The Smartphone Wars: Android measured at over 50% U.S. market share

Two consecutive smartphone-wars posts is unusual here, I know, but the latest marketshare news is high-explosive stuff. The Guardian in Great Britain has published data from a European market-research outfit, Kantar WorldPanel Comtech; the article is titled Nokia and RIM bleeding smartphone share while Android cleans up, and includes a very informative table of smartphone share numbers by country. They become even more interesting when set against the ComScore numbers covering Nov 2010 to February 2011, which I previously analyzed in The Smartphone Wars: Almost boring now….

The article does not say outright whether Kantar is reporting new-unit sales or total installed base, but the Kantar November 2010 figures for Microsoft, RIM, and Apple are so close to comScore’s from July 2010 that I think it must be the total installed base. Their differences max out at about 4%, which is not far above statistical noise level, and fit their known market-share trend lines for that period. Android is 19% in Kantar’s June 2010 results vs. 26% in comScore’s November 2010 ones, which can be accounted for by the (known) phenomenal growth between June and November.

So it appears the comScore and Kantar numbers are comparable. But if I’ve been fooled by a numerical coincidence, that’s not going to make a long-term difference. In a market turning over technology and units as fast as smartphones are, installed base doesn’t lag new-unit sales by a whole lot. If the Kantar figures aren’t describing total installed base now, they’re predicting it in 3 to 6 months.

At the risk of reducing the rest of this post to anticlimax, I will first report that Kantar measures Android market share in the U.S. at 54.7% in March 2011. This is a 21% jump from comScore’s February figures in a single month. It suggests that, a good six months sooner than I expected, Android is entering the final runaway to uncontested market dominance at 85% or up.

It’s certainly not looking like any competitor has what it takes to stop Android. Kantar has Symbian at 1.2%, Microsoft at 4.9%, RIM at 10.6% and Apple at %27.2. While regaining second place is good news for Apple, this is mostly because RIM is going down in flames; a gain of less than 2% since February isn’t going to do diddly to Android when Android’s growth rate is literally an order of magnitude higher.

Results from other countries show local variations on a similar pattern. Android is hammering the snot out of its competitors everywhere. The main difference from the U.S. is that while iOS is still posting tiny share gains here, it’s getting badly hurt elsewhere. In France it has actually lost 23% of share in the last year! It’s hardly doing better in Great Britain; -16%.

There are fewer surprises in the analysis above the table. “The key underlying trend is that Android is growing in every country,” said Dominic Sunnebo, described as “consumer insight director” for Kantar. He zeroes in on the Android army’s faster OODA loop and ability to place lots of different product bets at different price and capability points – quite a familiar theme to readers of this blog. He also identifies low Android cost as a major sales driver outside the developed world.

In a recent post I projected that serious zero-sum competition between Android and Apple wouldn’t begin until at least 3Q2011 after Android has chewed up the market share of all the soft targets – RIM and Microsoft, most notably. Kantar’s non-US figures give us a foretaste if what that’s going to be like, and they suggest that I may actually have overestimated Apple’s ability to retain share under Android pressure. It looks as though the disruptive collapse of iPhone has arrived in Europe.

The figures also point an increasingly bleak picture for the NoWin alliance, which has failed to stop both Nokia and Microsoft from incurring severe losses in market share. It is beginning to look as though the earliest possible time Nokia could ship WP7-based smartphones is going to be well after Android has achieved a strong supermajority. Since the minority competitor seems certain to be Apple, it’s difficult to locate a plausible point of entry for NoWin phones.

157 comments

    1. >and ask that I (please) be removed from the moderation queue.

      You aren’t in it. If your postings are getting binned, it’s because akismet thinks they’re spam.

  1. At the risk of reducing the rest of this post to anticlimax, I will first report that Kantar measures Android market share in the U.S. at 54.7% in March 2011. This is a 21% jump from comScore’s February figures in a single month. It suggests that, a good six months sooner than I expected, Android is entering the final runaway to uncontested market dominance at 85% or up.

    I think it’s fair to say that Android is rapidly reaching “unstoppable juggernaut” status. The last time we saw platform growth this fast in the iT industry was probably in the 1990s, with the rise of the then-800 lb. gorilla, Microsoft.And Android is achieving that dominance in much the same way Microsoft did — by having so many different OEMs and, as a result, being able to react to changing demand in the market more quickly and efficiently.

    Only the surprising thing is the timescale on which it is happening — it’s like a super-accelerated version of the Mac vs. PC wars. This also means that any predictions on the timescale in which smartphones will replace PCs will have to be adjusted as well because it shows, I think, an accelerated technology development curve.

  2. In the period from June 2010 to December 2010, the number of Android activations per day went from 160k to 300k. In the period from December 2010 to April 2011, the number of activations per day went from 300k to 350k. I would’ve expected Android activations to be at 400k per day by now, given that the smartphone market is still rapidly expanding.

    To compare: Apple went from selling 93k iPhones per day in the spring 2010 quarter to selling 180k iPhones per day in the fall 2010 quarter. For the winter 2011 quarter that ended on March 26, Apple is expected to have sold upwards of 200k iPhones per day. For a company that only releases a new generation of phones once a year, in summer, a slowdown of growth in winter is to be expected.

    OTOH, new Android phones are released all the time, so why the slowdown?

  3. Actually, Eric, I think it’s phones sold for the month, not installed base. They even have a total sales number that matches.

    Interestingly, according to this report, for the month of March, Android only outsold Apple 2 to1, but the similar number I extrapolated for the average of Dec/Jan/Feb from comscore was 3.25 to 1. So Verizon might actually have been a pretty good bump for Apple, although the 3.25 to 1 number wouldn’t count somebody replacing an Android phone with an Android phone or an Apple phone with an Apple phone, so doesn’t really track shipments that well.

    Still, with the caveat that we don’t know how many are replacements, you are absolutely right about the acceleration — 13.3 million new phones into a market with 69.5 M installed phones is 19% of the base in a single month.

    1. >Actually, Eric, I think it’s phones sold for the month, not installed base. They even have a total sales number that matches.

      You may well be right. However, very little of my analysis changes under that assumption. I reread carefully to check, and about all your theory does is make Android’s rise in March look a little less impressive and stretch out the expected time to supermajority by a few months.

  4. (btw, was talking about the US market in comment above)

    We can probably get close to estimating the Verizon bump, with this report and the comscore report, and the additional piece of information that a bit over a million iPhone Vs were sold in the last few weeks of Feb (which would have skewed the comscore moving average a bit), but I’m too tired to do that right now.

    Also will note that, with Apple looking this strong, if the other competitors disappeared today and the Android/Apple split in the US stayed static, the installed base would level off at 66% Android / 33% Apple.

    But that’s probably partly due to the one-time Verizon bump; other countries put Apple’s share at only 23 – 24% right now.

    I’m surprised at how well RIM is doing in most of Europe. It’s almost like some new data security directive made all the corporations sell off their Nokia and Apple phones for pennies on the dollar and invest in RIM phones :-)

  5. @Richard:

    > I would’ve expected Android activations to be at 400k per day by now,

    I think they must be. That report didn’t include any of Asia except Japan, and if you add up all the Android phones sold and divide by 31 days in March you get over 390K/day.

  6. > However, very little of my analysis changes under that assumption.

    Yeah, except this:

    > Android is entering the final runaway to uncontested market dominance at 85% or up.

    The whole exclusivity thing throws a monkey wrench in the analysis for the US, and the substantial share by other vendors (and RIM refusing to budge) throws a monkey wrench in the analysis for the rest of the world, but, for example, if you take my extrapolation from the comscore report that Android was outselling Apple by 3.25 to 1 in the previous quarter, you need that number to change to 5.7 to 1 to get to the 85%. It changed significantly in March, but downward, to 2 to 1, and it’s currently much lower in Europe — 0.91 to 1 in Italy (granted, not a significant number of phones).

  7. Only the surprising thing is the timescale on which it is happening — it’s like a super-accelerated version of the Mac vs. PC wars. This also means that any predictions on the timescale in which smartphones will replace PCs will have to be adjusted as well because it shows, I think, an accelerated technology development curve.

    I don’t see it as particularly surprising. The standard depreciation schedule for a PC is 5 years, and even the vendors haven’t been able to push anybody that I know of onto a faster than 3-year cycle for desktops. Smartphone owners can generally get a subsidized upgrade every 18 months—and not just smartphone owners but also “feature-phone” owners who decide to spring for a beefier model this time around. When you consider that the turnover pace is so much faster and that the fence between market segments is so low (smartphone/feature phone vs. computer/…television?), once this one tipped, it’s easily retroactively foreseeable how quickly it’s falling.

  8. @Patrick Maupin
    “That report didn’t include any of Asia except Japan, and if you add up all the Android phones sold and divide by 31 days in March you get over 390K/day.”

    Which is eerily close to predictions using an exponential growth function with a activation growth factor of 3.83/year, which gets to 395,000/day at the end of March:
    http://esr.ibiblio.org/?p=2961#comment-297134

    Extrapolating to the end of the year you get a million activations a day in the last week of the year and a total installed base of around 250 million Android phones. And an installed base of 1 Billion is reached in around 2 years from now. But only if all circumstances remain the same, which they won’t.

    @Patrick Maupin
    “66% Android / 33% Apple.” Or 23-24% Apple in other predictions.

    Be bold! Smartphones will replace dumbphones. That is 5 billion new low-low cost smartphones. They will neither run iOS nor QNX. And it is extremely unlikely that they will run WP7. So there will be 4.5 Billion Android phones and at most 0.5 Billion other phones. That is 90% vs 10% at the best for Android vs iOS.

  9. @Patrick> I think they must be. That report didn’t include any of Asia except Japan, and if you add up all the Android phones sold and divide by 31 days in March you get over 390K/day.

    The only problem with your math is that it can’t be supported by Google’s own claims:

    In last Thursday’s (4/14/11) conference call on (announcing earnings), Google’s Jeff Huber reported that Android is now recording 350K activations per day. Android hit 100K activations per day in May 2010 and 200K activations a day in September 2010. We can assume that
    the 350K/day number is quite recent (March 2011 at the earliest). I believe Google’s numbers, not yours.

    So from May to Sep (4 months), Android doubled in activation rate. But from Sep to March (6 months), the activation rate only increased another 75%. The growth is *currently* slowing, not accelerating. (Which also shows Winter’s predictions above to be misinformed.) Perhaps something will happen, and Android’s growth will recover to its former rate, but I’ve not seen any ideas for what might allow this.

    Apple’s slowing growth can be explained by it’s lack of new phones. Many of Apple’s existing and potential customer base were waiting (in vain) for a new iPhone variant (commonly referred to as the ‘iPhone V’, though for all anyone outside Apple knows, it’s named the iPhone 4GS or similar) to be announced in June. We now know that Apple won’t announce any new iPhone hardware at WWDC.

    Conversely, the various OEMs have new Android handsets coming out all the time. The carriers should always be able to sell the newest and greatest Android phone to their customer base.

    Many things are possible. We could have reached some (temporary) point of saturation in the Smartphone base. Customers could be holding out for better offers. It could be something else.

    But there is no pretending that the growth rate of both Android and iPhone haven’t slowed. (Apple should announce its activation rate during its earnings call Wednesday afternoon.)

  10. > You aren’t in it. If your postings are getting binned, it’s because akismet thinks they’re spam.

    Whatever it is, I can’t post a comment to this blog without being told “Your comment is awaiting moderation.”

    1. >Whatever it is, I can’t post a comment to this blog without being told “Your comment is awaiting moderation.”

      That’s what happens when akismet thinks your comment is spam. Sometimes (I don’t know why) I don’t even see such comments.

      It could also happen if I had flagged you as a nuisance. I haven’t done that.

  11. A small nugget not in the main article:
    Kantar’s figures suggest that only 1m Windows Phone devices were sold since they launched – around half the number that Microsoft has repeatedly said have been “shipped”.

    As we already knew, MS was lying big time with their “2 million shipped”. Only 1 million WP7 phones sold! That is only 3 days of Android activations.

  12. As we already knew, MS was lying big time with their “2 million shipped”.

    Nah, they did ship 2 million. Half are sitting in the clearance bin.

  13. > While regaining second place is good news for Apple, this is mostly because RIM is going down in flames; a gain of less than 2% since February isn’t going to do diddly to Android when Android’s growth rate is literally an order of magnitude higher.

    Companies don’t run on market share, they run on profit and cashflow.

    Business is all about trade, the exchange of value between two or more parties, and cash is the asset needed for participation in the economic system. For this reason – while some industries are more cash intensive than others – no business can survive in the long run without generating positive cash flow per share for its shareholders. To have a positive cash flow, the company’s long-term cash inflows need to exceed its long-term cash outflows. The concept of profit is somewhat narrow, and only looks at income and expenses at a certain point in time. Cashflow, OTOH, is more dynamic. It is concerned with the movement of money in and out of a business. More importantly, it is concerned with the time at which these movement of money takes place.

    If RIM can continue to be profitable *and* generate positive cashflows while bleeding market share, more power to it.

