s software, inevitably, the biggest point of differentiation one can offer is price. Shipping Windows creates what I like to call “the sea of sameness”. Other than some moderate hardware differences, which few care about, everything looks mostly identical. When everything looks the same, price becomes the driving factor in decision making for the mass market.

Android was to provide a different environment for hardware companies. One where they had some hope to avoid the sea of sameness. And this is exactly what HTC, Samsung, LG, Sony, Xiaomi, and others have done. By not shipping stock Android, they looked to create layers of value on top of Android to differentiate. However, we observed a tried and true reality in this environment — hardware companies are not great at software. The attempt to add value on top of Android often made it more bloated and lessened the experience. This is why we see a shift back from OEMs, not entirely, but closer to the Android experience Google intends. The designs and UI paradigms OEMs were adding onto Android are starting to look and feel closer to stock Android without completely allowing their software to look exactly the same to everyone shopping for a smartphone.

This is the Android world as it sits today for Google’s version of Android. Google’s version of Android has somewhere between ~1.3-~1.5 billion users today (hopefully this number gets update at I/O). 70-80% of customers using Google’s version of Android are not necessarily just looking for bottom-of-the-barrel cheap hardware and OEMs looking for some margins still have an opportunity to make money in hardware with this customer base. What happens next is where more uncertainty exists.

Myself and others have made the point that the next two billion consumers to come online via their first computer, a smartphone, are going to be nothing like the existing two billion. This creates both new opportunities and new challenges. How these will be met is a key question. More central to this is the role Google itself will play with Android as these new challenges and opportunities present itself. I’ve made the point that something like Cyanogen seems to be attractive to OEMs looking for a packaged non-Google Android solution. The vast majority of OEMs who are now starting to use China as contract manufacturing aren’t even going to try to build software, so they need someone else to do it while, at the same time, offering ways to customize their services, where their real revenue will come from.

Interestingly, Google’s weakness in China and among the China tech manufacturing ecosystem is posing quite a challenge for them with the next billion plus. Companies are looking to China to get into tablet or smartphone hardware and Google’s version of Android is not an option. There are way too many hoops for an aspiring brand looking to get into hardware to jump through that many have no intention of doing it. They would prefer to simply go to China, pick the hardware off the menu of options, slap their brand on it, and go to market. Because of this, over 30% of smartphones sold in India last quarter were not Google certified (GMS compliant) but did have Google services installed after purchase. Brands sold the phones, flashed with AOSP in China and consumers installed the Play store and maps on their own once they got the devices. This is not a tight integration of Google services and their value is minimal compared to something like Android One, which Google would prefer service the next billion.

The headroom to grow in this space is extraordinary. I created this chart of smartphone installed base by many major countries compared to their population.

Screen Shot 2015-04-21 at 8.25.11 AM

As you can see China, India, and the whole of the African continent are where the bulk of the next two billion new internet users will come from. What Android is and looks like for these groups is where all eyes are focused. Is this a Google company play or a non-Google company play? Is Facebook via WhatsApp and other assets better positioned to connect the next several billion or will it be something else? These markets will look very different than developed ones so will incumbents succeed or will it be new companies who rise to power? I see little reason at this point that some flavor, or even many flavors of Android, play a major role in this next phase. But the uncertainty around the details is fascinating considering we are on the cusp of a new global push to bring the internet to previously unconnected people.

The State of Samsung

As you know, Samsung has many different business units. However, the semiconductor and mobile group are the largest contributors to the company’s operating profit. Samsung’s tale has been quite the roller coaster and, while I feel things may be stabilizing for them, I don’t see their mobile division (the one that makes Samsung-branded smartphones) getting the company back to its peak revenue days, which happened some time ago. Here is what they pre-announced with regard to earnings.

Screen Shot 2015-04-06 at 5.32.14 PM

Interestingly, the report highlights that operating profits were up even though sales were not. There are a few dynamics playing into this.

First, the Korean Won has depreciated one percent vs. the US dollar during 1Q15, which was estimated by financial analysts to have a positive impact on component profits. I saw analyst forecasts upping their component estimates as high as 5% based on this point. Component revenues contribute a great deal to Samsung’s operating profits, sometimes more than 50%, and this clearly helped this quarter, perhaps more than many realize.

