Arrivals and Departures

I was listening to an episode of the Awl podcast this week. Guest Clay Shirky said something that struck me as important and profound. He was speaking specifically in the context of media (the Awl’s primary subject matter), but it seemed applicable throughout the tech world, too. Here’s the quote as I’ve transcribed it:

The arrival half of the media revolution happened much, much faster than the departure half.

He used email and the fax machine as an example of this phenomenon, suggesting that, even though it was immediately apparent to almost everyone that email was superior to faxes, fax machines were still being sold in this decade. In other words, even though email took off very rapidly, it took much longer for it to kill off one of the technologies it replaced. By implication then, what ends up becoming substitution very often looks like complementary or coexistence for an awfully long time at first. All of this got me wondering about what arrivals we’ve already witnessed and the departures they will ultimately cause but which might not yet be apparent.

Some recent arrivals

Let’s start out by thinking about some recent arrivals, although I’m going to use a fairly zoomed-out definition of “recent” for the purpose of this analysis. Consider the following list of technologies and products, ordered vaguely by their date of arrival in the mainstream consciousness. What departures might these arrivals eventually cause? Which have they already begun to cause?

  • E-commerce
  • Google search
  • Over-the-top video services
  • Facebook
  • Google Docs
  • Cloud computing
  • Wireless headphones
  • Amazon Fresh
  • Facebook Video
  • Uber
  • Drones
  • Smartwatches
  • Virtual Reality.

Look at the earliest items on that list – Amazon was founded in 1994, right at the beginning of the web era, along with other early e-commerce sites. But it took quite a lot longer to have a serious impact on brick and mortar retail and it was arguably only this year Amazon began to have a really broad-based impact on traditional retailers’ overall sales (as opposed to sales in specific categories like books). The impact of Google search was felt more quickly, especially by its direct competitors, but even then it took several years to reach its dominant market share and arguably even longer to have the disruptive impact it’s had on other indirect competitors. Over-the-top video services like Netflix debuted in the mid-2000s (with Netflix launching streaming in 2007), but it was only in 2015 pay TV cord cutting finally became a reality. For much of its history, Netflix existed as a complement rather than a substitute for pay TV, but that’s now starting to change.

Now look at some of the later items on the list. Most of these haven’t been around long enough to have prompted any departures, though there are exceptions, including Uber’s impact on cabs. But which products and technologies might these more recent arrivals have an impact on over the longer term?

Reasons for delayed departures

It’s also worth asking why the timing of arrivals and departures is often offset so significantly. I think there are several reasons:

  • Replacement cycles and sunk costs: Just because a new product is clearly superior to an older one, doesn’t mean we immediately throw the old one away and replace it. For both consumers and businesses, replacement cycles are real (though consumers may not articulate them in the same way) and sunk costs also cause people and companies to milk existing inferior technology for a period of time before buying the superior replacement.
  • Standards and culture: To take the example of Google Docs in the list above, it was launched in 2007 by Google (following the acquisition of the company that created its predecessor), but it took a long time to have a meaningful impact on Microsoft Office. One big reason is the Office applications are embedded in most companies and enjoy a position as de facto standards. Company and social culture often impedes the adoption of new products and technologies despite their obvious appeal.
  • Security and other concerns: New products are often considered to have inferior security or other features that make them less desirable in other ways, even if their core features and performance are superior. Competitor FUD often feeds these worries but, in many cases, the concerns are real. Over time, both the real and imaginary concerns are addressed, but that takes a while.
  • Apathy: I’ve often said apathy (and, perhaps more kindly, a preference for the known or the safe) is a significant force in adoption of new technologies and products. People would often rather stick with what they know rather than go through the personal disruption inherent in adopting a new product and learning how it works.
  • Absence of a trigger: Sometimes the transition from old to new doesn’t happen until there’s some significant event that drives it. Consider adoption of wireless headphones – they’ve been around for years now, the technology is pretty good, and yet, the vast majority of us are likely still using wired headphones. But what if Apple suddenly does away with the 3.5mm jack in the next iPhone, thereby putting into the market over the ensuing year a couple of hundred million new devices which won’t work with the headphones consumers already have?