    How,, it’s going to be able to manage to do that when it’s losing the upper end of the market, is a big question. How it’s going to retain customers during a major platform shift is another question.

    Eric may wish to look at the financial health of the major vendors of Smartphones.

    1. >If RIM can continue to be profitable *and* generate positive cashflows while bleeding market share, more power to it.

      This is very difficult in a market with fast turnover and strong network effects, though. That’s the lesser part of the reason I keep focusing on market share; in such a market, share losses tend to be self-reinforcing and slipping below a certain minimum share threshold is fatal. The greater part of the reason I continue focusing on market share is that my primary interest is in how open the smartphone infrastructure will be when the war is over. Thus, who the winner is matters a lot; how much money gets made along the way and by whom, not so much.

  14. Another factor that may be accelerating the technology cycle is that everyone is anticipating it. In this environment everyone is expecting disruption from below followed by network-effect lock-in. Android may be getting the full benefit of dominant share, not once it is achieved, but *as soon as everyone anticipates it will be achieved*.

    App count dominance is quite irrelevent; the question is what everyone expects app store sizes to look like a year from now.

  15. It’s almost as if Eric picks and chooses the sources he quotes based on whether or not they agree with him. Which is interesting in the case of Kantar, given this from a week or so back.

    The iPhone 4 is the top selling mobile handset on Verizon, according to the latest data released by Kantar Worldpanel ComTech, taking 30.1% of Verizon’s consumer mobile handset sales. The figures have been published for the four weeks ending 20 March 2011.

    1. >It’s almost as if Eric picks and chooses the sources he quotes based on whether or not they agree with him.

      I ‘chose’ the report from the Guardian because it bubbled to the top in Google News. I was unaware of the previous Kantar report.

      The report that iPhone 4 is the top selling mobile handset on Verizon has to be set against evidence noted here previously that it is being outsold by the HTC Thunderbolt. I am not aware of evidence contradicting Kantar’s most recent report, but this is perhaps because it is too new.

  16. The OODA is why Android beats Apple, but it is the ecosystem or lack thereof that will kill the rest and also determine the ultimate share.

    Both Android and iOS have devices for meidaplayers and tablet, over the air, wireless, or drag and drop to manage them. If you use a smartphone 80% as something else, having the 80% only in a less expensive or alternate device makes sense.

    Win Phone 7 will die if they don’t have millions of devices by summer – and they aren’t going to be phones. They need a Zune and tablet that run the phone apps, subsidize it like the Xbox if needed so that they can kickstart their ecosystem. Maemo-Meego might still be larger. I don’t think they will do this, but that is what it will take to remain relevant. Turning even every Nokia feature phone into a WP7 smartphone even if they do everything else perfectly will only drop sales for Nokia. (Not that I trust the company that did “plays for sure” will keep the ecosystem alive, only that requiring me to buy a phone or $500+ device unsubsidized to get into their ecosystem is a nonstarter, even if it has some unique advantages like xbox-live and good microsoft office integration).

    Android has the iPod Touch and iPad equivalents. Lots of them. I think the kitchen appliances shown at CES ran Android. This is part of the OODA loop – Phones are upgraded every 2 years. The cheaper media devices can go every few months.

    RIM’s playbook isn’t even part of an ecosystem.

  17. @Bryant
    “The iPhone 4 is the top selling mobile handset on Verizon,”

    Eh? The report talks about 26 million phones sold. How many of those were sold by Verizon? Even in the USA, Android phones were sold twice as often as iPhones.

    @tz
    “If you use a smartphone 80% as something else, having the 80% only in a less expensive or alternate device makes sense.”

    And you are predicting the Android “gadgets” will be more expensive than the iOS or WP7 gadgets? Why?

    @tz
    “Win Phone 7 will die if they don’t have millions of devices by summer – and they aren’t going to be phones. They need a Zune and tablet that run the phone apps, subsidize it like the Xbox if needed so that they can kickstart their ecosystem. ”

    At the end of the year an expected 250 million Android phones will have been sold. To counter that, MS will have to subsidize WP7 appliances in the multi billion dollar range. Make that tens of billions of dollars. I expect the stock holders to balk at the mere thought of that. Half the apps have something to do with phones/GPS/Internet access. The Zune does not come to mind as the best gadget for such apps. And who would build these tablets? Millions of WP7 tablets?

  18. @Winter:

    > Be bold! Smartphones will replace dumbphones.

    Of course. But not in the next 3 or 4 months.

  19. Winter: compare and contrast “Alas, [iPhone 4] sales out of the gate were unimpressive.” with “The iPhone 4 is the top selling mobile handset on Verizon.”

    1. >Winter: compare and contrast “Alas, [iPhone 4] sales out of the gate were unimpressive.” with “The iPhone 4 is the top selling mobile handset on Verizon.”

      Let’s suppose the second claim is true (there are reasons for doubt). Why treat these claims as facially inconsistent? It could be that the Verizon network is simply not very good at generating handset sales, so even “top selling” doesn’t signify a lot. I can think of several reasons this might be the case.

  20. @Bryant
    “compare and contrast “Alas, [iPhone 4] sales out of the gate were unimpressive.” with “The iPhone 4 is the top selling mobile handset on Verizon.””

    I do not see the contradiction:
    – iPhone4 was not *only* sold on Verizon. And elsewhere it was *not* outselling Android
    – Was the iPhone4 a hit “out of the gate” even at Verizon?

    If the iPhone4 was only a smashing hit with Verizon, and no-where impressive “out of the gate”, then eric is still correct. But maybe you know better, and iPhone4 was a real hit everywhere in the world, out of the gate?

    Then the numbers in the linked paper must be completely wrong.

  21. @Bryant:

    Winter: compare and contrast “Alas, [iPhone 4] sales out of the gate were unimpressive.” with “The iPhone 4 is the top selling mobile handset on Verizon.”

    Context is everything. From all we know now, they seem to have been unimpressive compared to what had been forecast. 30% really isn’t all that great for the long-awaited Messiah smartphone. According to pundits, you should have been able to expect 60% from people who said “oh, I can get an iPhone for the same price as an Android? Why not?” and that’s before the much-touted wave of millions of refugees from AT&T’s horrific network.

    It’s almost as if Eric picks and chooses the sources he quotes based on whether or not they agree with him. Which is interesting in the case of Kantar, given this from a week or so back.

    The iPhone 4 is the top selling mobile handset on Verizon, according to the latest data released by Kantar Worldpanel ComTech, taking 30.1% of Verizon’s consumer mobile handset sales. The figures have been published for the four weeks ending 20 March 2011.

    Nobody said that there wasn’t some pent-up demand for the iPhone for Verizon. But speaking of picking and choosing, the same article you quote from also says “We’re expecting Apple’s leading share on Verizon to drop back over Q2, unless we see a price drop for the iPhone on Verizon. Currently, Android devices are satisfying Verizon consumer demand amongst lower cost Smartphone handsets.”

    But we might as well just grab some popcorn and sit back for a few days:

    http://www.fiercewireless.com/story/verizons-iphone-effect-dominate-q1-carrier-earnings/2011-04-13

  22. @Richard Thompson:

    BTW, I get the waiting in the moderation queue thing sometimes, and then it may show up on its own 10 hours later. Weird.

    > The only problem with your math is that it can’t be supported by Google’s own claims…

    Possibly. Sales are not the same as activations. Apparently google only counts people who sign up for google services (like the marketplace) and does not count phone upgrades. It may be that as more people who were carrying dumbphones convert, the percentage of eager app users goes down. Or it may be, as you speculate, that growth stalled somewhat, which could partially be due to pent up demand for CDMA iPhones.

    1. >Possibly. Sales are not the same as activations. Apparently google only counts people who sign up for google services (like the marketplace) and does not count phone upgrades.

      This is an interesting point. I have three Android phones, and have registered for Google services only once. They’d have undercounted me twice compared to the way sales are normally tallied up. Google may be low-balling its own growth numbers by making moving a SIM to a new handset so easy.

  23. I don’t see the network effects being all that strong. Any phone can connect with any other and they all let you read email from any source; unless someone develops a killer app that will only run under one OS, network effects of phone OSes seem pretty minor.

  24. @Richard Thompson:

    Many things are possible. We could have reached some (temporary) point of saturation in the Smartphone base. Customers could be holding out for better offers. It could be something else.

    But there is no pretending that the growth rate of both Android and iPhone haven’t slowed. (Apple should announce its activation rate during its earnings call Wednesday afternoon.)

    It could be something as simple as this:

    http://www.nytimes.com/2011/04/20/business/global/20yen.html
    http://www.bloomberg.com/news/2011-04-19/european-april-consumer-confidence-weakens-more-than-forecast.html

    These people think that 2010 was an anomaly with pent-up demand from a weakened 2009. Of course, they also think Windows Phone is a serious contender, so they’re smoking some good shit:

    http://www.dmwmedia.com/news/2011/03/29/report-global-smartphone-market-grow-50-year

    Nonetheless, they might actually have a correct thought (though who knows if their reasoning behind it is any good or not).

    IF everybody else’s market share is headed into the toilet, AND both Android and Google’s growth rates are slowing, THEN the overall smartphone market must be taking a breather.

    I actually think the proliferation of Android handsets could have a lot to do with that. Even though the cost of getting it wrong is small compared to buying the wrong car, a phone does become a big part of your life, and people are probably researching what’s available now, what the reviews are, what’s coming out in a few months, etc.

    As with a car, most people already have a phone, so there is no definite required timescale for the upgrade process, just whatever they’re comfortable with.

    It’s also possible for the same reason that the proliferation of Android handsets actually helps Apple sell more units (not more than if there were no Android, but more than if there just a few highly differentiated Android handsets). With the current chaos, someone could easily throw up their hands and say “I’ll just get an Apple! It might not have the latest flux capacitor upgrade, but at least I know millions are happy with it.”

  25. @William B Swift:

    > I don’t see the network effects being all that strong.

    Network effects on android (deliberately, IMHO) are not that strong. But Apple tries really hard to use network effects to its advantage. For example, you can share apps, as well as music, with up to 5 other users.

    1. >Network effects on android (deliberately, IMHO) are not that strong.

      You don’t think that consumer perception of the depth on the bench in the apps and handset market counts as a network pull? I do. I think Google has engineered for minimal product lockin but strong network effects. I know that’s a little difficult to grasp because we’re so used to those two traits traveling together.

      Then, too, even relatively naive consumers these days understand that tech markets tend to have a winner-take-all dynamic. That’s why market share is so unstable – changes have a positive feedback.

  26. @Winter

    First I think I saw a story about one model of android phone on Verizon outselling the iPhone 4 on verizon

    +tz “If you use a smartphone 80% as something else, having the 80% only in a less expensive or alternate device makes sense.”
    +And you are predicting the Android “gadgets” will be more expensive than the iOS or WP7 gadgets? Why?

    No,, they will be cheaper. Or the continuum will probably be lower quality and cheaper, so you might get a $50 Sansa or a $80 HD Flip type device.

    +tz “Win Phone 7 will die if they don’t have millions of devices by summer – and they aren’t going to be phones. They need a Zune and tablet that run the phone apps, subsidize it like the Xbox if needed so that they can kickstart their ecosystem. ”

    + At the end of the year an expected 250 million Android phones will have been sold. To counter that, MS will have to subsidize WP7 appliances in the multi billion dollar range. Make that tens of billions of dollars. I expect the stock holders to balk at the mere thought of that. Half the apps have something to do with phones/GPS/Internet access. The Zune does not come to mind as the best gadget for such apps. And who would build these tablets? Millions of WP7 tablets?

    I don’t know, but they did that with the Xbox for years. It wouldn’t be a zune, but something like one, more like an iPod touch, except running WP7. If they want WP7 to have a chance of going positive anytime in the next 3 years they will need to do such a subsidy. The stockholders aren’t going to like yet another expensive failure either. The android tablet hardware isn’t that different so they could have something out quickly for another billion or two.

  27. >The greater part of the reason I continue focusing on market share is that my primary interest is in how open the smartphone infrastructure will be when the war is over. Thus, who the winner is matters a lot; how much money gets made along the way and by whom, not so much.

    I agree that market share is the most important figure to watch, because handset profitability tells you little about how attractive a platform developers will find it. But you talk as if profitability is just about who gets their name on a charitable foundation.

    Look at the Modigliani–Miller theorem: allowing for doubts about how exactly the theorem translates to real life, it tells you that companies get to choose between making investor dividends and raising fresh investment. So highly profitable companies are in a much better position that weakly profitable companies to try to explore novelties or break barriers to entry into new markets. That is, it is something to keep an eye on when predicting who the race will go to.

    Of course, that doesn’t mean they will be successful. Look at the huge number of dud projects Microsoft have pioneered in the last ten years.

    1. >So highly profitable companies are in a much better position that weakly profitable companies to try to explore novelties or break barriers to entry into new markets. That is, it is something to keep an eye on when predicting who the race will go to.

      I agree. But profitability matters only in proportion to the degree that financial mass is actually necessary to “explore novelties or break barriers to entry into new markets”. I don’t think this is a market where financial mass is very important to innovation, and never have thought so. If it were, Microsoft and Nokia would have made a better showing by sheer weight.