In my view, the component side of Samsung’s business is the most important and has been for a long while. I’m fond of saying Samsung is a component company and that the brand has historically been used to move those components in volume. But as Samsung is finding out, it is beginning to require more than their brand to move key components in volume. This is ultimately why I feel Samsung is getting aggressive on the semiconductor side of things. Their flash memory and semiconductors business has long been a key contributor thanks to their partnership with Apple. However, it seems they are looking to broaden their semiconductor efforts.

Some time ago, Samsung acquired an ARM architecture license which gave them the ability to fully customize, in ways Qualcomm, AMD, Apple, Nvidia, Broadcom and others have, ARM cores. The benefit to being an ARM architecture licensee is differentiation. The companies above, with the exception of Apple, sell their cores to customers and add their own proprietary value to separate their offering from the traditionally lower priced generic ARM core ones. Making this move can only mean one of two things. One, Samsung intends to focus more heavily on the premium parts of the smartphone market or two, they intend to compete with Qualcomm, Mediatek, and even Intel, in selling their cores more broadly to other OEMs. Of course, a mix of both could be true, but development costs surrounding being an architecture licensee and customizing an ARM core to create a proprietary ARM architecture are not cheap. Therefore, one has to monetize this investment. I’m not sure Samsung can do this by just focusing on premium, let alone their own premium offerings.

Regardless, this side of the business may have more upside than their branded mobile division and it will be interesting to see how they navigate through it.

On to Samsung branded smartphones. Most analyst estimates have Samsung smartphone shipments in the 80-82m range. Interestingly however, the release stated earnings were a beat on estimates but sales were not. The sales figures of smartphones are not disclosed, so those of us who estimate will have to do the hard work.

Samsung is tough to estimate because they promote shipments not sell-through. While these numbers generally correlate at some point in time, we know Samsung likes to stuff the channel to make their shipment numbers look good. Sometimes in excess of 5m units or more could be pushed to channel and not sold through. So I generally take any estimates with a grain of salt until we get a clearer sell through picture. Personally from my own channel discussions, I’m not sure Samsung got to the 80m range but rather mid 70s perhaps — flat for the quarter. This will still put them back as the number one smartphone seller by volume but most of those sales came from the low end.

Samsung is at a crossroads in my view. They are losing low end smartphone sales faster than they are losing sales in premium. They don’t actually sell a lot of premium smartphones, less than 100m per year, but they make 70-80% of their profit from these devices. One could argue Samsung should just focus on the premium market and make a go at growing this business and the premium middle, devices in the $400-500 range, which is likely to become the new Android premium price point. This is the price where many vendors will focus their “premium” efforts and it makes sense for Samsung to focus in this area and the super high end against Apple rather than chase the sub-$200 market for growth. If they are doing what it seems like they are doing with their semiconductor strategy then this path seems plausible.

Lastly, Samsung may, and likely should, perhaps buy some of the less dominant brands in other markets. Oppo in China, for example, is one that is rising quickly, or Micromax in India, or even a smaller vendor in the region who is trying to separate with a brand identity. These companies don’t want just to chase the bottom but are looking to build a regional brand appealing to regional players. Samsung could buy these companies and use them to ship their own components like memory, chipsets, displays, etc., and thus have more brands helping their component business grow. This seems like the best way to address their brand struggles to move components in volume and help them avoid channel conflict with their own brand while trying to sell low end and high end smartphones.

2015 is going to be a defining year for Samsung as they navigate these waters. The choices they make this year will dictate their chances to be a global player in some capacity going forward.

The Devices of Our Lives

One of the single most important observations of the post-PC era has been how humans now distribute their time among many digital screens. During the age of the PC, there were primarily two screens in a consumer’s life they looked at for long periods of time — the TV and the PC. Those were simpler times for everyone. The TV was the primary entertainment screen and the PC was the primary productivity screen. With the advent of smartphones and tablets, both the TV and the PC had their primary roles challenged. Entertainment could now be on any number of screens. Similarly, productivity could now be done on any number of screens.

This fundamental shift in time is one based solely on convenience. Certainly, a large 50″ HDTV is likely the best place to watch movies or TV for long periods of time. However, it is not always the most convenient screen to do so. The PC is also the best place to crank out many long emails. However, it is not always the most convenient screen to do so. What the post-PC era has brought us are any number of new ways (screens) to fulfill the technological jobs in our lives. Now, it is up to us to decide the solution that works best for us.