Once these delaying factors are overcome, it’s worth noting that, once the departure phase of the incumbent technology begins, it often accelerates rapidly, taking the incumbents and observers by surprise, especially if they’ve convinced themselves the new thing was, in fact, a complement rather than a substitute. It’s almost as if these technologies, having been delayed in their departure, are determined to make up for lost time.

Upcoming departures

We began by reviewing “recent” arrivals, so let’s end by reviewing some upcoming departures. Some may be far more obvious than others. Feature phones are clearly on the way out, thanks to the emergence of smartphones over the past 10-15 years. Pay TV in its current form seems on its way out too, though exactly what form its replacement will take is still up in the air, and it may look more similar to the current model than some people are assuming. Sometimes we can’t see the potential for a departure because we’re still mentally seeing a technology as an arrival. But what about smartphones themselves? I listed both smartwatches and VR among my recent arrivals earlier – both of these along with other input and display technologies could eventually obsolete some of the core functions of the smartphone or at least turn it into something very different. How soon will that happen? Not for a number of years but it’s quite likely that, when the time does come, it will take a lot of people by surprise and happen far faster than many of us imagine.

Published by

Jan Dawson

Jan Dawson is Founder and Chief Analyst at Jackdaw Research, a technology research and consulting firm focused on consumer technology. During his sixteen years as a technology analyst, Jan has covered everything from DSL to LTE, and from policy and regulation to smartphones and tablets. As such, he brings a unique perspective to the consumer technology space, pulling together insights on communications and content services, device hardware and software, and online services to provide big-picture market analysis and strategic advice to his clients. Jan has worked with many of the world’s largest operators, device and infrastructure vendors, online service providers and others to shape their strategies and help them understand the market. Prior to founding Jackdaw, Jan worked at Ovum for a number of years, most recently as Chief Telecoms Analyst, responsible for Ovum’s telecoms research agenda globally.

13 thoughts on “Arrivals and Departures”

  1. I’m unclear if we’re talking about individual products (“Google Search”, not Altavista, Bing, DuckDuckGo…) or whole categories (“wearables”, not Gear S, gGlass, Fitbit…).

    There’s also a reporting bias. Journalists/Bloggers like new stuff, more ad $$, more thoughtful pondering to do, and more news to disseminate. Media coverage is usually disproportionate to economic importance: Amazon Fresh might be new, my corner butcher and grocer have always done deliveries, and my supermarket has for over a decade.

    I’m also doubtful specifically about wireless headphones: I’ve bought mid-range ones alongside my also mid-range wired pair for over two decades, yet the charging requirement mostly condemns them to a drawer, I just can’t be arsed to charge Yet Another Gizmo. I’m also told there are still sound quality (even AptX is re-compression on top of the MP3/AAC compression, why don’t they just send the source file ?) and connection reliability issues.
    Plus wired headphones are required as antenna for phones with an FM radio (we could also discuss whether FM is superior to online for radios, maybe during a hike, that usually makes for a very short debate).
    To portray wireless headphones as better new entrants only kept down by lack of impetus is overreach, I think. And not similar to replacing Serial, Parallel and ADB with USB: that switch added no charging requirement and no quality/performance degradation, on the contrary. Plus even LapLink had a USB version ^^

    1. Wireless anything is great until it isn’t. Then it just sucks it big time. You don’t even get to splice over the problem. And for anyone who cooks, anonymous delivery of groceries is always a variable most cooks would rather not depend on, never mind the simple joy of grocery shopping, which is one of my little pleasures in life. And I’m not entirely sure how drones are an arrival type product in the sense of the over all article.

      That said, while I do think some of the examples of potential AR/DP are questionable, the concept is quite intriguing. In the example of email vs fax, what made email superior was when scanning became more ubiquitous and easily accessible as a digital file when copiers and faxes became networkable. There may be some ingredient missing in the other examples as well.

      “We have to arrive to depart. But we have to depart so we can arrive”, Snake Dupree arriving at airport in the movie _Big Trouble_.