      The truth is that in markets that are design-intensive (whether software or hardware) small nimble companies are at least as good at exploration as large ones, and arguably much better. This is why both the pharmaceutical and IC industries have shown a strong tendency to organize as networks of boutique design shops with contract links to manufacturing behemoths. In markets like this, knowing who looks good on the quarterlies is simply not very helpful in predicting innovation or future market share.

      Part of the genius of Google’s grand strategy is that the Android army in some ways collects economies of scale as though it were a single company with huge finacial mass even while the individual outfits in it can remain small and nimble. Case in point: NVidia and Qualcomm think it’s worthwhile to create Android SoCs to sell to the army even though most of its troopers probably wouldn’t present large enough demand to individually justify that investment. At the same time, the troopers are small enough to turn on a dime in a way that a really large company generally cannot.

  28. @winter

    > And who would build these tablets? Millions of WP7 tablets?

    Nobody. Microsoft has made it clear [http://goo.gl/c0IQX] that Windows tablets will run traditional Windows (i.e. Vista 7 modified for tablets), *not* Phone 7. They may, of course, change their mind once they realize that this continues to be a losing strategy, but there’s still a lot of internal culture in Redmond that knows PC-tablets were Bill Gates’ pet project so they will be hesitant to disassemble their Great Leader’s vision, even if he’s not at the helm anymore.

  29. @esr:

    You don’t think that consumer perception of the depth on the bench in the apps and handset market counts as a network pull?

    I absolutely think it is a huge, non-coercive pull, so gentle that many might not realize they are being pulled. I mean, the app network effects have a leveling out effect against Apple, and somebody might see the handset that catches his eye, but the average consumer isn’t going to choose Android because there is an app that is only available there, or because the handset is one of many Android ones available. I do see a push — people straining against the lock-in of Apple, or the withering ecosystems of the other non-Android vendors.

    I’ve written about google’s Android judo moves multiple times. It’s just like the Data Liberation Front — the fact that you can get data away from them very easily means that you don’t have to agonize over leaving your data with them.

    > I think Google has engineered for minimal product lockin but strong network effects.

    Possibly. I think google’s mostly about making the phone transparent to the actual network. To the extent they succeed with this, they leave the door open for non-locked-down Meego, for example. Not that Intel and the handset makers worried about google would know what to do with google’s munificence in this instance, but still…

  30. Another quarterly prediction article:

    http://www.thestreet.com/story/11085995/1/atts-subscriber-growth-nears-zero.html?cm_ven=GOOGLEN

    We’ve discussed prepaids before, but, like Android, they are really picking up steam. Carriers subsidize prepaid phones much less heavily than postpaid phones, but consumers are getting smarter and weaning themselves from the carrier contracts in record numbers.

    This is the sort of thing that Apple is up against, and this is their challenge.

  31. >We’ve discussed prepaids before, but, like Android, they are really picking up steam.

    I think the problem pre-paid faces is twofold. As noted, the component costs of these phones are not cheap, but they will be coming down considerably. But more harmful at this point is that almost all the carriers (save Virgin Mobile) are afraid of pre-paid because I think they see it far what it is, a fundamental change in the power and pricing of cellular service. A la carte scares the cell providers in a big way, and so for the time being prepaid smart phones are as bad and in some cases worse than the contracted ones. I was really hoping that T-Mobile would lead the way down the new path, but it looks like Virgin mobile is trying for that ($25 / month, pre paid, 400 minutes, unlimited tex / web). To be honest, I don’t think you’re going to see “cheaper” iPhones until such a plan is available on one of the bigger carriers and until those carriers are all using SIM devices (or alternatively if the iPhone just comes with both antennas). And then I personally envision a sort of merge of the iPod touch and the iPhone. You’ll buy it like you would buy an iPod touch, and if and when you want to use it on a cell network, you’ll prepay $25 for a month of web / text and some voice minutes, on whatever network you choose.

    Incidentally this is also where Google could have a huge breakthrough with google voice. One of the big issues with pre-paid is you don’t really own your number. If you let your service lapse, the cell company can re-assign that number and when you re-up you can be given an entirely new number. Imagine if instead, you have your portable android device, it’s hooked into your Google Voice account so you can get texts and such even over the web, and then when you activate your cell service, it activates your Google Voice number rather than whatever number the cell company provides. I’m positive there are some real technical roadblocks to that happening, but that to me seems like the killer feature of the pre-paid future.

  32. @tmoney:

    You realize that in the US Virgin is owned by Sprint, as is Boost, right? Sprint’s not waiting for anybody else — they’re busy competing against themselves:

    http://www.rcrwireless.com/article/20110406/CARRIERS/110409966/boost-mobiles-smartphone-further-clouds-sprint-nextels-prepaid-plans

    I wrote about that earlier in a couple of comments starting here.

    BTW, that phone I mentioned is the #3 selling prepaid at Amazon. Numbers one and two are only dumbphones, but they are on AT&T’s network, so afraid or not, they are entering the fray.

  33. @tmoney:

    I do like your suggestion re google voice, though. Seems like it wouldn’t be too hard for an android phone to be told to prefix all dials with a number that would route through google voice, then google voice could discern the calling phone, map the temporary prepaid number to a more permanent google number, and push the call out. This would make caller ID work right — nobody you call would see your disposable phone’s number if you didn’t want them to.

  34. @ Patrick

    Actually, I hadn’t. I thought they were still an MVNO. Still, I somehow can’t see Steve Jobs on stage offering the newest iDevice and describing the great cellular service you can optionally get from Boost Mobile (“Where you at?”). It’s a shame, but we really need to see those types of plans under the AT&T or Verizon names before I think it will really take off. Too many people are on under those two umbrellas and I think too unwilling to risk their service on an off brand provider, even if they are wholly owned by the bigger company.

  35. @tmoney:

    Disagree completely. Maybe not about Jobs — Apple will be late to the prepaid game if they play at all, but TracFone added 1.1 million new prepaid subscribers in Q4 2010, and:


    Sprint ended 2010 with 33.1 million total retail postpaid subscribers and 12.3 million retail prepaid subscribers. Sprint has continued to leverage its prepaid business to compensate for postpaid losses–Sprint ended 2008 with 36.7 million retail postpaid subscribers and 3.6 million retail prepaid subscribers.

    Crunch those numbers: Sprint’s overall retail is up 12% over the two year period, which isn’t fantastic, but would have been abysmal without its prepaid being up 241% over the same period, because its postpaid is down 10%. And right now, 27% of all Sprint’s retail customers are prepaid.

    The previous article I linked to showed things trending that way for AT&T and Verizon as well. Apple ignores prepaid at its peril — that’s probably part of why the rumor that the iPhone 5 will be a lower-cost refresh.

  36. Can I haz Cheap Chinese Junk Android for Tracfone? Please?

    I pay $84/year (cost of 4 90-day airtime cards) with the dumbphone. Adding some data into the mix would totally rock…

  37. @JB:

    You’re getting ripped off. I only pay $64.95 (4 * $15 * 1.0825 local sales tax) for 360 days for Virgin Mobile. They act like you have to add $20 every 90 days, but they don’t enforce it — $15 works fine. (OTOH, maybe I’m grandfathered in or something).

  38. “TracFone added 1.1 million new prepaid subscribers in Q4 2010.”

    Maybe not….the thing about Tracfone is you can change your phone number just by buying a new phone for ten bucks. I’ll bet a lot of people do that, or lose or break their old phones and go the new route because waiting for a human on their 800 number is too painful, or they don’t think they can handle the instructions to keep their old number, or, whatever…they all get counted as new subscribers.

    That said…I have to say that I’ve been using a Tracfone for years. Coverage in the New York City area is just fine. I would suggest to JB, that, if his phone is fairly new, and he hasn’t already, get one of those double minutes cards and put it in…you’ll cut your airtime costs in half.

    We recently gave a family member a new Tracfone. I forgot what model it was, but it is a flip-style phone with a nice display, a web browser and it came with double minutes, all for fifteen dollars (US).

    I’ve mentioned this before, but this is an opportunity for NoWin. Sell Tracfone a Win7 phone made by Nokia cheap enough, and you’ll capture market share from the bottom up. It’s also a backdoor entry to the Asian market. (They are GSM phones sold as a loss leader. Asian immigrants have been driving Tracfone crazy by buying them in bulk and shipping them to associates across the Pacific, where the SIM cards are swapped out and the phones are sold at a profit.)

    1. >comments on these two stories that were all over the web today?

      The first one gives numbers that are pretty meaningless in isolation. Without knowing the product mix of that 18.1M units we can’t even guess how many of those are new customers, how many are upgrades, and so forth. I will note that the analyst’s 18.4M for the whole quarter equals about a month of Android activations at the current rate.

      The second one was quite rightly called out by Patrick Maupin earlier as an example of one of the sorts of fanboy delusional thinking I didn’t cover in my previous post – anybody who thinks iPods still count. Low-margin business, and probably a negative share impact on iPhone and iPad sales.

  39. Excerpts from Wall Street Journal today:

    Amid Windows Shift, Eyes on Nokia’s Basic Phones

    “The company has cautioned that the transition [from Symbian to WP7] will take time, and analysts and investors expect Nokia to suffer steep market share losses.

    “As Nokia’s market share in high-end smartphones continues to slide, investors and analysts are taking a closer look at its basic mobile-phone business, anchored mostly in emerging markets. While smartphones get most of the attention, because they are driving industry sales, most people around the world still use basic phones…

    “Nokia sales [of dumbphones] in the fourth quarter reached $5.86 billion, or 48% of its revenue, and even more of its profits… That low-end business is Nokia’s last line of defense for sales, profits and market share, and could help the company better stay afloat as its higher-end smartphone business transitions to Windows.

    “‘They do have to rely on their [basic] phone business, not only for the next year, but for the next several years while they ramp up their smartphone business,’ says [an analyst].

    “Still Nokia can’t survive on its low-end business alone, analysts say….Nokia’s low-end business is important, because currently it is the company’s ‘bread and butter, but [an analyst] cautions that the business faces tremendous challenges from all sides.”

    This looks to me like a fairly risky strategy. Time is precisely what they don’t have, and Android is clearly a threat to their low-end phone business, especially in the U.S., Europe, and Japan. The article doesn’t call the Microsoft deal a Hail-Mary-pass, but given the discussion on this blog, one wonders what the analysts say in private when not being quoted. :-)

    Nokia may actually be the biggest victim of Android. Apple is managing as well as can be expected, and as a higher-end, higher-margin business they can survive for a while. Windows Phone was doomed anyway. But Nokia, which was never really threatened by WP or the iPhone, suddenly had an Android appear out of nowhere and position itself to take on not just the high-end, but the low-end that is their “bread and butter.”

    http://online.wsj.com/article/SB10001424052748704740204576272912937511914.html?mod=WSJ_Tech_LEFTTopNews

  40. @Patrick Maupin

    The Netherlands nor Indonesia are going to save RIM.

    RIM’s business model is more like Apple than Android. And they are not going to win from Apple.

  41. @Cathy
    We know there are 390k Android activations a day, which makes for around 35M phones for a quarter. That is roughly 2:1 Android activations per iPhone sold.

  42. Wow. So market share is what counts. Until it comes out that iOS devices (not just phones) have a larger base by a huge amount over android devices. People have been pointing out the undercounting for quite a while, since about the iPad came out. Margin doesn’t count, until iOS devices have a larger share. Then it’s all attributable to “low-margin” iPods! Eric, I think they were counting iOS devices, not iPods. Don’t have an axe to grind here. Think the competition is good for both firms, but I come by here every few weeks to hear about Apple’s imminent demise and Googles ingenious strategy. Firm values tell a completely different story however.

    1. >Eric, I think they were counting iOS devices, not iPods

      Remember what I’m interested in – who will own the platforms of the future. iPods are irrelevant to that. So are iOS devices that happen to be desktops and laptops; that’s today’s story, not tomorrow’s. What’s relevant to that question is smartphones and, maybe, tablets. I say “maybe” because it’s not clear that having a big marketshare in tablets will amount to hill of beans for any player who doesn’t also own most of the much, much larger smartphone market.

      I expect that as Apple’s problems mount in the smartphone market there will be more attempts to confuse this issue by eliding different categories. I don’t intend to let that take my eye off the ball. You shouldn’t either.

  43. @Ian Fairchild
    “Eric, I think they were counting iOS devices, not iPods.”

    It depends on what you are interested in. I for one am looking towards the way my future computing needs are going to be served. And the best possible outcome would be for me to have a small, powerful portable device which allows me to carry around my data, applications, and settings securely and allows me to access networks and “compute” everywhere.

    That does sound a lot like a next generation smartphone with optional external screen and keyboard connections.

    To be useful to me I *must* be able to install and develop everything I want, including command shells. More like a modern Linux system. Which points towards FLOSS as the platform.

    So the evolution of both phones and tablets is important for me. I have many problems working on OSX and even more on Windows because of their closed nature and culture of walled gardens. I could not care less about music players and other one-trick gadgets. So if there would be 10 iPods for every phone, I still would not be interested as these are not going to fulfill my computing needs.