This is what makes the market landscape so fascinating. There will be a great deal of variety for end users when it comes to their computing solutions. How many screens will depend entirely on the needs of the person and the jobs they need fulfilled. Today, we have a decent idea of the size of the markets for these different screens. PCs are in use by a little over a billion people. Tablets are in use by than less than 500m people. Smartphones are used by about two billion people. PC growth remains flat to declining. Tablet growth has slowed dramatically but is still growing slightly. Smartphones still have the most growth of any screen. We know smartphones are the one primary computer four billion plus humans will own some day. For the time being, it seems tablets and PCs have normalized and are not growing or declining steeply in any direction. For these three categories it seems the dust has settled and we have a somewhat firm understanding of what the next few years will hold.

For most people on the planet, the smartphone is central. The PC used to be the center of consumer’s digital lives and now it is the smartphone. Each smart screen other than the smartphone they choose to own flows from the smartphone. Consumers are getting savvy about this. They are starting to understand how their time is shifting and beginning to think through the screens in their lives — the ones they need and the ones they don’t.

What is interesting about all the screens in our lives is how they demand attention. While the global average total time spent per day on a smartphone is now over two hours, those two hours are the accumulation of many shorter interactions with a smartphone every day. Data from an observational study I did observed a range of tasks done by people with their smartphones and revealed the average session time of a US consumer was 2.5 minutes. We similarly studied session times based on tasks and found that, depending on the task (checking email, using social networking, reading the news, checking the weather, etc.), it varied the amount of time per session. However, consumers rarely felt incentivized to turn their phone on, launch an app and do just that one task. More often than not, once they turned their phone on, they started with one task then moved to a few others, making the average time much higher per engagement. My overall takeaway from this emphasized my theory about how engaging and immersive each screen in our life is. This is why I think the potential of the smartwatch is interesting.

While time has been shifting from PC to smartphones, many central tasks remain. People still sit in front of their PCs for hours at a time, but we found many consumer simultaneously use their smartphone while doing so. The reason for this is certain tasks are more efficient on the smartphone. Looking at the weather, checking stocks, even viewing a social network. Many consumers admit to doing these types of tasks while sitting at work in front of their PCs. The reason? Convenience.

Now apply this to the smartwatch. My thesis going into studying smartwatches over the next few years will be based on this convenience of information display and micro-interactions which we observed to some degree as a dynamic between PCs and smartphones. The always visible or easily and quickly accessible smart screen on your wrist may add a dimension that did not exist before. A screen which is not designed to suck us in for 2.5 minutes or longer.

The smartwatch fits into the overall post-PC theme. It will be one of many digital screens that exist as an option for consumers to put together a solution of smart screens that work best for them. Each screen has a role to play, even if that role evolves over time.

Bullish on Cyanogen

I’ve spent a bit of time over the past few months digging into Cyanogen. Cyanogen is an Android ROM that now has aspirations to be a full mobile OS Android alternative. Cyanogen is running on devices such as the OnePlus and the Micromax YuPhone in India. I have many relationships with Silicon Valley VCs who often come to me for assistance doing due diligence on companies looking to enter markets I have expertise in. My insights and research are often part of the process investors use to decide whether to invest in a company or not. One of those I’ve provided assistance on is Cyanogen.

My overall thesis on the mobile landscape is that the dominant mobile operating system provider is at a crossroads. I wrote about this in my article the “Android Schism”. In short, Google is in a tough spot as they try to have Android appeal to both their most lucrative customers who are part of their profitable existing 1.3 billion active users. When we wrestle with the observation that the most profitable customers on the planet are all now online and using primary devices to access the Internet with smartphones and PCs, then we recognize all the growth of new Internet users is going to come from customers with less disposable income, less income overall and, from a monetization standpoint, dramatically lower average revenue per user.

Interestingly last year, in what seems to be counter to what many assumed about mobile growth, both China and India saw declines in smartphone shipments. So we ask ourselves why? The answer lies in the availability of low cost smartphones. Essentially, those who can afford smartphones have them. The rest will come online when prices get low enough. In many cases, we just weren’t there yet to continue the growth in those markets. 2015 will bring a different set of circumstances.

