A Google Analytics Case Study in Internet Explorer Usage – June 2013

BeSeen / Blog  / A Google Analytics Case Study in Internet Explorer Usage – June 2013
Google-Analytics-icon

A Google Analytics Case Study in Internet Explorer Usage – June 2013

In another post on the BeSeen blog, I outlined how we dropped official support for Internet Explorer 7 (IE7) for new website builds (unless we have a very compelling reason to include IE7 support).

This is a (somewhat delayed) follow up post to give an overview of some real life Google Analytics data from some of our clients (kept anonymous). If you haven’t axed IE7 support yet, and you feel like its dragging you down, hopefully this article will help you decide to escape the shackles of IE7 forever! If your clients ask why, feel free to reference this article!

The Data

Ok so here’s the raw data on Internet Explorer usage for a selection of our clients’ websites (and our own). Instead of giving the business name or URL, I’ve given a business sector/description which may be of interest to you.

Data is given as a percentage of total visits over the past 6 months (26th Dec 2012 – 26th June 2013).

EDIT (25/7/2013): I sourced data for a London borough council website and added this to the table below and updated the values accordingly. The number of visits that this site generated YTD was over 3.5 million. The number of visits with IE7 was 178,537, the number of IE8 visits was 512,775 and the number of Chrome visits was 128,805.

Business Type % of visitors using IE7 % of visitors using IE8
Digital marketing agency (our website) 0.5% 3.7%
Industrial motor company 3.8% 23.6%
Roofing supplies company 1.7% 9.0%
Guitar parts company 0.9% 6.5%
Women’s fashion blog 0.1% 4.0%
A London borough (local government) 5.0% 14.4%
Average: 1.4% 2% 9.4% 10.2%

The total data sampled amounted to over 300,000 3.8 million visits.

The Interpretation

So what does all this mean? Well here are a few bullet points for rapid digestion:

  • The overall % of visitors using IE7 is low!
  • The % varies across different markets and sectors. The target audience evidently has an effect.
  • In some cases, IE6 had slightly more users that IE7 for a site!
  • The stats indicate that IE6 and IE7 are definitely both legacy browsers with a very small user base (no kidding!)
  • IE8 still has a significant user base, so we won’t be dropping that quite yet!
  • ADDED (25/7/2013): government / council websites not only have a higher % of IE7 visitors than the other sectors analysed in this case study, but the number of visitors are also significant in their own right. YTD, 178,537 visitors using IE7 visited the local council website used in this case study. Not a number to be sniffed at!

Hope you enjoyed the read. Please do leave comments with feedback and any of your own stats or experiences. What % of IE7 and IE8 users are your customers receiving?

Until next time friends…

  • June 26, 2013

    Interesting. I’ll have the check my stats out and see if they tell the same story. I tend to find these days that people don’t ask for IE support (at any level). I don’t cater for less then IE 8 unless it’s specifically requested, which is hasn’t been for at least a year now. I don’t see any point in catering for IE 8 either as there is 9 (and 10?) and stable sites (our own not clients) don’t cater for it. People need to move with the times. Software updates quickly these days (even IE!) and people should be adopting new, better, safe,software faster.

    I wonder if the “Industrial motor company” visitors are corporate visitors? Always seems it’s corporations that have a convoluted IT policy which run old(er) software.

  • June 26, 2013

    Those are interesting stats, and they fairly closely mirror our experience at GPMD.

    We no longer support IE7 (unless there’s a strong case for it), but although we would love to drop support for IE8 there is still enough traffic to not be able to do that at the moment. Many of our clients are in the ecommerce sector, so we simply can’t ignore 5-10% of their potential customer bases.

    Am I right in thinking that Microsoft have now stopped ‘supporting’ IE8 though? And with many SAAS apps (not least Google Apps) dropping support I can see a situation not long from now where we no longer have to worry about it – Goodbye polyfills!

  • David Smith

    Leave reply
    June 26, 2013

    Interesting figures which are roughly the sort of numbers I’m seeing across the sites where I have access to the data.

    My view is that current browser and one version back is good enough for browser support and then ensure that the site degrades gracefully in previous versions older than -1

    Chrome and Firefox with their better experience and also auto-updating are makeing a big difference for developers (to the good for now)

    If of course a client has a specific browser requirement (often this happens with intranet work) in which case we take account of the additional work with a cost uplift (ie you want IE 6 support it will cost you and you can’t have a lot of the “cool” stuff)

  • June 28, 2013

    What does “support” mean? We don’t test it? The site won’t work at all? The site might be partially usable but we’re not going to break our backs to get it to work? Google has dropped support for older IE versions, but my understanding is you can still try and use the site. It may or may not work. They’re just not going to listen if you raise a complaint/bug/problem.

    I think it all depends on the target(s) for your website.

    Those browsers are still out there. Government could well still be on IE6 although we have just moved to IE8 which incidentally is the latest IE version that Windows XP can support (and there is still just under a year to go of Windows XP support from Microsoft – which means IE6, 7 & 8 are technically still in support).

    It’s a call you have to make. From a government perspective we are trying to promote digital services by default and digital inclusion. This will mean that we will probably have to continue to support some legacy browsers for far longer than other commercial sites (if only to support our own staff accessing the site!). This will vary by region though. We cover a fairly affluent area and are now seeing around 50% mobile usage to our website with that split pretty evenly between Android and iOS.

    What is the cost of not supporting an older browser? For an estore or a government site potentially high. It could be lost sale(s) or in the case of government, someone calling up to perform the same transaction – which is much more expensive. This is all vs the additional development & testing costs.

    There is something to be said for dropping support for older browsers as well. People will never move if stuff still works well enough. Now that I think every newer browser version has the option of continuous automatic updates we hopefully won’t get in this situation in future. I’m finding they’re all rendering pretty close now so for the most part stuff just works in them all without any CSS hacks etc.

    As a developer I also don’t want to be stuck supporting older browsers that require stupid hacks and have illogical nuances. I want to concentrate on building shiny new things using the latest stuff.

    Whilst we’re on the subject of stats and browsers. Another interesting stat to look at is the ISP. We can say with a reasonable degree of confidence that most mobile users are infact browsing to our site using a fixed broadband connection. Why is that of interest? Well it means we can look at using more scripts and graphics without there being a penalty in terms of download. We can also perhaps be a little lazy and not optimize things as much as we could because it’s not worth the effort for the saving.

LEAVE A COMMENT