    I have a feeling that Eric and many other commenters share at least some of that (des)interest.

  44. Patrick: I didn’t know that, thanks…I don’t know of a way to buy cheaper than $20 ($21 w/tax) cards for TF, since 60 min/90 days are the lowest denomination.

    LS: Got that! Minutes aren’t a problem, I’ve got plenty – the effective “maintenance fee” is the “90 day renewals.”

  45. Winter, an iPod Touch may not fulfill your computing needs, but it’s hardly a “one-trick gadget.” It can download apps from the App Store, which makes it an important part of the iOS ecosystem, and a contributor to some network effects that work in Apple’s favor.

  46. @Winter> We know there are 390k Android activations a day

    And just how to ‘we’ know that? Google only admitted to 350K activations/day on April 14.

    I’ll just leave this here:

    JavaScript PDP11 emulator running Unix V6
    http://pdp11.aiju.de/

    Tips to anyone else trying this out:
    # stty -lcase
    # chdir usr
    # chdir games
    # wump

    (Yes, I know it’s off-topic.)

  47. @esr> I will note that the analyst’s 18.4M for the whole quarter equals about a month of Android activations at the current rate.

    Say what? 18,400,000/30 = 613k/day.

    Google only claims 350k/day (on 4/14, during an earnings call).

    Where are the other 275k coming from, Eric?

    1. >Where are the other 275k coming from, Eric?

      I was using Paul Maupin’s activations number. I wasn’t aware of the one from the 4/14 earnings call.

  48. #iPhones sold during Q1 = 18,400,000
    / 90 = 204,444.444

    So 204.4k/day.

    Google said on 4/14 that they’re at 350k activations/day. Using Google’s claims of activations/day, Android’s rate of growth appears to be slowing.

    Again, from June 2010 to December 2010, the number of Android activations per day went from 160k to 300k. In the period from December 2010 to April 2011, the number of activations per day went from 300k to 350k.

    160k -> 300k = 1.875x
    300k -> 350k = 1.167x

    Apple went from selling 93k iPhones per day in the spring 2010 quarter to selling 180k iPhones per day in the fall 2010 quarter, to selling 204k iPhones/day in the latest quarter, so their rate of growth is slowing, too.

    93k -> 180k = 1.935x
    180k -> 204.4k = 1.135x

    Apple claims 108M iPhones sold

    Android claims (these are activations)
    Oct08: Zero
    May09: 30,000/day
    Feb10: 60,000/day
    May10: 100,000/day
    Jun10: 160,000/day
    Aug10: 200,000/day
    Dec 10: 300,000/day
    Feb 11: 350,000/day

    If I take the activation rate for these days, multiply it by the entire period (so way over-estimating volume), I come up with 107,280,000 Android devices sold.

  49. @PapayaSF:
    I know the iPod touch is a phoneless iPhone. So it or other phoneless devices could qualify as significant. Still, I don’t think the iTouches are outselling phones yet.

    However, I do expect a barrage of WiFi only devices, with ubiquous VoIP and free WiFi access. Simply because the demand for bandwidth outstrips what telcos can deliver over the air.

    Part of the anti piracy politics seems targetted at outlawing free broadband internet access to protect the telcos. You see it happen in Canada and the UK.

  50. @Richard Thompson

    The estimated 390k activations was from a comment from Patrick Maupin.

    This is a nice set of Android activation data. Reworked into days from 15 May 2009

    Date Days #Activ log(Activ)/log(2)
    05-09 0 30000 14.87
    02-10 275 60000 15.87
    05-10 365 100000 16.61
    06-10 396 160000 17.29
    08-10 457 200000 17.61
    12-10 579 300000 18.19
    02-11 640 350000 18.42

    If you look at the log plot, you will observe a jump around June 2010. To get the correct growth rate minus this one time bump, I have added 0.5 to the logarithms before June 2010 and subtracted 0.5 from August 2010. Then you get a correlation line of:

    A log vs days correlation gives
    log(#Activations/day)/log2 = y = 0.003995 * x + 15.43 (R^2 = 0.96)

    This is a growth rate of 2^(365 * 0.003995) = 2.74760573 /year
    This growth rate is somewhat lower than I anticipated (my estimate was around 3.8). The sudden doubling of the activation rate around June 2010 messes up most estimations.

    Just adding the days * mean activations (geometric) comes down to somewhat over 80 million Android activations up to last February. The expectations for the end of the year are then 850,000 activations per day and around 180 million activations between February and new year’s eve.

  51. @esr> I wasn’t aware of the one from the 4/14 earnings call.

    Even though it was pointed out above? http://esr.ibiblio.org/?p=3137#comment-304512

    Here’s the other wonky thing about the 350k/day Android activation number:

    Eric Schmidt stood on stage at MWC on Feb 11 and said 350k activations / day.
    source: http://www.engadget.com/2011/02/15/live-from-eric-schmidts-mwc-2011-keynote/?sort=newest&refresh=60

    Then on April 14, Jeffrey Huber says “over 350k/day” on the Google earnings call.
    http://seekingalpha.com/article/263665-google-s-ceo-discusses-q1-2011-results-earnings-call-transcript

    So… are we to believe no real growth in the activation rate for two months?

  52. @Richard Thompson
    “So… are we to believe no real growth in the activation rate for two months?”

    In 8/9 weeks the activation rate should have grown to around 410,000/day.

    Which is actually over 350,000. More likely, they did not have the newest exact numbers at hand. And I expect quite some variation due to holidays and “events” all over the world. How to account for, e.g., Chinese New Year (Feb 3, 2011), or the aftermath of the Japanese earthquake?

  53. Just some nice pictures:

    8 Android Tablets for 2011 to be Excited About
    http://www.techdrivein.com/2011/04/8-android-tablets-for-2011-to-be.html

    Best Android Tablets for 2011
    Until recently, Samsung Galaxy Tab was the only Tablet offering that had Android OS. But that is all going to change now. Android 3.0 Honeycomb OS is optimized for tablets and almost all major manufacturers of smartphones are planning for Tablet device based on this latest Android release. Here is a quick listing of best Android tablets for 2011(in no particular order). Also read: Top 5 Android Tablets Showcased During Mobile World Congress 2011.

  54. >>Where are the other 275k coming from, Eric?

    > I was using Paul[sic] Maupin’s activations number. I wasn’t aware of the one from the 4/14 earnings call.

    Mine wasn’t an activation number. It was taken from the report that the blog post was about, simply by adding up Android phones sold in March in US, Europe and Japan (excluding China, Taiwan, Korea, and other areas not covered by the report) and dividing by 31 days in March. Unless I messed up my math, that came to more than 390K units/day.

    Which means that the report is off, or that activations are becoming a less useful proxy for Android usage, or that we simply need to somehow figure out what percentage of phones are being activated as more dumbphone users convert.

    Remember that “activated” means “using google services.”

  55. More likely, they did not have the newest exact numbers at hand.

    Nitpicking over the numbers stated in speeches by some busy Google execs is silly. I’m sure Huber either didn’t bother asking for the exact numbers or his staff was unable to get them for him by the time he gave his speech. These guys don’t just give speeches, they have a business to run.

  56. Actually, if you dig a tiny bit deeper towards the bottom of the page, the retail adds were slowed considerably, and AT&T is doing a bunch more with resellers (probably pre-paid) to make up for it.

    Holy crap, AT&T must have sold a lot of used iPhones at $0 – $50! (and/or activated a bunch of used ones people sold on ebay)

    If I read their PR correctly, they activated 3.6 million iPhones, but sold only 5.5 million smartphones. Unfortunately, I have a lot to do today, so won’t be digging in any further, but obviously AT&T is doing all it can to staunch the loss to Verizon — 23% of those 3.6 million activations were new customers.

    The data in the full report will merit several hours of analysis; unfortunately I cannot do that this week.

  57. @Patrick Maupin:

    Good place to start looking:

    * Best-ever first-quarter smartphone sales of more than 5.5 million
    * iPhone activations increased nearly 1 million year over year to 3.6 million, with 23 percent of subscribers new to AT&T; iPhone subscriber churn unchanged year over year

    With iPhone 4’s release last year in 1Q10, a large percentage of those 3.6 million activations will be upgrades.

    From those numbers, it appears that Android might be outselling iPhone on AT&T.

  58. > With iPhone 4?s release last year in 1Q10, a large percentage of those 3.6 million activations will be upgrades.

    I was sort of thinking the opposite — all the upgrades happened last year, because AT&T was giving a really cheap upgrade because they knew iPhone v was coming. I think most of these are selling the used returned units, which you could get for $0, $19, or $49, depending on capacity and cosmetics.

    > From those numbers, it appears that Android might be outselling iPhone on AT&T.

    Possibly, but I calculated in a comment on an earlier blog post that that was already happening last year. I think that the really cheap iPhones may have reversed it, or at least evened it out a bit.

  59. > More likely, they did not have the newest exact numbers at hand.

    Google doesn’t have computers? Andy Rubin doesn’t have an app that tracks this stat?

  60. AT&T reported it added a net 62,000 post-paid wireless customers in the first quarter, compared to the 512,000 subscribers it added a year ago when it sold the iPhone exclusively. An 88% drop, year over year.

    Subtracting the 58,000 tablet contract customers added in the quarter, AT&T narrowly avoided — by 4,000 users — a net loss of post-paid phone subscribers for the quarter.

  61. Morgan Greywolf Says:
    > From those numbers, it appears that Android might be outselling iPhone on AT&T.

    This would be really surprising given the paltry Android options offered by AT&T (until the last 4 weeks or so).

  62. I was sort of thinking the opposite — all the upgrades happened last year, because AT&T was giving a really cheap upgrade because they knew iPhone v was coming.

    Could be, but people typically sign up for 2-year contracts. That usually means new phones will affect sales for about 2 years rather than just 1 year, but maybe I’m just crazy. You’re right that a lot of those will also be $50 iPhone 3GSs.

  63. @Michael Hipp:

    > This (Android outselling Apple on AT&T) would be really surprising given the paltry Android options offered by AT&T (until the last 4 weeks or so).

    I don’t think this happened this quarter. Nonetheless, I think it came closer than you would expect as long ago as Q3 last year — I did some math on that here:

    http://esr.ibiblio.org/?p=2813#comment-289218

    I also think Android outselling Apple on AT&T may have happened in Q4 (after they converted everybody to iPhone 4 for really cheap, before they ramped up sales of used iPhone 3), but I don’t really remember, and can’t find the blog comment I would have made on that.

    @Morgan:

    > Could be, but people typically sign up for 2-year contracts. That usually means new phones will affect sales for about 2 years rather than just 1 year, but maybe I’m just crazy. You’re right that a lot of those will also be $50 iPhone 3GSs.

    Except that when AT&T saw this coming, last quarter and quarter before last, they were letting people still under contract buy an iPhone 4 for cheap (sometimes $19) just with a 2 year contract extension. They were even sending out emails and texts to their customers offering the promotion. I think they collected a lot of used iPhone 3s during that process.

    I took a quick look to try to tease out the numbers from last quarter to this quarter, but unfortunately, AT&T changed how they report these things. This could be completely innocent, but I would not be surprised at a bit of deliberate misdirection here.

    They used to report “integrated devices” which, by their description, I assumed were smartphones, but which it now appears included feature phones. Now, they report actual smartphones. Makes it hard to compare but here’s the apple and the orange:

    – At the end of Q4 2010, 61% of 68.0 M postpaid subscribers were using integrated devices
    – At the end of Q1 2011, 46% of 68.1 M postpaid subscribers were using smartphones

  64. @Michael Hipp:

    Found it:

    http://esr.ibiblio.org/?p=2898#comment-294235

    I don’t think that Android ever outsold Apple at AT&T, but the trend was clear: in Q3, 65% of integrated devices sold were iPhones, and in Q4, 55% of integrated devices sold were iPhones. Now they’ve switched over to “smartphones” as a category, and dumped a boat load of cheap used iPhones into the marketplace, so last quarter 65% of smartphones sold were iPhones.

    In one very real sense, “smartphones” today as a sales category might almost be fungible with the “integrated device” of a quarter ago. Both categories for their time include people who pay a bit extra over the basic dumbphone (“integrated devices” include a virtual or real QWERTY keyboard, and there probably aren’t that many of those that aren’t smartphones around any more), so at the margin we could treat these categories on these timeframes the same for sales numbers (but not installed base).

    So, best case, iPhone sales went 65% – 55% – 65% at AT&T, within this slightly shifting category. (To the extent AT&T still sells integrated devices that aren’t smartphones, you would have to adjust that last number down to make it relevant.)

    As you point out, more and better Android devices are coming, and the stock of used iPhones must be about finished, so I think this is a very temporary bump. Absent some other change (like faster integration of T-Mobile) I expect minimal iPhone sales at AT&T next quarter, and even the quarter after that if the iPhone 5 is really such a minimal refresh.

  65. Wouldn’t AT&T’s “integrated devices” also include iPads and Kindles? Not sure what US sales of the iPad were, but if they were say 6 million units over the past three quarters (out of what, 20 million or something?) that would probably decrease the “iPhone” share in Q4 (where iPad sales were highest, and the majority were on 3G contracts).