So why does Cyanogen matter? The bottom line is nearly every handset vendor I talk to is looking for every possibility to not have to partner with Google. When I say partner with Google, I mean with Google services not necessarily Android. So this brings us to Android’s fork in the road. In this next phase of mobile, the next two billion consumers we bring online are new to the Internet and the smartphone is their first computer. This customer group has dramatically fewer demands on their devices and only need a few simple apps and services. I could make a strong argument that Google’s services over-serve this next coming two billion. This is where handset OEMs see an opportunity to move away from what they view as a stranglehold Google has had on them. And I believe this opportunity exists for them as well. This is where Cyanogen comes in.

What makes Cyanogen interesting is that an OEM can choose to use them to create a mobile OS solution that includes Google services or not. Cyanogen offers choice in a way Google refuses to for handset OEMs. In this coming era, where the business model is shifting from hardware to services, choice is going to be essential. Hardware companies have to learn to be services companies and services companies have to learn to become hardware companies. When the bulk of new smartphones sold in the next four to five years are at prices less than $150, money will not be made in hardware but made in other areas. Cyanagoen offers handset brands, or services companies looking to get into handset brands, an opportunity to layer value on top of Android in ways Google does not want to allow.

The other key point for Cyanogen is how, outside a very small few handset OEMs, most have no software expertise. Companies like Samsung, HTC, LG, and Xiaomi, have some internal capabilities to customize Android software but most do not. Cyanogen offers hardware companies or services companies who want to get into hardware a customizable and turn-key solution with or without Google’s services. Outside of Cyanogen, I’m not sure who else is offering such a thing.

The counter to Cyanogen is that if Google figures this out and adjusts their strategy accordingly to offer a similar package like Cyanogen. Or if Google adjusts their policies and embraces open Android in new ways. However, what we are observing is Google is trying to control and actually “close” open Android with each new Android update. While Google thinks this deals with the open Android issue, it really just validates it and bucks the overall trend rather than embrace it. In fact, what Google is doing to control open Android is actually pushing more people toward Cyanogen than away.

One last point for Cyanogen. Google is not in China, yet the Chinese ODM ecosystem is essentially opening up the equivalent of contract manufacturing for smartphones. I’m fond of saying it isn’t the Chinese brands likely to go global but the Chinese manufacturing ecosystem under the guise of many local and regional brands. Wico in Europe, Micromax in India, SmartFren in Indonesia, and Cherry Mobile in the Philippines. Cyanogen has been able to penetrate the Shenzhen ecosystem both through relationships with Qualcomm and MediaTek. This is a huge deal, and one that nearly seals Cyanogen’s opportunity to actually be a third OS in this next phase of mobile.

I’m convinced Google is going to focus more on strategies to increase their monetization tactics of their existing, most profitable Android customer base. The only variable that could alter Cyanogen’s plans is if Facebook provides a similar solution. Yet, I think it is more likely Facebook is a potential Cyanogen partner rather than competitor.

New Hardware Means a Lot, but Software Means More


I spend what is probably a silly number of hours a day reading, or at least glancing at, tech articles on the internet. They are concerned endlessly with hardware. The problem is, it is software and services that really matter in an era where quality hardware, especially for smartphones and PCs, is widely available.

apple-watch-inviteThis will surely be obvious on Monday when Apple holds its scheduled announcement on the Apple Watch. It is an unusual event since Apple introduced the product in September. The Apple Watch was described and shown off in much detail, including all three designs and two sizes. What Apple has to talk about on March 9 is the apps and services.

As has so often been the case, the Apple Watch will enter a field dominated by dozens of competitors, from Pebble, Samsung, Google, and Motorola to Chinese manufacturers that appeared at CES without advance warning. We will see how the competition does once Apple announces more details of Apple Watch on Monday.

Apple has clearly worked intensively and for years to top the field of watch designs. If you want the depth of detail on the effort, read The New Yorker‘s profile of Jony Ive, “The Shape of Things to Come“. But as much time as Ive denotes to the physical design, he is also obsessed by the software and services, probably more so than ever after the death of Steve Jobs.

