Browser Trends May 2012: IE9 Strikes Back

Contributing Editor

It’s been one month since we last looked at the browser market. Since then, Firefox 12 and Chrome 18 were released and Microsoft has embarked on an aggressive television, movie and online marketing campaign for IE9. Has any of this affected usage patterns in the worldwide StatCounter statistics?…

Browser March April change relative
IE 9.0+ 14.56% 15.67% +1.11% +7.60%
IE 8.0 16.00% 14.69% -1.31% -8.20%
IE 7.0 2.91% 2.54% -0.37% -12.70%
IE 6.0 1.34% 1.17% -0.17% -12.70%
Firefox 4.0+ 21.11% 21.38% +0.27% +1.30%
Firefox 3.7- 3.88% 3.48% -0.40% -10.30%
Chrome 30.92% 31.29% +0.37% +1.20%
Safari 6.71% 7.14% +0.43% +6.40%
Opera 1.76% 1.70% -0.06% -3.40%
Others 0.81% 0.94% +0.13% +16.00%
IE (all) 34.81% 34.07% -0.74% -2.10%
Firefox (all) 24.99% 24.86% -0.13% -0.50%

The table shows market share estimates for desktop browsers. The ‘change’ column shows the absolute increase or decrease in market share. The ‘relative’ column indicates the proportional change, i.e. another 12.7% of IE6 users abandoned the browser last month. There are several caveats so I recommend you read How Browser Market Share is Calculated.

Chrome’s monthly growth has fallen below 1% for the first time since October 2011. Other vendors dream about a market share increase of 0.37% but it’s unusually low for Chrome. I’m not convinced adoption rates have begun to plateau and it could be a statistical blip. That said, the advantages Chrome holds over the competition are more marginal than they’ve ever been.

This month’s biggest winner is IE9 and it’s risen 3.6% since the start of March. Three factors have contributed to it’s success:

  1. Automated browser updates.
  2. Microsoft’s glossy marketing campaign.
  3. Businesses migrating to Windows Vista/7.

However, Microsoft cannot be complacent. IE8 is falling faster than IE9′s gains and total IE usage dropped by 0.74%. Chrome is still poised to overtake during the summer of 2012.

Safari had another good month and the browser is slowly creeping toward double-digit usage figures. The iPad accounts for 2.12% of the 7.14% total which means it now has more users than IE6.

Firefox and Opera’s market share barely changed. Perhaps this isn’t surprising; there are few major differences between the top five browsers and we could be entering a period of stability with little significant movement in browser market.

Mobile Browser Usage

April’s mobile usage increased to 9.58% of all web activity. This is partly owing to better weather in the northern hemisphere — although the UK had its wettest April for 100 years.

The primary mobile browsing applications are:

  1. Opera Mini/Mobile — 21.52% (down 1.34%)
  2. Android — 21.31% (up 0.15%)
  3. iPhone — 20.04% (down 0.06%)
  4. Nokia browser — 11.42% (down 0.44%)
  5. UC Browser — 7.77% (up 1.42%)

There’s little change within the top five although UC Browser has helped itself to another slice of Blackberry.

While it’s great to see healthy competition in the mobile market, it remains frustratingly difficult to test web sites and applications on a range of popular devices. Do you bother?

Free book: Jump Start HTML5 Basics

Grab a free copy of one our latest ebooks! Packed with hints and tips on HTML5's most powerful new features.

  • Sujay Krishna Suresh

    Why isn’t Chrome for Android considered in the Mobile browsers?

    • http://www.optimalworks.net/ Craig Buckler

      Because it’s currently sitting at position 21 in the chart of the top 42 mobile browsers with a 0.03% market share. It’s outpaced by IE, Firefox, the Sony PSP and even the Nintendo 3DS.

      • http://www.dave-woods.co.uk Dave Woods

        I agree, Android Chrome can’t be considered yet due to it only being available on ICS (as far as I’m aware). I’m sure that as users upgrade their Android OS or Google release it to all versions of Android that those stats will pick up and it’ll be at least 4th on that list very soon.

  • Patrick

    Testing a range of browsers is annoying, but not so much as it used to be. Tools like Adobe’s BrowserLab make checking the visuals of a page easy. Using it, you can check how a page appears on a wide range of browsers and operating systems and quickly pinpoint problems. Testing interactive features (i.e. JS) is a bit more challenging, but I use jQuery for most sites, and it takes care of browser compatibility issues for me. Pick a JS library, and trust it to do its job. Testing mobile browsers is a little tricky, but I tend to just test in one (I usually only have iOS available to me) and trust that it probably displays okay in other mobile browsers too.

    I also find that as long as you write good code and stick to well-established features, you don’t run into many issues now. IE8+ and every other browser will display almost everything correctly. IE6/7 or other older browsers will sometimes suffer the odd rendering glitch, but it’s rarely enough to make the pages unusable in those browsers. I generally don’t bother fixing them – it gives those users another incentive to upgrade. Maybe I’m just writing better code than I used to, but browser testing is now a relatively minor task that I tick off at the end of a project, right before public deployment. The only exception is if I’m working on a really important site, in which case I do break out a few different operating systems and hand-test the site in dozens of browsers. But I rarely find myself needing to fix anything, and I can’t remember the last time I wrote browser-specific code (largely thanks to the demise of IE6).