    Just a guess.

  66. > Wouldn’t AT&T’s “integrated devices” also include iPads and Kindles?

    No. They called (and still call, I think) those “connected devices.”

    From http://www.att.com/Investor/Financial/Earning_Info/docs/4Q_10_IB_FINAL.pdf

    Integrated devices are handsets with QWERTY or virtual keyboards in addition to voice functionality
    and are a key driver of wireless data usage.

    and:

    Continued expansion in new wireless growth areas; connected devices
    up a record 1.5 million; iPad- and Android-based tablets up 442,000.

    I thought the definition of “integrated devices” matched “smartphones”, but their latest quarterly has a definition of smartphone:

    Smartphones are voice and data devices with an advanced operating system to better manage data and Internet access.

    If you think about it, their definition of smartphone could theoretically exclude anything that RIM or Nokia currently makes…

  67. They act like you have to add $20 every 90 days, but they don’t enforce it — $15 works fine. (OTOH, maybe I’m grandfathered in or something).

    The $15 deal is if you’ve registered your credit card with them to auto top-up, which is what I do. That’s kind of the Holy Grail of the service business: When you don’t have to do anything for them to get paid, inertia means they’ll get more money than if you have to do something every few months. Your apathy works in favor of them getting paid this way, so in the long run they make more money.

  68. > The $15 deal is if you’ve registered your credit card with them to auto top-up, which is what I do.

    Ah, that makes sense.

    > Your apathy works in favor of them getting paid this way, so in the long run they make more money.

    Yeah, but the model also works in favor of the phone just working on the odd occasion I want to use it. Win-win.

  69. If you think about it, their definition of smartphone could theoretically exclude anything that RIM or Nokia currently makes…

    “Smartphone”, as of summer 2007, is a radial category with the iPhone in the exact center; inasmuch as a device becomes more iPhone-like it becomes more of a smartphone.

    1. >“Smartphone”, as of summer 2007, is a radial category with the iPhone in the exact center; inasmuch as a device becomes more iPhone-like it becomes more of a smartphone.

      That’s not true for me. The center of my smartphone category is the Nexus One. The G-1 was wonderful at the time, but once I saw the N1 it seemed unpolished and awkward, no more than a worthy prototype. Every phone I’ve seen since has seemed like a pretty small incremental step forward by comparison.

      And I actually evaluate the iPhone the same way, as a worthy prototype that failed to reach full potential. Opacity and lockdown make the iPhone maddeningly un-smart for my purposes.

  70. I wonder where Nokia’s upcoming, supposedly last-ever Maemo/MeeGo phone N950 will fall in all this. By the time it launches, it should be even clearer that they’ve already lost to Android. Might they try to do something with the N950 and possible successors? I guess they’d have to fire Elop first and suffer yet another change of direction.

    Meanwhile, Nokia’s contractors are preparing to scale down their businesses. Ahead of the big announcement that everyone is expecting from Nokia, Digia has announced that it will lay off 63 people and Ixonos has started the process to lay off up to one hundred. Both companies are based in Finland and have done Symbian and Maemo/MeeGo -based development for Nokia.

  71. More from the AT&T Q1 earnings report:

    “Excluding the impacts of the Alltel and Centennial integration migrations, postpaid net adds were approximately 165,000.”
    “Postpaid churn was 1.18 percent…”
    “More than 5.5 million smartphones were sold in the first quarter…”
    “iPhone activations increased … to 3.6 million, with 23 percent of subscribers new to AT&T”
    “Approximately 65 percent of postpaid sales were smartphones.”
    “At the end of the quarter, 46.2 percent of AT&T’s 68.1 million postpaid subscribers had smartphones…”

    I’m concentrating on postpaid here. AFAIK, AT&T never sold postpaid smartphones until this week, so it’s probably safe to assume they’re all postpaid.

    So, 968K new postpaid AT&T subscribers (165K net adds + 1.18% churn), and 828K of those (23% of 3.6M) bought iPhones, with the other 140K buying dumbphones or Android. Those cheap iPhones must have been really enticing.

    AT&T sold 8.5 M postpaid handsets in the quarter (5.5 M smartphones / 65% smartphones), so for existing customers, they sold:

    2.8 M iPhones (3.6 M * 77%) — 37%
    1.76 M – 1.9 M Android phones — 23% – 25% (depending on mix of 140K new customers)
    2.86 M – 3.0 M dumbphones — 38% – 39% (depending on mix of 140K new customers)

    So, absent any additional change to entice existing customers to use smartphones (such as reduction in the price of data plans or reduction in the price of smartphone handsets), over two-thirds of AT&T’s dumbphone customers are going to continue using dumbphones. Currently, the number of dumbphone customers must be 54%, with last quarter’s dumbphone replacement sales at 38-39%.

    I think this is a huge advantage for whatever ecosystem can get the price of a smartphone down to where it’s viable even without a data plan. A lot of phone users probably wouldn’t mind a combination dumbphone / wifi terminal access device. AT&T is not going to drop their data rates too quickly — they would be leaving money on the table from existing data customers, without picking up enough extra revenue to make it up on the revenue side, never mind after they lay out more investment for bandwidth. And Apple is not going to drop the $600 iPhone price too far, and AT&T is not going to subsidize smartphones for non-data customers. So that’s advantage Android, for whatever handset comes out cheaply enough for the dumbphone user to buy off the shelf and move his old SIM card over to.

  72. BTW, using AT&T’s definition of smartphone, I assigned their non-Apple smartphone sales to Android, and lumped RIM, etc. into the dumbphone category.

  73. @Fackbook is evil:

    AT&T reported it added a net 62,000 post-paid wireless customers in the first quarter, compared to the 512,000 subscribers it added a year ago when it sold the iPhone exclusively. An 88% drop, year over year.

    Subtracting the 58,000 tablet contract customers added in the quarter, AT&T narrowly avoided — by 4,000 users — a net loss of post-paid phone subscribers for the quarter.

    Nope. 62K post-paids was after including Alltel and Centennial (which were losing customers). Preexisting business gained 165K, and tablet contract customers are a completely different category.

  74. WSJ is claiming that Apple’s quarterly profits jumped 95% as iPhone sales “soar” to “record sales”, accompanied by “strong sales” of the iPad.

    iPod sales were down 17%, which is a pretty telling indicator that the consumer is moving on to integrating phone/data with music and is less likely to be looking for a standalone music player.

    Apparently gross margins are slowly declining, in part because “analysts have said the iPad isn’t as profitable as other products.”

    http://online.wsj.com/article/SB10001424052748704570704576275350324616010.html?mod=WSJ_hp_LEFTTopStories

    1. >WSJ is claiming that Apple’s quarterly profits jumped 95% as iPhone sales “soar” to “record sales”, accompanied by “strong sales” of the iPad.

      Wow. That article reads like it was written by an Apple PR flack.

  75. Effusive language or no, 95% y/y profit growth is pretty strong, especially given that they didn’t sell all that many iPads (they claim to be supply-constrained, a claim which is probably confirmed by the still-going long lines)

  76. Google showed a 28% revenue growth and a profit reduction for the same period. HTC profits nearly tripled, meanwhile, so clearly phone hardware is a profitable game right now.

  77. > Wow. That article reads like it was written by an Apple PR flack.

    Not unusual, most ‘early’ stories will essentially re-word the Apple PR. Later analysis will dig a bit deeper.
    Quoting Apple’s own PR of it’s recent results:

    Apple sold 3.76 million Macs during the quarter, a 28 percent unit increase over the year-ago quarter. The Company sold 18.65 million iPhones in the quarter, representing 113 percent unit growth over the year-ago quarter. Apple sold 9.02 million iPods during the quarter, representing a 17 percent unit decline from the year-ago quarter. The Company also sold 4.69 million iPads during the quarter.

    That’s 207k iPhones/day, and another 100k iPods/day. On the conf call, Apple admits that over 50% of these are iPod Touch devices. So iOS activations are easily above 260k/day.

  78. Terms like “soaring” and “record sales” are quite appropriate when you’re talking about a 95% quarterly revenue growth year-over-year. That’s awfully good for any company, but for a vendor of Apple’s scale, it’s almost unheard of.

    BTW, contrary to Eric’s assertion that the Verizon iPhone launch fell flat, Dan Mead said: “in just our first two hours, we had already sold more phones than any first day launch in our history.” http://www.appleinsider.com/articles/11/02/04/preorders_of_apples_iphone_4_break_verizon_sales_record_in_2_hours.html

    1. >Dan Mead said: “in just our first two hours, we had already sold more phones than any first day launch in our history.”

      I’ve heard this bit of PR-speak before. I also know what the evidence on the ground indicated. I’ll continue to believe the evidence, and not the PR-speak – especially since Apple evaded actually giving a Verizon iPhone sales number in its earnings call.

  79. @Bennett/Some Guy:

    Of course there is 95% year over year growth for Apple. How many iPhone 3GSs were they selling in 1Q2010 while the public was eagerly awaiting iPhone 4?

    You guys are outright silly.

  80. You’ve misread the reports about the Verizon iPhone (please either stop referring to it as the iPhone V or else stop referring to the upcoming iPhone 5 as the iPhone V, as I’ve seen commenters who are understandably confusing the two). Nowhere was it stated that it’s being outsold by the HTC Thunderbolt on Verizon — what was said was that it’s being outsold by the HTC phone _in Verizon stores_. Store sales presumably only account for a small % of total CDMA iPhone sales on Verizon. It’s dangerous to assume of course but not only does the aforementioned Kantar report support this, but Verizon themselves have stated that the iPhone broke first day sales records within a two hour period from 3am to 5am (they had to close pre-orders after that). This is your definition of a product launch with “lackluster sales”? Ok, it’s only first day figures, but if they can do that in 2 hours in the middle of the night it gives at least some indication of the demand on Verizon for the iPhone.

    I still don’t understand your reasoning in discounting iPod Touch devices when calculating marketshare Eric, particularly given your insistence that what you’re concerned about is the platform. iPod Touch devices are in and of themselves fully-fledged members of the iOS platform, of the “walled garden”. On top of that, they will presumably have at least some halo effect on later mobile/cellphone purchases, especially given that if you pick an iPhone you can freely migrate all your purchased apps onto the new phone.

    “Wow. That article reads like it was written by an Apple PR flack.” How so? Can you cite specific examples? It seems pretty detached and well balanced to me, even to the point of remarking that Apple’s revenue was lower than “some analysts'” estimates, instead of stating that it was in fact well above Apple’s own guiding estimates.

    As you can no doubt guess from the points I’m focusing on above, on the iOS vs. Android debate I come down on the side of iOS to the extent that I personally prefer it (though am perfectly happy to see Android do well as a viable competitor). You’ve always been open about which is your favoured platform, but it’s important to understand your own bias when analysing the state of affairs, so that you can better guard against it unduly influencing your perspective. FWIW, your “Smartphone Wars” blog posts, while always interesting to read, do sometimes come across like you could do well bearing this in mind.

    1. >(please either stop referring to it as the iPhone V or else stop referring to the upcoming iPhone 5 as the iPhone V, as I’ve seen commenters who are understandably confusing the two).

      My terminology is conistent. Perhaps your reading is not.

      >I still don’t understand your reasoning in discounting iPod Touch devices when calculating marketshare

      They do nothing to help Apple’s smartphone share rise – in fact they hinder it, as someone who goes the iPod route is likely to be be substituting it fot a iPhone that can do everything it does but cost a lot more. And PMPs in general are a dying category; smartphones are eating them.

  81. Morgan, if the expected launch of the next generation iPhone was slowing sales in Q1 2010, you’d expect a similar effect to also slow sales in Q1 2011. The rumors about an iPhone 5 delay didn’t show up until after Q1.

  82. but Verizon themselves have stated that the iPhone broke first day sales records within a two hour period from 3am to 5am (they had to close pre-orders after that). This is your definition of a product launch with “lackluster sales”?

    But Verizon is very careful in how they state this. It didn’t break other iPhone first day sales records. Rather, it broke some Droid? Droid X? Razr? sales records.

    Apple and AT&T always told us how many iPhones sold at launch before — lately, millions. This time? Crickets. And a comparison against some unknown number. The best I can find is some estimate that Droid X sold 300K at launch, but they sold out on the first day so you don’t know how many they could have sold.

    The thing is, the Droid is one of a whole army of Androids, but we’ve been hearing for over a year about the pent-up demand for the iPhone on Verizon. If it couldn’t beat any single phone that Verizon ever shipped in the past, especially when it was known well in advance it was coming, then that would have truly been game over, Apple.

    We may know a bit more tomorrow, but if Verizon plays their normal game, we’ll have to extrapolate. It’s their own damn fault if everybody extrapolates low…

  83. >iPods are irrelevant to that. So are iOS devices that happen to be desktops and laptops; that’s today’s story, not tomorrow’s. What’s relevant to
    >that question is smartphones and, maybe, tablets. I say “maybe” because it’s not clear that having a big marketshare in tablets will amount to
    >hill of beans for any player who doesn’t also own most of the much, much larger smartphone market.

