With an ever increasing range of browser technologies, versions, flash, HTML5, migration of applications to the cloud ensuring that your code is 100% browser compatabile is more important than ever. Browser testing is a key mitigation for some of these issues.
Indeed browser user experience is a key component and indictor of other wider problems that can have a detrimental impact on your site SEO and wider performance and usability.
Which Browsers Should I Test?
CPC Training. Are you using cutting edge technologies (HTML5), writing plug-ins or add-ons, testing out browser betas, or simply trying to make sure your site looks and works properly across the mainstream browsers?
There are several sites dedicated to tracking browser usage. They all differ slightly in their numbers, but they generally agree. However, their numbers are not always applicable. What I mean by this is if you are a web design london website, you probably need to test later versions of a browser than a static site advertising a local business.
Also take a look at:
and read the sections about over and under estimation. These too skew the results.
So while we can’t ever know the exact percentages, we should be able to get a good idea of what to test by looking at several of these sites and averaging out their results. It is a best guess and that is about as good as you’ll get.
If you follow the link above, they have a nice table of the latest stats:
Hmmmm… If you look at the data from earlier this year, you’ll see a significant change. Most notably, the fact that Chrome has stolen market share from both IE and Firefox. As a matter of fact, on http://gs.statcounter.com, if you just look at browsers without the versions, you’ll see that Chrome has, in fact, surpassed Firefox!
So what versions of each browser should you be testing? Again, it depends. (see above) If you are looking for just mainstream use, StatCounter has a really nice graph to show you that:
Source: StatCounter Global Stats – Browser Version Market Share
Top 10 browser test?
- Internet Explorer 8
- Firefox 3.6
- Internet Explorer 7
After this, things get tricky. These were the easy ones. After this, the sites referenced start to disagree a bit. In some cases more than a bit. I’ll say that those top three are the most used browsers and are in order of popularity. I’ll list the rest in relative order of importance, but not everyone agrees on the exact usage. Still, I think most would agree that these should be included in your testing.
- Internet Explorer 9
- Firefox 4
- Internet Explorer 6
- Chrome 12
- Safari 5
- Firefox 5
- Chrome 13
If you include all of the above browsers in your battery of tests, you should hit the majority of the browsing population. Though take note that on this graph, 20% are classified as “Other.” This is not an insignificant statistic and presents a bit of a problem for us all in trying to determine exactly which browser/version combinations that should be tested other than the 80% shown above. Though if you follow the 80/20 rule, then the above set is probably your best bet.
There will always be something else to test of course. The numbers are changing as I type this. New versions of the browsers are coming out monthly. And of course now you have all the mobile platforms to check too. They may be statistically way down on the list, but you might want to throw them in just to make sure your site is at least usable by the millions of mobile users. Our site makes it easy to do that so why not?
I hope you find this information useful. If you want to dive into the details, be sure to follow the link to wikipedia and check out their sources. They have plenty of numbers to look at I assure you. We wanted to put this page together to help our users focus their testing and get the most value from our service.