All quiet on the PC front, Android passes iOS (maybe) in mobile

All quiet on the PC front, Android passes iOS (maybe) in mobile

Summary: It's the first of the month, which means it's time for another look at the state of the PC and mobile markets, using the latest data from two large web analytics firms.


It’s a longstanding tradition on the first day of each month for tech reporters to look at the latest releases from Net Applications and StatCounter, trying to sort out who’s ahead and who’s falling behind in the PC and mobile horse races.

This month, in desktop operating systems, the continuing story is the replacement cycle spurred by the end of support for Windows XP three months ago, evidenced by a sharp drop in XP usage and a slight increase in Windows 7's share of usage. Both analytics companies say the percentage of PCs running some version of Windows is still around 90 percent and essentially unchanged from a year ago.

In that time, both companies agree, Windows 8 and 8.1 combined now make up at least one out of every eight PCs in the world, somewhere just shy of 200 million. As I've explained many times in presenting these numbers, it's more important to look at trends than to focus on each month's statistical blips. Here’s a chart comparing the figures for the past year from both firms.


Using StatCounter’s figures, which are based on pageviews, Windows 8 is probably a month away from replacing Windows XP as the second-most-used PC operating system, behind Windows 7. The Net Applications (NetMarketShare) numbers, which purport to measure unique PCs, have XP still solidly in second place with Windows 8.x flat over the past quarter.

Meanwhile, StatCounter says Chromebooks account for 0.19 percent of worldwide usage, with Net Applications still reporting that Chrome OS usage is below the 0.01 percent threshold.

In the second horse race on the card, Net Applications says Chrome has displaced Firefox as the second most-popular browser in the market, 20.4 percent to 15.1 percent. Last year at this time, Firefox was ahead by roughly the same margin. Internet Explorer continues to hold its 50 percent share on the desktop. In an alternative universe, StatCounter says Chrome is used twice as much as second-place Internet Explorer, 48.7 percent to 23.5 percent, with Firefox in third at 19.2 percent.

And in the final race of the day, mobile operating systems, StatCounter says iOS is far ahead among tablets and Android has a commanding lead in smartphones and similar mobile devices, with nothing resembling a clear third-place contender. Net Applications, which combines Mobile and Tablet into a single group, says Android has now passed iOS by a fraction of a percent, with the two OSes combining to make up 89 percent of total usage.

Credit: Net Applications monthly report for July 2014

Those mobile stats are highly unlikely to bear much more than a passing resemblance to actual device usage, largely because so little mobile usage happens in browsers, and web analytics companies can’t easily track app usage.

If you’re curious about why two analytics companies can have such wild results and how reliable their numbers really are, see “Net Market Share vs. StatCounter: Whose online measurements can you trust?”

Topics: Operating Systems, Mobile OS

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Other than "out of tradition" or "force of habit", why should one bother

    with these types of projections or reports from these third party analytical firms?

    For example, in a recent Appleinsider editorial entitled, "Apple, Inc's double digit U.S. Mac growth contradicts IDC & Gartner reports of a Mac sales slump", Daniel E Dilger writes, "Apple reported "strong double digit growth" in it's Mac sales in the U.S., directly contradicting the earlier estimates published by IDC and Gartner that stated Apple's U.S. Mac sales fell year-over-year in the June quarter and calling into question the legitimacy of market estimates that the tech media uncritically presents as factual."

    As you stated, Ed, the firms you site in this blog will publish stats and opinions that "... are highly unlikely to bear much more than a passing resemblance to actual device usage."

    Perhaps the only persons that should be interested in reading those figures are the company execs that run the firms that are being reported on. And if I had any advice to give to those execs, it would be this. Read those reports with more than a grain of salt.
    • Yes, those stats are stats

      In other words they measure certain things (web usage patterns)...

      For certain populations (USA for StatCounter, more "global" usage for NetStats)
      For certain devices (NetStats still clobble together smartphones and tablets)
      For certain websites (IIRC neither one is used on top 100 most popular web sites)

      Etc. Etc. Etc.

      And last but not least we try to guess completly different (while 'somehow' related) metric, by aproximating those stats. (Operating System usage vs Web Browser usage!!!)

      But with all that said, one can still see some patterns.

      Like with above article, which try to judge trends and be weary of pitfals of choosing "suitable" numbers.

      (Counter examples would be picking "bigger" number for the **preffered** software. Or ignoring regional biases present in both surveys)
  • Sad really ...

    ... That they still haven't learned to count Chrome books.
    • They have.

      They did have it on their previous graph, but the number was so low people just assumed it was a mustard smudge in the corner of the photo, so they removed it. Here, we'll recreate it -

    • They don't bother because

      Chromebooks aren't selling well enough to make a difference, and Google will probably kill off Chromebooks as they move everything over to Android.