    This seems horribly short sighted. What is a modern smart phone but an ultra portable computer with voice communication? Speaking of the future only in terms of “smart phones” seems to me like being in 2000 and speaking of the future only in terms of full sized laptops. But let’s say that you’re right and the future is defined entirely in smart phones and tablets. Isn’t an iPod Touch simply an iPad Nano? What is the fundamental difference that would make an iOS based iPod or other non smart phone device not matter? Yes, cellular data is likely a major part of the future. But a device can have cellular data access without being a smart phone and with technologies like Google Voice and Skype, voice communication is hardly limited to cell phones.

    >I think this is a huge advantage for whatever ecosystem can get the price of a smartphone down to where it’s viable even without a data plan.

    I think they already are viable without one. I don’t have one for my iPhone. The problem is, the cell providers won’t let you buy a smart phone without buying a data plan (full price pre paid excluded). Some people just aren’t willing to spend $50-70 / line for their cell phones.

    >Of course there is 95% year over year growth for Apple. How many iPhone 3GSs were they selling in 1Q2010 while the public was eagerly
    >awaiting iPhone 4?

    I imagine relatively the same as were being sold in 1Q2011 when the public was eagerly awaiting 1Q2011 while the public was eagerly awaiting iPhone 5. And again in 1Q2009 while the public was eagerly awaiting iPhone 3GS.

  84. This seems horribly short sighted. What is a modern smart phone but an ultra portable computer with voice communication?

    Ummm, an always connected ultra portable computer with voice communication? (For those who care.)

    Seriously, though, the form factor makes a huge difference. There are actually some 7″ tablets with voice functionality. It is my understanding that they are not much fun to use as a phone, but (like the B&N Nook) are much better for reading and media consumption than a phone form factor.

  85. And, BTW, I said this earlier (as did Eric) but the standalone PMP category seems to be dying. A 17% drop in sales for the company that owns the market is huge, and is not a one-time event. Expect this to accelerate.

  86. “The problem is, the cell providers won’t let you buy a smart phone without buying a data plan (full price pre paid excluded).”

    I don’t understand your point here. If you want to buy a smart phone without buying a data plan, obviously you would go for a full price pre-paid. That’s precisely the person whom full price pre-paid is for!

  87. Here is a really good estimate of Apple’s performance from last week:

    http://blogs.barrons.com/techtraderdaily/2011/04/15/apple-dont-sweat-large-numbers-says-jpmorgan/

    The reason I point this out is that he has numbers for iPhone V of “2 million or more” for the quarter, and since his other estimates were pretty darn good, I have to believe that he’s got a better handle on the actual situation than most.

    But you have to remember, when we started pooh-poohing the impact of iPhone V, it’s because we were seeing a lot of articles like this:

    http://www.pcworld.com/article/218034/verizon_iphone_sales_could_hit_25_million_this_year_analyst.html

  88. @Cathy:

    I don’t understand your point here. If you want to buy a smart phone without buying a data plan, obviously you would go for a full price pre-paid. That’s precisely the person whom full price pre-paid is for!

    We’re not looking at it from the point of the carrier; we’re looking at it from the point of ourselves as customers. We’re not broke, but we’re cheap (hmm, maybe there’s a correlation there). Anyway…

    Imagine a small WiFi media player / browser — an iPod, if you will.

    Now imagine a dumbphone. With a rock-bottom dumbphone price plan.

    Now imagine a customer who doesn’t expect a subsidy, and is willing to pay extra to combine these in a single device, but still just wants the rock-bottom dumbphone price plan from the network, with no data.

  89. A iPod-Dumbphone or an Android-Dumphone with no expected data plan requirement (contract or prepaid), for $200 or less, would be…glorious. It’d be worth $300 easy.

    It could come with voice and wifi, and even with 3G onboard but disabled (so if you *wanted* to pay monthly for it, sign up anytime). It’d be perfect in terms of don’t-waste-money till you need to.

  90. >Ummm, an always connected ultra portable computer with voice communication? (For those who care.)

    And this is different from an iPod touch with a cell modem how?

    >Seriously, though, the form factor makes a huge difference.

    Absolutely, And just like a 7 inch tablet makes a bad phone, a 3 inch phone makes a bad tablet, and a bad notebook computer and an even worse desktop. That’s why I don’t think that smart phones are the absolute future of the “post-PC” computer revolution. And to be honest, I don’t see the phone + dock = computer model taking over either. There’s still too many tradeoffs in that form of computing, especially when the dock costs as much or more than the phone, and is useless without it. Reminds me too much of the RIM Playbook.

    >And, BTW, I said this earlier (as did Eric) but the standalone PMP category seems to be dying. A 17% drop in sales for the company that
    >owns the market is huge, and is not a one-time event. Expect this to accelerate.

    But a non smart phone iOS or Android device doesn’t have to be a PMP. An iPod Touch is as much of a PMP as an iPhone is. The only practical difference between the two is the lack of the cell modem.

    >I don’t understand your point here. If you want to buy a smart phone without buying a data plan, obviously you would go for a full price pre-paid.
    >That’s precisely the person whom full price pre-paid is for!

    There are, I believe, plenty of people in that group that are still buying dumb phones, who don’t mind buying on contract, or don’t want to spend $150+ on a good smart phone (of which, when I asked a few months ago, I was told there still was not), and don’t want to spend $30+ /month for data access. I would gladly re-up my cell contract (well, pending AT&T merger not withstanding) if I could buy an android phone from T-Mobile on contract without attaching a data plan to it.

  91. @tmoney: There isn’t necessarily a fundamental difference that separates an iPod Touch and and iPad, but there is (in my opinion) a consumer difference. If you buy an iPhone or an iPad you’re getting a device that you intend to use for some sort of computing (at the very least you’ll browse with it). If you’re not happy with its computing functions, you are unlikely to be happy with the device.

    An iPod Touch is different. It is perfectly reasonable to buy an iPod Touch as a pure media player (with a larger screen and built-in speakers as its advantages over an iPod Classic or Nano) and many (I’d argue most) people do. In this case, computing functions are a bonus, not the core purpose of the device. If you agree with that context, then it is reasonable to apply a substantial discount to the iPod Touch as part of the installed base of iOS computing devices.

    How much of a discount should be applied? I don’t know. You’d need surveys of browsing and app usage by iPod Touch owners who do and don’t own iPhones (and the relative breakdown) in order to reliably estimate the discount. As far as I can tell, esr and others make the argument that iPod Touches should be discounted entirely for two reasons: (a) the intrinsic discount to apply is large (i.e. most iPod Touches are pure media players) and (b) the media player market is shrinking, not growing, so they’ll be increasingly irrelevant going forward. I wouldn’t necessarily go that far my self (the iPod Touch does add something to the iOS ecosystem), but I do think that discounting the iPod Touch entirely is closer to the truth than counting each iPod Touch as the equivalent of an iPhone or an iPad is.

  92. >An iPod Touch is different. It is perfectly reasonable to buy an iPod Touch as a pure media player (with a larger screen and built-in speakers
    >as its advantages over an iPod Classic or Nano) and many (I’d argue most) people do. In this case,

    I can’t speak for actual statistics, so take this for what it’s worth, but what I was selling them, I had plenty of customers who bought the iPod touch because they wanted the iPhone, but either they didn’t want to switch carriers (a group obviously going away) or bought because they wanted “an iPhone without the phone”

    That said, I should make it clear that I agree that every iPod touch is not equivalent to an iPhone, but I think discounting them entirely, and especially writing off any ultra portable computer, iOS or Android based as irrelevant because it isn’t a “smart phone” is short sighted.

    BTW, iPod touches make particularly useful corporate gifts, all the fun of an iPhone and no contract. At the company I work for now, there are plenty of people who have android devices, but they still have iPod touches because they got one as a gift. These people use their iPods regularly, and are just as much a part of the iOS ecosystem as they are of the Android one.

  93. Re whether the iPod touch matters or not.

    A lot of people are missing the point.

    Whether something that looks and smells kind of like a PMP is useful and used in the future or not is irrelevant to current trends.

    Samsung court documents show that Apple shipped 60M iPod touches. Apple’s quarterly shows that they shipped 9 M iPods (more than half of which were touches) last quarter, down 17% from a year ago.

    This decline is only going to accelerate. There are 5 billion cellphone users in the world — a waiting market for smartphones.

    If Apple managed to keep selling 9M iPods per quarter, then about 5 1/2 years from now, the number of iPods out there would be equivalent to about 5% of the cellphone market. That’s not going to happen — iPod sales will only continue to decline, and that will accelerate once iTunes is no longer a virtual monopoly for reasonable music sales and management.

    On the other hand, within 5 1/2 years, it is not at all unimaginable that there will be 3 billion smartphones (60%) on the planet.

    Eric has said many times that this is about where things are going, not where they are.

    And where they are going is — smartphones are eating PMPs for lunch, and tablets are still a bit of an unknown in terms of quantity.

  94. > And, BTW, I said this earlier (as did Eric) but the standalone PMP category seems to be dying. A 17% drop in sales for the company that owns the market is huge, and is not a one-time event. Expect this to accelerate.

    … Until the point where the iPod touch is what the market gives the kid as a ‘training device’ and portable gaming console. With wifi.

  95. I see the iPod Touch as an original example of what will be the future of personal computing: A WiFi/Bluetooth/USB portable computer with a small touch screen. Given the capital layout needed to set up over-the-air broadband over cell-phones, there will be problems to get 3 Billion and more humans on “data-plans” in only a few years. However, many places on earth are wired-up.

    What is more natural than to make that a franchise and let people use lcoal WiFi for their data needs? There are many entities who would love to supply you with “free” WiFi access. Starting with coffee houses, pubs, and public transport. And with VoIP, who needs cell phones if there is ubiquous WiFi in some form?

    While “the” original example of such a device, the iTouch is too locked down to do normal computing tasks. I think that is the primary reason the sales are going down. And, of course, the fact that Apple really wants all iTouch owners to move on to iPhones.

  96. @Winter:

    The places on Earth that are “wired-up” are the places that already have cellular infrastructure. Building cell towers is far, far cheaper wiring individual buildings.

    Here’s an article about the African mobile market: http://3g4g.blogspot.com/2009/10/african-mobile-market-grows-550-in-5.html The key quote is this: “At the end of 2007, there were eight times as many mobile phones as fixed lines in the least developed countries.” And fixed phone lines are the oldest, cheapest way wiring things up!

    I agree that the future of personal computing is going to be a portable computer with a touch screen. But for the majority of humanity, that computer is going to connect to the rest of the world using some sort of wide-area wireless technology. In other words, the future of personal computing is the smartphone.

  97. @Ravi,

    I agree that in Africa, the outlook is rather dim. However, in Asia they are pulling fiber everywhere. So if you are living in a city, cell towers might not be you most cost effective broadband. Especially when I hear stories about USA providers being unable to cope with the bandwidth demands of iPhone users.

  98. @Winter,

    Even in Asia, not everyone lives in cities. From the statistics I could easily Google, over half of the Chinese population and over 70% of the Indian population are rural. And that’s before we get to Indonesia, Thailand, Vietnam, Malaysia and so on

    Beyond that, there’s a *huge* gap between the more developed Asian cities (like Singapore and Hong Kong) that are hooking up fiber everywhere and cities like Bombay where even if you did hook up every building (which isn’t happening in the near future) you’d still miss a significant fraction of the population. And then there’s the fraction of the population that might well have access to wired broadband, but decides to use mobile broadband regardless.

  99. @Ravi
    “Even in Asia, not everyone lives in cities.”

    Obviously. But how does that make the telcos invest enough to beef up over-the-air broadband enough to link up billions of people on data-plas. And that within, say, 5 years?

    It is *much* easier to use voice only when you are literally in the field, and get in reach of a WiFi spot to get data access. And that is instant help for all those that already live in cities (over half the human population now). I know enough people who use their iTouch that way.

    It is simply much cheaper to set up hotspots for those in urban areas than to get broadband over-the-air coverage. So I really expect that this will in the end be the way it will happen. But if everyone can get cell phone broadband for cheap, even better.

  100. Verizon results are in.

    Their profit exceeded expectations, but their postpaid adds didn’t.

    They sold 2.2 million iPhones, which, when taken against AT&T’s 3.6 million iPhones, shows what several of us have been saying — there is a huge, pent-up demand for good and cheap smartphones.

    They reported 906M postpaid adds, but they counted 364M tablets as part of that, so cellphone adds would have been 542M.

    Total network additions was 1.8 million, 10% less than AT&T.

    There’s lots of numbers there, but the best one, considering how much iPhone V was going to help them kick some serious ass:

    “Wireless revenue was $16.9 billion, up 10.2 percent from a year ago.”

    That 10.2% number, ladies and gentlemen, is (a) not all that big considering their supposed data growth opportunities; and (b) identical to AT&T’s growth.

    So Verizon remains the largest wireless outfit, about 10% bigger than AT&T by revenue, and iPhone barely helped them not lose ground to AT&T.

  101. They sold 2.2 million iPhones, which, when taken against AT&T’s 3.6 million iPhones, shows what several of us have been saying — there is a huge, pent-up demand for good and cheap smartphones.