  • http://www.redubl.com Dustin James

    To answer your question at the end of the article – not really.

    Coding work arounds for IE 6 and 7 is increasingly irrelevant. All that work for less than 4% of the market? I can’t imagine a scenario where that might be important anymore. Often, people argue that IE 6 + 7 should be considered when developing for enterprise solutions but again…less that 4% of the total market – are these really the users we’re catering to? I don’t even think you can consider them late adopters anymore…just not adopters at all :)

    Mobile is increasingly important, but again, most new devices can expand/shrink content on their own. I’m all for fluid layouts, but when it comes to mobile I ask one question and code for one thing – what’s the purpose of this site and make sure that’s above the fold.

    In any case, with increased standards support it is slowly, painfully but surely becoming less and less important to make something work in all of the ‘legacy’ platforms. In reality, those users are not likely our target anyways.

  • Steve

    “April’s mobile usage increased to 9.58% of all web activity. This is partly owing to better weather in the northern hemisphere — although the UK had its wettest April for 100 years.”

    Was this an attempt at humor? Mobile usage increased because of the weather?

    • http://www.optimalworks.net/ Craig Buckler

      Absolutely. More people venture out or have vacations in warmer weather so they’ll use the mobile web more. Mobile web usage patterns are affected by the seasons — it’s higher in the summer than the winter. There’s usually a blip in December too because people receive mobiles and tablets as presents.

  • smithy

    “Often, people argue that IE 6 + 7 should be considered when developing for enterprise solutions but again…less that 4% of the total market – are these really the users we’re catering to?”

    - You have to know your audience … we still tested in Mac browsers before mobile, when that was ‘only’ a 3% share, just because some Mac-based studio might see our studio’s site. I just heard that a particular dept. at the BBC still use IE7: if you were doing business with them you’d want to make sure your site degraded gracefully otherwise they’d think you’re a muppet (the irony, etc – but sometimes the browser upgrade is prevented by the O/S – and what do sysadmins care about standards-compliant browsers?) Your client may have IE9 at work and Safari god-knows-what at home. Always pays to ask.

    “I also find that as long as you write good code and stick to well-established features, you don’t run into many issues now. [snip] Maybe I’m just writing better code than I used to..”

    I agree – back when IE6 was big I tried not to use CSS that would cause a major cross-browser headache. You do get into a certain style – unfortunately you end up writing inoccuous IE6 fixes like “display:inline” and eventually forgetting why. Nice to be able to break out a bit now!

  • http://www.stratusbusinesssolutions.com Kurt

    It would be very nice and extremely useful if someone would undertake a survey of the browser use by business users. Are business users as apt to use Chrome or Firefox, when they may not control their IT decisions?

    My guess is that the overall browser statistics are heavily skewed by the home user, who have the leisure of being able to experiment with one browser after another.

    • http://www.optimalworks.net/ Craig Buckler

      It would be useful, but it’s not particularly easy. First, what’s a business user? I’m a sole trader and can use whatever browser I like — but I’m still a business. Even if you determine what defines a business, how do you separate the user groups?

      It’s generally assumed most business users use what they’re given and, historically, that would have been IE. That’s less likely today but I suspect business users skew the statistics just as much. IE usage drops during the weekend.

  • http://crowdogs.com Peter Schreiner

    No mention of Maxthon? The #1 in HTML5Test.com. Surely it out-paces IE6, which had a mention.

    • http://www.optimalworks.net/ Craig Buckler

      Maxthon accounts for 0.11% of the browser market. IE6 has 10x more users. Pace is not an indicator of popularity.

  • Fred B

    “While it’s great to see healthy competition in the market, it remains frustratingly difficult to test web sites and applications on a range of popular devices. Do you bother?”

    Yes, I “bother”! I’m building all new sites using adaptive/responsive design and gradually converting many of my clients’ older sites.

    I feel it’s a responsibility to my clients that visitors to their sites, using any device, have an equally good experience.

    • http://www.optimalworks.net/ Craig Buckler

      But do you test your sites on a range of iPhones, Android, Blackberry, Symbian, etc. phones using a variety of browsers? How large is your mobile test suite?

  • http://JohnWolfBooks.com John Wolf

    The only browser testing I rely on is my own. IE has a peculiar hold on some banking sites and the like. Other browsers just don’t run some of their banking functions like online check deposit. So I had to use IE for that. I used FireFox for years as the preferred, but with the latest Chrome all is changed. It really does a nice job. I have switched to Chrome. I use WinXP, so I’ve abandoned IE altogether. The banking program I use works with Chrome and now the latest FireFox. The browsers seem to be converging to become indistinguishable except for funky font differences. I’ll stick with Chrome for now.

  • http://www.rackservers.com.au/ Rob RS

    it’s really a question of predicting what your client’s clients will be using… and do your clients know what their clients will be using? usually not but a bit of thought can save some coding workarounds.

    and with so many people still using xp and older computers, we are still in a catch-all environment. one day our economic culture of planned obsolescence will be abandoned and we can spend our energies on real development instead of catering for such a variety of uniqueness!

  • http://www.xoogu.com/ Dave

    I got 6.9% and 5.1% of users on IE6 for my two main websites in April :(

    Maybe if I stop supporting IE6 those figures will drop?

    • http://www.optimalworks.net/ Craig Buckler

      Well, yes, it almost certainly would. But how will that help if your biggest customer is an IE6 user?!