      1. re: headsets, maybe Apple could do things well:
        1- send straight AAC directly to the headphones, apparently that’s technically feasible, headphones and phones can agree on any codec/format
        2- support both wired and wireless operation, at least on over-the-ears models, so that if you forget to charge you’re not stuck musicless (the horror !) but can just hook up your trusty cable.

        Bonus question: will said cable be included or a $29.99 extra ^^

      2. I think the main issue with AR/DP is selection bias. Of course stuff that has been successful takes a long time to disappear: it has been successful for . Most stuff never arrived, or barely. AOL didn’t take that long to depart, neither did Clippy, nor Newton (got to get my Apple jibe in ^^).
        And the category vs individual product confusion doesn’t help either: Ashton Tate’s Framework, Lotus’ 123 and Agenda, even dBase and Turbo Pascal did depart as products, yet they created whole categories that are relevant and alive to this day. Which reminds me, where’s Notes at ? :-p

        1. What I take AR/DP to mean from the original example is Arrival=people are consuming/downloading video in ways that the content presenters are hardly able to provide (the Departure half).

          I think Napster is a better (though not necessarily as a sustainable business model) example of Arrival out pacing Departure. The consumer is getting the product no matter how the provider is actually providing, almost in spite of.


  2. This is clearly a discussion that Christensen’s Disruption Theory explicitly addresses. I have no intention to go into that here, but I just want to point out a huge omission in this argument.

    In the article, it is assumed (except from a single isolated sentence) that the new technology is superior to the old one from the onset, and that none of the “reasons for delayed departure” pertain to the immaturity of the new product itself. This, in the vast majority of examples is totally false. In fact, the immaturity of the entrant product and the potential to subsequently improve, is a huge part of what makes a product “Disruptive” in the Christensen sense.

    So let’s look the example that has been around the longest, and that is E-commerce.

    Amazon may have been around since 1994, but remember that at that time, most of us were accessing the Internet on 33MHz 80486s with 4MB RAM on 14-inch desktop CRTs connected via 28.8 kbps modems. Not to mention the vast improvements in parcel delivery, that were most likely enabled with adoption of IT by the delivery industry. Depending on how much you wanted to shop around, driving to the local mall was arguably faster than waiting for the pages on Amazon to load. Add slow delivery times, and it’s no wonder that Amazon wasn’t as popular as it is today.

    Now anybody who has been paying attention knows that although e-commerce sales are cutting into retail, it is still less then 10% of total retail in the USA. Hence it is likely that we need even more technical advances if e-commerce is to “replace” traditional brick & mortar retail. VR may play an important part here.

    The other examples are more difficult to address since they have been around much shorter. However, I am sure that the success that they have had up till now is strongly related to improvements in the product, or the infrastructure that surrounds it. For example, Google Docs would be much less convenient without good wireless connectivity. Facebook would be far less important without widespread smartphone adoption.

    The current article looks at the technology adoption lifecycle argument (early adopters, laggards, etc) for an explanation of “late departure”. However this is hardly the only angle one should look at. You should also be looking at how the products themselves improve, and the factors that enable that.

    At the same time, you should also look at the products that have little propensity to significantly improve further, because these are far less likely to replace the incumbent even in the long term future. For example, one should wonder how Moore’s law could increase the gap between regular taxis and Ubers. If you can’t come up with a good idea, then Uber might not replace taxis after all.

    1. Interesting that you think the obstacle for e-commerce is technological and not sociological or anthropological.


      1. It depends on what the technology is capable of. Regarding the sociological factors, if you could go shopping with friends and family but do that virtually, as a social experience, from your living room, then they may be overcome. Of course, a simple TV app alone is unlikely to be sufficient.

        On the other hand, listening to music used to be a social experience with every family member listening to the radio in the living room. Now it is more often than not, very private. Technology can change what was once a social experience into a private one.

        Things change, and technology strongly influences these changes.

        I’m not sure what you mean by anthropological factors.

Leave a Reply

Your email address will not be published. Required fields are marked *