    In 2/3rds the time. If you extrapolate Verizon’s sales to the full quarter, you’d be at 3.3 million iPhones. You could claim that Verizon’s sales numbers are simply due to pent up demand, but since Eric said iPhone sales on Verizon were “anemic,” that’s still inconsistent with previous conventional wisdom around here.

    Other interesting facts: AT&T sold 2.7 million iPhones in Q1 2010. Thus, their sales were up year over year by around 25%. The interesting thing about this is that it suggests that Verizon’s sales were not at the expense of AT&T, as Patrick notes. But the obvious conclusion, which he did not draw, is that there were a significant cohort of Verizon customers who were waiting for the iPhone.

    Oh, and I think we can put the nail in the coffin on that whole rumor that HTC Thunderbolts outsold iPhones. 260K Thunderbolts; 2.2 million iPhones. It is entirely possible that the Thunderbolt had a great first couple of weeks, over a month after the iPhone was released… but the overall numbers are crystal-clear.

  102. @Bryant:

    > But the obvious conclusion, which he did not draw, is that there were a significant cohort of Verizon customers who were waiting for the iPhone.

    I already responded to this here: http://esr.ibiblio.org/?p=3137#comment-304539

    But since I’m dealing with somebody who loves to repeat himself, I guess I’ll have to repeat myself: NOBODY SAID THERE WASN’T SOME PENT-UP DEMAND FOR THE iPHONE FOR VERIZON. It’s just that the pundits were claiming a lot (many, many millions) of pent-up demand, and it turns out that the pent-up demand plus the natural run rate for half a quarter is slightly less than AT&T can sell (of mostly older model iPhones) in the same amount of time if they put their mind to it and cut the price a bit.

    > The interesting thing about this is that it suggests that Verizon’s sales were not at the expense of AT&T…

    Actually I’m sure some of them were. If you look at net postpaid subscriber adds on Verizon and AT&T vs. recent quarters, the expense to AT&T was probably measured in the 300K – 400K subscriber range. Not the droves that people were claiming would defect from AT&T for the “better” network.

    > since Eric said iPhone sales on Verizon were “anemic,” that’s still inconsistent with previous conventional wisdom around here.

    I showed you an article from January that predicted 25 million for the year. There were several like that. Are you really saying that 2.2 million in the first 7 weeks isn’t “anemic” compared to that expected run rate?

    > Oh, and I think we can put the nail in the coffin on that whole rumor that HTC Thunderbolts outsold iPhones. 260K Thunderbolts; 2.2 million iPhones.

    Not sure about that.

    > It is entirely possible that the Thunderbolt had a great first couple of weeks, over a month after the iPhone was released… but the overall numbers are crystal-clear.

    No, that’s probably backwards. The iPhone V and the Thunderbolt were released at the same time, but the iPhone V was pretty much a known quantity, and the Thunderbolt might have needed some word-of-mouth and reviews to get ramped up. The analysts have said that the Thunderbolt outsold the iPhone V in March. If true, that would mean that 2 million or so of those iPhones were sold in February, and the current run rate is as anemic as predicted.

  103. Really sad to see that Toyota is planning to use Microsoft in their new infotainment car system – “Entune”. Can’t believe they are putting Bing in their vehicles. Cross Toyota & Lexus off my shopping list.

  104. > Really sad to see that Toyota is planning to use Microsoft in their new infotainment car system

    They’re hoping that association with Microsoft’s legendary rock-solid software development skills will help alleviate the public perception that their sudden acceleration problems were caused by wayward programmers.

  105. As I pointed out earlier, one of AT&T’s interesting statements was:

    “At the end of the quarter, 46.2 percent of AT&T’s 68.1 million postpaid subscribers had smartphones…”

    One of Verizon’s statements was:

    “At the end of the first quarter, 32 percent of Verizon Wireless’ retail postpaid customer phone base were smartphones, up from 28 percent at the end of fourth-quarter 2010.”

    I don’t know the exact number for “retail postpaid customer phone base” because Verizon doesn’t break that out cleanly like AT&T, but even if it were only the size of AT&T’s (70 million), that’s 10 million missing smartphone upgrades on Verizon’s network compared to AT&T’s. (46.2% smartphones vs. 32% smartphones.) The discrepancy is actually much probably much bigger — the number of “missing” smartphones on Verizon is probably more like 11.5 to 12 million.

    It looks like AT&T’s early exclusive on iPhone did indeed help put the hurt on Verizon, and AT&T’s continued aggressive smartphone pricing and increased Android options are going to keep the pressure there. Smartphone customers pay more (data plans cost money), so Verizon wants more of those. That discrepancy between smartphone takeup on the two networks is probably one reason why a lot of analysts were projecting huge gains for iPhone V, but they appear to have overestimated consumer preference for Verizon’s network, and underestimated the willingness of the well-heeled who lust after data to migrate from Verizon to AT&T and stay there.

    So, to the extent that Verizon’s network really is preferable, I expect some adjustment of customers switching from AT&T to Verizon, but AT&T has an awful of customers locked up for another year and a half on iPhone 4, and a lot more customers locked up for two years starting last quarter on used iPhone 3. That gives AT&T some breathing room to integrate T-Mobile, flip a couple of switches to use the extra bandwidth, beef up their network, and do whatever other magic is required to increase customer inertia and reduce churn.

  106. >I showed you an article from January that predicted 25 million for the year. There were several like that. Are you really saying that 2.2 million in the first 7 weeks isn’t “anemic” compared to that expected run rate?

    2.2 million in 7 weeks would correspond to over 16 million for the year (yes, it’s not likely to sell at the exact same rate all year – you will get a big surge in the first seven weeks, but you will also get a bigger one in the holiday season). Are you really saying that 16:25 is an ‘anemic’ ratio??

  107. I showed you an article from January that predicted 25 million for the year. There were several like that. Are you really saying that 2.2 million in the first 7 weeks isn’t “anemic” compared to that expected run rate?

    That’s a straw man argument: you’re taking the far end of the spectrum and pretending that it represents the entire range of predictions. It doesn’t even represent the consensus analyst position.

    Note that the relative positions of the Thunderbolt and the iPhone on Verizon doesn’t even have huge relevance to the question of Android vs. iOS marketshare. It is entirely possible for the iPhone to destroy the Thunderbolt without changing anything substantial about the marketshare; in fact, it’s likely. That’s one of the big advantages Android has: it’s not limited by a single vendor. I’m perfectly willing to admit this, because it’s obvious. I am not entirely sure why it’s so hard for Eric and Patrick to say they were wrong about Verizon iPhone sales when it’s not going to change the overall trajectory of the marketplace.

  108. Of course, the other thing Verizon can do is to drop the contract price of its cellphones and its data plans. This would entice customers back to Verizon, and would also entice upgrades from a few of the tight-fisted voice-only dregs that currently make up a majority of Verizon’s customer base.

  109. That’s a straw man argument: you’re taking the far end of the spectrum and pretending that it represents the entire range of predictions. It doesn’t even represent the consensus analyst position.

    No, but it fairly accurately represents the position of the fanbois who were posting here back then…

    In any case analyst consensus was hard to come by:


    Analysts’ estimates for Verizon’s iPhone sales this year vary widely, from 5 million to 13 million — and some of that would come from what AT&T would have sold. The iPhone is big business for AT&T: The carrier activated 11.1 million iPhones in the first nine months of 2010, the latest figures available.

    Many analysts estimate that Verizon would be the largest seller of iPhone in the U.S. this year, outdoing AT&T as it satisfies pent-up demand. Verizon has been doing its best keep up with AT&T by selling smart phones other than the iPhone, but it’s still been lagging.

    If, in fact, Verizon sold fewer iPhones in March than the 260K Thunderbolts, then they aren’t yet on track to hit the bottom of that range.

    Note that the relative positions of the Thunderbolt and the iPhone on Verizon doesn’t even have huge relevance to the question of Android vs. iOS marketshare. It is entirely possible for the iPhone to destroy the Thunderbolt without changing anything substantial about the marketshare; in fact, it’s likely. That’s one of the big advantages Android has: it’s not limited by a single vendor. I’m perfectly willing to admit this, because it’s obvious. I am not entirely sure why it’s so hard for Eric and Patrick to say they were wrong about Verizon iPhone sales when it’s not going to change the overall trajectory of the marketplace.

    Well, I did say that, based on historical behavior, I didn’t think that Verizon would break out iPhone sales for the quarter. I was wrong about that. What else, exactly, did I say that has so far been shown to be “wrong?” Point it out, give evidence why it is wrong, and I will gladly admit it. But I’m not remembering any other mis-predictions at this point.

  110. 2.2 million in 7 weeks would correspond to over 16 million for the year (yes, it’s not likely to sell at the exact same rate all year – you will get a big surge in the first seven weeks, but you will also get a bigger one in the holiday season). Are you really saying that 16:25 is an ‘anemic’ ratio??

    Yes, especially if March sales are as lackluster as some analysts seem to think. Again, you have to look at context. The FUD which was being slung around here for a very long time was that the only reason that Android sold any devices ever is because of the AT&T exclusive, and that as soon as Verizon had iPhone, Android was toast.

    In any case, if I have engaged in a small amount of hyperbole, it is nothing compared to the hyperbolic predictions of Android’s demise and Apple’s ascension at the rapture.

  111. @I wrote: “I don’t understand your point here. If you want to buy a smart phone without buying a data plan, obviously you would go for a full price pre-paid. That’s precisely the person whom full price pre-paid is for!”

    @Patrick Maupin: “We’re not looking at it from the point of the carrier; we’re looking at it from the point of ourselves as customers. We’re not broke, but we’re cheap (hmm, maybe there’s a correlation there). Anyway…

    Imagine a small WiFi media player / browser — an iPod, if you will.

    Now imagine a dumbphone. With a rock-bottom dumbphone price plan.

    Now imagine a customer who doesn’t expect a subsidy, and is willing to pay extra to combine these in a single device, but still just wants the rock-bottom dumbphone price plan from the network, with no data.”

    I am looking at it from a consumer point of view (me!), but I was thinking of a higher price-point. Now I understand your argument, and it does make sense in the short term. In the long term, a true smartphone will be so cheap that there’s no benefit to getting the dumbphone.

    @twilightomni: “A iPod-Dumbphone or an Android-Dumphone with no expected data plan requirement (contract or prepaid), for $200 or less, would be…glorious. It’d be worth $300 easy. It could come with voice and wifi, and even with 3G onboard but disabled (so if you *wanted* to pay monthly for it, sign up anytime). It’d be perfect in terms of don’t-waste-money till you need to.”

    This, on the other hand, seems to be already covered by phone/plan combos like the Virgin Mobile LG Optimus V ($200 for the phone, $25/month for full voice/text/data) and Boost Mobile Samsung Prevail ($180 for the phone, $2/day for voice/text/data when used, or full voice/text/data for $50/month dropping to $35/month after a few months, no contract). (These do assume that you live in an area with Sprint coverage.)
    I thought Patrick’s point above referred to a much lower price point than $200+ and $2/day or $50/month.

    For comparison, my husband and I are currently paying Verizon about $45/month for a voice-only plan on an ancient dumbphone. I plan to ditch my Tracfone for a Samsung Prevail soon (as soon as it hits stores; it’s been announced but no one has it in stock yet), and I’m encouraging my husband to ditch the Verizon phone for either an Intercept or a Prevail. Yes, I’m cheap, but I also care about value-for-money, and I’m not getting a good deal right now.

  112. Willian B. Swift

    > Does Android have this feature, too. Tracking Your Every Move: iPhone Retains Extensive Location History.

    See

    http://www.socialaw.com/slip.htm?cid=20416&sid=121

    Is the relevant Google code in the handsets under the Apache license? If so, it should be easy to see how much location data it stores locally, but regardless, it seems like Google has lots of data of people’s whereabouts on their servers.

  113. >If, in fact, Verizon sold fewer iPhones in March than the 260K Thunderbolts, then they aren’t yet on track to hit the bottom of that range.

    To reiterate, Verizon reported that they activated 2.2 million iPhones in Q1, which basically means February and March in this case. Given that I reported this number before you wrote that, it seems like you’re losing touch with reality. Your gut feelings are out by a factor of four.

    1. >To reiterate, Verizon reported that they activated 2.2 million iPhones in Q1, which basically means February and March in this case. Given that I reported this number before you wrote that, it seems like you’re losing touch with reality. Your gut feelings are out by a factor of four.

      Dunno how Patrick will respond, but I’ll admit that is higher numbers than I was expecting – my estimates based on first-week activity were in the 1.8M range.

      I’m not sure it makes any substantive difference, though, because I also underestimated Android new-unit sales by a larger factor. Even on the very optimistic assumption that Verizon sustains that pace through Q2, Android phones are selling so much faster in aggregate (ratio of about 10:1) that iPhone V is barely going to budge the needle on the market share numbers (if that).

      I’m going to be scrupulous about pointing out the possible ramifications of my mis-forecast now so there won’t be any accusations later that I’ve tried to sweep it under the rug. 2.2M units in 60 days means the plausible range for Q1 and Q2 together tops out at about 6M units; a more realistic estimate would be 4M. They’re likely to beat 2.6M that I set as a threshold for “anemic” pretty easily, like in another three weeks. And if Android volume now had been where I thought it would be when the iPhone V launched, this would be pretty serious news. But it isn’t; it’s much higher.