Apple has understood software is at least as important as hardware for decades. The iPod was better designed than all the MP3 devices in the market but it needed both its appearance and the services of iTunes to steal the market.

The iPhone was the notable exception. Apple brought the first iPhone to market as a phone of limited voice quality with weak data services, just one carrier, limited availability of software options, and no provision for the addition of third party apps. The iPhone had a lot of attraction for being a real, if limited, handheld computer. But the market was dominated by BlackBerry, with a great voice phone, keyboard, and email. It was only in the second year, when Apple added third party apps and 3G wireless, the iPhone really took off.

As the apps grew, the iPhone smashed the competition. BlackBerry tried to attract third parties, but its operating system and limited display made the work very difficult. After a very rough start in its start up years, Android has finally come up with a flood of apps, though it is still better at matching iPhones through quantity than quality. Samsung has finally come to realize with the Galaxy S6 the best way to improve its software is to dump the lousy apps it had included in its recent offerings. Meanwhile Microsoft, with a new operating system and Nokia designs, has much improved the Windows Phone but it still struggles for third party apps.

Apple’s history has been to move late, from MP3s to tablets, into markets where many competitors have tried but failed to accomplish much and succeed with good design and even better software. Stay tuned for this to happen again with watches.


Research/Data: Phablet Impact Continues to Grow

The device market continues to evolve and grow, following paths that aren’t always as obvious as one might presume.

PCs were dead, and now they’re not. Tablets were taking over, and now they’re not. Phablets were only going to be for a small minority, and now they’re not.

Despite the challenges, one of my roles as an analyst is to create market forecasts and try to bring some sense of order to what often seems like chaos. In the process of creating my firm’s latest predictions for smart connected devices (a term I coined while I was an analyst at IDC to describe the combination of PCs, tablets and smartphones), it became obvious to me that one over-riding factor is now driving a dramatic shift in the device landscape. In a word, “phablets,” which I define as large smartphones with screens that are 5” and greater.

The phablet phenomenon has only had a modest impact so far in the US, but in many other parts of the world—particularly the fast-growing Asian markets—large smartphones are quickly become the common choice of most smartphone buyers. For some of those buyers, in fact, a large smartphone is their first true computing device and serves as their entré to the wonders of the Internet.

While in the US, phablets represented just 13.1% of Q4 2014 shipments, worldwide they already captured 27% of smartphone shipments. In fact, with shipments at just over 100 million units last quarter, more phablets shipped than all PCs or all tablets for the same time period

Plus, the momentum for the category is really only just starting. Here in the US and certain other markets, the 5.5” iPhone 6 Plus finally “legitimized” the phablet market and the momentum behind it is now enormous.

Not surprisingly, I’m predicting very strong growth for large smartphones and believe they will surpass 1 billion shipments in 2019. In fact, they’re the sole reason the smartphone market is continuing to grow because I believe small smartphones (those with screens under 5”) peaked in 2014, and will continue to decline through the end of the decade.

Phablets are doing more than just driving smartphone market growth, however. Large smartphones have also had a negative impact on tablets. Worldwide tablet shipments barely eked out any growth last year and US tablet shipments fell between 2013 and 2014. The small tablet market (defined as those with screens from 7”-7.9”, like the iPad Mini), in particular, is already taking a beating from larger smartphones and the situation is expected to get worse over the next few years. Many people are finding that large smartphones can easily take the place of smaller tablets for most of their activities. Plus, large smartphones have the added benefit of an always-on data connection through the cellular network. As a result, I’m forecasting that worldwide tablet shipments peaked last year and will continue to decline over the next 5+ years.

Interestingly, because of this decline, I believe phablets are also impacting the PC market. When people were predicting that tablets would take over for PCs (the Post-PC Era anyone?), the PC market suffered. Now that it’s clear that wholesale PC replacement is not going to happen, we’re seeing stabilization worldwide and even modest growth in the US PC market. Many people are discovering that PCs and large smartphones make for great companions and meet all the computing and mobility demands that most people have.

The chart below summarizes my new TECHnalysis Research forecast for all the major smart connected device categories through 2020. As you can see, the phablet phenomenon is not only real, it’s reshaping the entire device landscape.


You can find out more about the TECHnalysis Research forecast by reading the press release here and downloading a summary presentation of the results here.