      As Tablets mature and gain more processing power, memory, etc, a tablet with an attachable keyboard or dock will become the new portable computing platform and Google will then focus their efforts solely on Android and quietly kill off Chromebooks. Chromebooks will be nothing more than a short lived transitional device with limited appeal.
      • Actually, Google is moving native Android apps to Chrome OS

        This makes more sense to me than putting Android, a mobile OS, on a laptop or desktop PC form factor. In addition, I believe that Google included more security design up-front with Chrome OS than it did with Android.
        Rabid Howler Monkey
        • Seems odd given the popularity of Android

          I guess we will see. But my money is on Android over Chrome OS
    • Or

      Maybe it's because Chromebooks make up less than half a percent of net assuage and are thus under "other".
      Michael Alan Goff
  • We know that android is over 80% and iOS 12%

    I wonder where's the interest of issuing numbers that say they are close of being 50/50?
    The difference is like 7 times bigger vs a tie!
    These numbers are becoming more about the visitors of some few sites than any valuable data.
    • *Le sigh*

      You've kind of answered your own question.

      The information *we know* is the Wall Street journal article that says 85% of smartphones sold last period were Android. Not 85% of smartphones in the world are currently android.

      So we then look at these graphs over equally long periods, and oh! What do you know? They match! See how the ios graph is plummeting and the android one is skyrocketing? That pretty much confirms the Wall Street journal article research right? - android sales growing faster than ios even though ios sales continue to grow themselves.

      So why does it matter? Well if you were a developer, and only had your source, you would assume that android devices made up 85% of usage and that would affect how you developed your site, targeted ads, even the average screen size your audience is using.

      What this is saying is that whilst the numbers are heading that way (I see them balancing at around 90-91% android globally) we are not there yet. If you want accurate information on your customer base, it is currently closer to 50/50, but plan for that to change drastically is the current message from these reports.

      Usage stats have been the holy grail for developers for as long as there have been rival platforms (so since day one basically) in the eighties and earlier, all they had to go on was device unit sales by manufacturer combined with unit sales of software and accessories for the platform.

      With the internet, and app stores we now have considerably more data, but it is still a guess. Each study has it's flaws - the 85% that ios doesn't compete in the budget market, the 50/50 that actually the trend is very much down ward.

      So basically, yes. Lots of studies are needed so that companies, developers and the like have a good idea how their customers use their service and can improve the delivery of their services to their customers.

      News articles about the research for some kind of childish popularity contests aren't actually why people find the research.
      • WEP/Dumbphones/Featurephones

        Also its important to state that "smartphones" by themselfs account for ~50% of sold phones.

        So there are tons of WAP-enabled phones that are not on that graph cause both bean counter do not count WAP web.

        Global migration to smartphones should end by the 2020's. So its still some time when such stats can even account for "whole" picture.
        (Dumb/Feature phones being rought equal to smartphones in unit sales being bigger then "IBM PCs" - PC&notebooks - being bigger by twice then tablets. So those stats can not account for 20-30% of devices *sold right now*, and even more in installed base)
  • Apple skewed

    Agreed. And Ed Bott's contention you can't go by mobile browser useage is laughable. I use my mobile browser several times a day too look up stuff. he's only saying that because he's doesn't want to admit Apple is losing 85% to 12% in the mobile market.
    • ignore the type

      "to", not "too". Typing TOO fast.
    • Wait a minute...

      Wait, you're saying I'm an Apple fanboy!

      Ed Bott
    • I don't use the browser all that much on my smartphone.

      That's why everyone is talking so much about app count. Individual apps have basically replaced website for many mobile apps.
    • Wait, what?

      When did Ed Bott become an Apple fanboy?
      Michael Alan Goff
      • You Must Have Missed It

        AKH had the Pope fly in and they had a secret ceremony (robes and all). Mr. Bott was christened: Cardinal Pomme du Cafe' avec Latte'.
    • Nice try, but it doesn't fly.

      Being this blog used to be called "The Ed Bott, Microsoft report", or something close to that. I do not see Ed as a "Crapple Fanboy", no matter how you try to paint him in that light.
      I hate trolls also
    • well if we want to go by browser stats

      there's a lot of ifs. first, we have to assume that people on different OSs use the browser a similar amount (whether that amount is low or high doesn't matter as long as it's the same). second, we have to assume that people on different OSs visit websites counted by these groups the same amount. both iOS and android have big user bases so it's somewhat reasonable to assume that these variables should even out for the most part, but it should still all be taken with a big grain of salt.
  • If we give importance to these stats

    There is a curious aspect about netmarketshare numbers - windows 8 is down again and now being below vista regarding adoption.