I don’t understand what the heck is going on with browser usage statistics. I honestly have no clue which browser has the most market share. If you look a different sources you get radically different numbers.

W3 SchoolsNet Market ShareWikipediaGlobal Stats counterW3 Counter
Internet Explorer11.80%57.79%20.47%28.56%23.90%
Firefox28.20%18.58%17.71%18.36%17.80%
Chrome52.90%15.98%46.02%40.80%31.60%
Safari3.90%5.77%3.10%8.52%14.20%
Opera1.80%1.47%5.45%1.16%2.40%
That’s hard to understand so let’s throw out some graphs. The one which caught my eye right away was Net Market Share. They show an overwhelming lead for Internet Explorer

net market share

On the other hand everybody else shows a lead, of various degrees, for Chrome

w3 counter

global stats
wikipedia

w3schoolsThe divergence is because of the different methods used to get the data. For instance Wikipedia and W3 schools look only at the statistics on their site. Because both of them are used by people with a fair degree of technical ability they reflect a higher degree of usage by Chrome. The interesting ones are the first three, W3 counter, Global Stats Counter and Net Market Share. They are all aggregators of a large number of sites. I’m socked to see such a high degree of variability. Each of these sources use millions of page views to gather their information so a variation of more than a couple of points seems unusual.

It feels like the take-away here is that the browser usage statistics are garbage. As an industry we’re totally failing to measure the most basic of statistics about how people interact with the Internet as a whole. We should be ashamed of ourselves and we should do something about it. In the meantime it seems like we’re going to have to continue to support at least 3 possibly 4 different browsers to say nothing of the various versions of the browsers. The trends don’t reveal anything of any use either. There seems to be some momentum behind Chrome and IE and less for Firefox, Opera and Safari – but who really knows. The Internet is not homogeneous so we see different browser statistics when we slice our data geographically and topically. I bet the usage statistics on hacker news are interesting.

Unfortunately all of this means that you’re going to have to look at the statistics on your own website to see which browsers should be concentrated upon. I hate supporting old browsers but if your market is 6 guys sitting in their Unix holes* using Lynx then you’re supporting Lynx. Best of luck to you!

*Unix hole – it is a thing, trust me.