      The iPhone didn’t need just moderate success on Verizon, it needed the massive multicarrier breakout various Apple fanboys were confidently and gloatingly predicting for over a year on this blog. It needed to rack up sales numbers high enough to exceed Android’s growth over the same period to dent Android’s momentum. That didn’t happen, and quibbling over the odd million or two error in unit sale projections is not going to change the fact that it didn’t happen.

      In fact, it didn’t even come within an order of magnitude of happening. 6 million units, the top end of the now-plausible Q1+Q2 range, is – depending on how you predict rate of growth in Android activations – between 14 and 17 days of Android sales.

      (Hm. I think this deserves to become a post.)

  114. @Bennett:

    To reiterate, Verizon reported that they activated 2.2 million iPhones in Q1, which basically means February and March in this case. Given that I reported this number before you wrote that, it seems like you’re losing touch with reality. Your gut feelings are out by a factor of four.

    Verizon reported that they sold 260K Thunderbolts, and two different analysts say that the Thunderbolt outsold the iPhone V in March. If both those statements are true, then the number of iPhone Vs sold in March would have to be under 260K, with the other 2M iPhone Vs sold in February. This is simple reasoning, based on a couple of widely reported (though as-yet unproven) statements, which is why I preceded my statement with “If, in fact,”. I described this entire line of reasoning in a post I wrote well before the one you are quoting from here, and I don’t have a dog in this hunt.

    I can learn stuff from really smart assholes, and can sometimes teach stuff to really nice idiots, but when you can’t follow simple reasoning, and then insist that I’m losing touch with reality, that puts you squarely in the idiotic asshole category, where no communication is possible. Whether you stay there or not is entirely up to you.

  115. The iPhone didn’t need just moderate success on Verizon, it needed the massive multicarrier breakout various Apple fanboys were confidently and gloatingly predicting for over a year on this blog.

    As I described in an earlier comment, I think part of the problem is that everybody who absolutely had to be a smartphone early adopter already shifted to AT&T. As did everybody who had to own an iPhone.

    That’s why smartphones are only 32% of total handsets on Verizon while they are 46% of total handsets on AT&T.

    Then Droid came along, and people who would like a smartphone, but didn’t necessarily have to have one, snapped that up.

    So when the iPhone V finally hit, the core market wound up being the yuppies who have to have Apple, who are living in those red states where AT&T isn’t available…

    Sure, others will buy as their old phones go off contract. Some will even migrate from AT&T as they go off contract. But AT&T has executed brilliantly with their iPhone trade-up project to stave off that day as long as possible, and then flooded the market with cheap iPhone 3s.

    Look at the math on the tradeup: By default, an iPhone 3 or an iPhone 4 would give them $200 up front, plus 2 years of contract, so somebody who upgraded when their contract was out would give them $400 + 4 years.

    But if they let them upgrade a year early for $50, and then resold the iPhone 3 for $50, they would lose $100 up-front revenue (plus some restocking labor), but would gain 1 total contract year on the two phones because the iPhone 4 would be under contract for 2 years, and the iPhone 3 would be under contract for a year, then another 2 years with the new owner.

    If AT&T executes with the bandwidth upgrade opportunity that is the real reason for the T-Mobile transaction, then they will have the network to beat before all these people go off contract. And if Verizon’s smartphone adoption rate rises to match AT&T’s in the meantime, we may find out exactly how bad or good their network is — remember, these are the people who were proactively instituting data caps before the first iPhone even showed up on the network.

  116. Summary of the patent:

    A method and apparatus for performing storage and retrieval in an information storage system is disclosed that uses the hashing technique with the external chaining method for collision resolution. In order to prevent performance deterioration due to the presence of automatically expiring data items, a garbage collection technique is used that removes all expired records stored in the system in the external chain targeted by a probe into the data storage system. More particularly, each insertion, retrieval, or deletion of a record is an occasion to search an entire linked-list chain of records for expired items and then remove them. Because an expired data item will not remain in the system long term if the system is frequently probed, it is useful for large information storage systems that are heavily used, require the fast access provided by hashing, and cannot be taken off-line for removal of expired data.

    Erm, isn’t this just “Robin Hood hashing”?

  117. JB Says:

    > …I don’t know of a way to buy cheaper than $20 ($21 w/tax) cards for TF, since 60 min/90 days are the lowest denomination.

    I have seen an option when I re-fill my tracfone (like you, $20 pre-tax every 90 days) for a year with zero minutes.
    But, since it is not MUCH cheaper ($70 pre-tax IIRC), I keep going with the $20 option.

    BTW, for an emergency phone, I paid $9.95 at a Staples for a Motorola flip-phone with an awesome battery life, and $20 pre-tax every 90 days, it is a good solution.
    And I do not think their customer services is significantly worse than Verizon or Comcast.

  118. Bennett,

    > To reiterate, Verizon reported that they activated 2.2 million iPhones in Q1, which basically means February and March in this case. Given that I reported this number before you wrote that, it seems like you’re losing touch with reality. Your gut feelings are out by a factor of four.

    Patrick is a careful, well informed disputant. You can argue with him (I’ve tried) but you have to have your facts straight and read carefully or you are likely to look foolish (I did, IIRC). I’m normally careful myself, but not always when something makes me mad. Oops!

    You can see that he isn’t loosing touch with reality and he doesn’t have a problem with his gut:

    >Verizon reported that they sold 260K Thunderbolts, and two different analysts say that the Thunderbolt outsold the iPhone V in March. If both those statements are true, then the number of iPhone Vs sold in March would have to be under 260K, with the other 2M iPhone Vs sold in February. This is simple reasoning, based on a couple of widely reported (though as-yet unproven) statements, which is why I preceded my statement with “If, in fact,”. I described this entire line of reasoning in a post I wrote well before the one you are quoting from here, and I don’t have a dog in this hunt.

    Bennet, my advice is to upgrade the care with which you are commenting on this subject. I know you can be very careful about what you write, because I’ve seen you do it.

    I say this because seeing two careful commenters going at it hammer and tongs is entertaining! Mmmmmm. Free ice cream.

    Yours,
    Tom

  119. @Patrick
    “Two different analysts say that the Thunderbolt outsold the iPhone V in March.”

    I’m asking for it, but why believe the analysts here?

    Assuming they’re correct, the only conclusion to be drawn would be that Verizon sold less than a third of a million iPhones in an entire month, which seems unlikely considering that the “anemic” Verizon iPhone sale rates reported at the end of February only *matched* AT&T’s (and hence, an anemic non-shutout). To then say that the iPhone’s sales rate suddenly dropped by a factor of 6 (for example, 1.8 million in February, then the remainder < Thunderbolt's sales) in March seems a little unrealistic.

    1. >To then say that the iPhone’s sales rate suddenly dropped by a factor of 6 (for example, 1.8 million in February, then the remainder < Thunderbolt's sales) in March seems a little unrealistic. I actually agree with this. It's normal for sales to ramp down after a consumer-electronics product release, but not by that much. Somebody's numbers are way off; we just don't know whose yet.

  120. @Tom DeGisi:

    You flatter me greatly :-)

    But I must say that it really is fun to come here and have all my assumptions and reasoning challenged.

    @twilightomni:

    I’m asking for it, but why believe the analysts here?

    I never said I believed the analysts :-) OTOH, their story might be believable. More anon.

    To then say that the iPhone’s sales rate suddenly dropped by a factor of 6 (for example, 1.8 million in February, then the remainder < Thunderbolt's sales) in March seems a little unrealistic.

    Possibly. But they “sold out” on some unknown number of preorders. Seemingly informed analyst speculation was that the preorders were only 500K or so, which I believe is one of the reasons that Eric postulated 1.8M for a couple of quarters. Even Eric himself said he was stunned at the apparent lack of interest at the time. Possibly GIGO.

    But what if the preorders were really bigger than that, say even 1.5M units? That would go a long way towards explaining why Apple got all bent out of shape and decided to forgo preorders in favor of photogenic long lines for the iPad 2 rollout — really a pretty drastic step, if you think about it. This is not an unrealistic possibility — according to suppliers Apple itself expected to ship 6 million CDMA phones for the quarter, probably at least half of those to Verizon:

    http://www.digitimes.com/news/a20101226PD203.html
    http://www.gizmocrave.com/3508-apple-iphone-shipment-goal-for-2011-q1-set-higher/

    Some analysts expected that Verizon would pick up up to 6 million AT&T refugees and/or sell almost 25 M units this year:

    http://www.sfgate.com/cgi-bin/article.cgi?f=/c/a/2011/01/01/BUJA1H1FFR.DTL&tsp=1
    http://www.infosyncworld.com/reviews/cell-phones/verizon-iphone-sales-potential-sets-new-record/11743.html

    With this sort of hype and expectations, and with the backstop of being able to sell CDMA handsets internationally (which apparently Apple is ramping fairly slowly, reading behind the lines on what Tim Cook said), it might have made sense for Apple to be ready to sell 2M handsets on opening day. Maybe they really did come close to that; maybe they weren’t lying when they said it was a phenomenal rollout for them, and they really did sell 2M handsets in February. I think I saw an article that said they actually built 3M CDMA units way back in December, but I can’t find that right now.

    If you go back and read what both Eric and I said before the opening day, I don’t think you can show that either of us would have been surprised by a couple of million units of initial sales. And maybe that’s what happened. If so, we’re certainly not the only confused ones, and we were confused by the seeming lack of interest on opening day.

    There is no question that there are Verizon customers who simply had to have an iPhone, but didn’t live in AT&T’s coverage area, for example. How many, I don’t know.

    So far, some of the best predictions I have seen for where the market is going come from polls about events that have already occurred. Unlike the analysts’ predictions of 500K preorders (which they were apparently just pulling from a dark smelly place), the “Thunderbolt outsells iPhone” meme supposedly originated from polls of knowledgeable people about what really happened. Of course, those need to be discounted a bit too, because they are only retail stores, and don’t reflect corporate Verizon internet orders.

    So it may not be for another 3 months (when we can see everybody’s quarterly results for the current quarter) that we can more accurately discern the trajectory of iPhone sales for last quarter. At least, we apparently won’t have the iPhone 5 messing up the calculations too badly.

    1. >Some analysts expected that Verizon would pick up up to 6 million AT&T refugees and/or sell almost 25 M units this year:

      See my next post. Even 6M unit sales wouldn’t haul Apple out of the strategic hole it’s in now.

  121. > See my next post. Even 6M unit sales wouldn’t haul Apple out of the strategic hole it’s in now.

    6M refugee sales wouldn’t have any long-term results. “refugee” implies somebody who just had to have an iPhone, but will ditch AT&T in a heartbeat. In other words, any iPhone V sold to such a person would simply be a one-time deal, replacing his preexisting AT&T iPhone.

  122. @Cathy

    Seems to be already covered by phone/plan combos like the Virgin Mobile LG Optimus V ($200 for the phone, $25/month for full voice/text/data)

    I think the Optimus is actually one of the coolest Android lines. It’s simple and affordable, and I really like the styling on Sprint (AT&T’s is a little wonky). I could easily see tons of people using them off-contract.

  123. Well, as an outsider of the programming world (I Stumbled upon here and chose to keep track of this, it looks interesting), I find amusing that Apple fanboys really pretend that the most expensive product in the market be the best selling one too. That doesn’t make sense anywhere. I am myself an iPod 4 user at the moment but I see it as a Mercedes – Benz – Would Mercedes drivers be sad if the vast majority has Volkswagens? It is the obvious state of the market: if iPhones actually managed to top sales at a given moment, there were many unnatural amount of factors going on, and now the market is slowly but steadily returning to reality.

    Apple and its fanboys should be satisfied with a minor but strong niche, but yeah, just like Mercedes, Apple is screwing up. There are still some flaring deficiencies on iPhone / iPads / iPods like the need of another computer to add video directories. That, combined with their stubborn attempts at monopolizing multimedia content are the reasons why the iPhone may lose even the very own niche that would otherwise love to keep it alive.

  124. “I have seen an option when I re-fill my tracfone (like you, $20 pre-tax every 90 days) for a year with zero minutes.
    But, since it is not MUCH cheaper ($70 pre-tax IIRC), I keep going with the $20 option.”

    Yeah…it makes no sense to pay up a year ahead in case you lose the sucker…

  125. JB Says:

    >> I have seen an option when I re-fill my tracfone (like you, $20 pre-tax every 90 days) for a year with zero minutes.
    >> But, since it is not MUCH cheaper ($70 pre-tax IIRC), I keep going with the $20 option.”

    > Yeah…it makes no sense to pay up a year ahead in case you lose the sucker…

    Although, they will let you transfer unused minutes (I have not tried transferring days) from one device (still starting at 9.99 USD) to another.
    And my two experiences with live customer service were no worse than typical with Comcast or Verizon.
    No financial stake for me…I just think they perfectly serve the niche for an emergency cell phone.

Leave a comment

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