Some stats on OS and browser share

Warning This article was written over six months ago, and may contain outdated information.

Last week on Twit­ter I shared some brows­er and OS sta­tis­tics from a site I man­age. These turned out to be quite pop­u­lar, so I’ve decid­ed to expand on them a lit­tle fur­ther, and also add the stats from anoth­er site I man­age, to broad­en the base num­bers a lit­tle. I’m not try­ing to make any point here, just shar­ing a lit­tle bit of ana­lyt­ics data. If there’s any inter­est in my doing so, I’ll pro­vide fur­ther updates in the future; leave a com­ment if there’s any­thing in par­tic­u­lar you’d like to know.

Reporting notes

Stats are tak­en from Google Ana­lyt­ics from the peri­od 1st Jan­u­ary 2013 to 26th April 2013 (it took me a few days to write this post). I’m going to give the mean aver­age of stats between the two sites, not­ing where there is sig­nif­i­cant vari­a­tion — and I’ll spec­u­late on the vari­a­tion at the end of the post.

I decid­ed not to use the stats from this blog as it tends to skew towards a very non-typ­i­cal audi­ence, con­sist­ing in large part of the tech-savvy and ear­ly adopters. Instead I’m using two sites I man­age which I think have a more ‘nor­mal’ set of vis­i­tors (no offence intend­ed, dear reader).

To keep the sites anony­mous I’ll refer to them as Site A and Site B. Site A is for news and opin­ions on Arse­nal foot­ball club. It’s Lon­don-cen­tric but has an inter­na­tion­al audi­ence, and had an aver­age of 1,935 unique vis­i­tors per day in the select­ed peri­od. Site B is a hyper-local blog for a Lon­don neigh­bour­hood, and its audi­ence reflects that. It had an aver­age of 120 unique vis­i­tors per day. Both sites are updat­ed semi-fre­quent­ly but irregularly.

Selected stats

The per­cent­age split between mobile (includ­ing tablet) and non-mobile is 31.2 to 68.9. Of those mobile vis­i­tors, iOS account­ed for 68.4 per­cent and Android for 23.3. Black­ber­ry was a dis­tant third, with oth­er mobile OSes bare­ly registering.

Pie charts showing Mobile/non-mobile visitors and popular mobile OS

In over­all OS stats, Win­dows is still by far the dom­i­nant plat­form, with more than dou­ble the vis­its of the sec­ond-placed iOS. Some strong vari­a­tion to note here: Site A had 12.2 per­cent of vis­i­tors using OS X, with Site B on 27.1.

Most pop­u­lar OS
Oper­at­ing system % of total
Win­dows 44.7
iOS 21.2
OS X 19.7
Android 7.3

In over­all brows­er stats, Safari is the per­haps sur­pris­ing leader, thanks to the large share that iOS holds. Some very strong vari­ance to note: Site A actu­al­ly had Opera Mini in fourth place with 7 per­cent share, beat­ing Android into sixth, while on Site B there were no hits at all from Opera Mini.

Most pop­u­lar browser
Brows­er % of total
Safari 34.1
Chrome 21.6
IE 20.2
Fire­fox 12.3
Android (stock) 5

Break­ing Inter­net Explor­er into the dif­fer­ent ver­sions, IE9 is the most pop­u­lar fol­lowed close­ly by IE8. Very grat­i­fy­ing­ly, IE6 is bare­ly exis­tent and soon to be gone alto­geth­er. There was a point of strong vari­ance here: Site A had IE8 on 42.1 per­cent of the IE share and IE9 on 41.6, while Site B had the same ver­sions on 35.3 and 50.3 per­cent respectively.

Inter­net Explor­er versions
IE version % of IE share % of all browsers
IE9 46 9.3
IE8 38.7 7.8
IE7 8.5 1.7
IE10 6 1.2
IE6 1 0.2

Final­ly, in my orig­i­nal tweets I not­ed that over 98 per­cent of vis­i­tors run­ning iOS devices used their phone in por­trait mode, but this is incor­rect; in fact Safari for iOS only shows dimen­sions for the por­tait mode, even when in land­scape, so there’s no way to get this fig­ure. Android devices don’t fol­low the same pat­tern, how­ev­er, so there we can see that an aver­age of 92 per­cent of vis­its were made using a device in por­trait mode. I can’t be sure where the false iOS fig­ures came from; per­haps oth­er browsers on iOS, or jail­bro­ken devices.

Speculation on variance

There are a num­ber of points of strong vari­a­tion: in vis­i­tors using OS X; in the use of Opera Mini; and in IE8 vs IE9. My broad the­o­ry to explain this is that Site A has a glob­al audi­ence, includ­ing a not insignif­i­cant num­ber in emerg­ing mar­kets across Africa and Asia, where­as Site B is almost exclu­sive­ly vis­it­ed by (com­par­a­tive­ly) wealthy Lon­don­ers, which explains the high­er num­ber of Macs in Site B, and the use of Opera Mini on Site A (from areas with poor­er net­work cov­er­age and a high­er inci­dence of low-end hand­sets). As for IE, I’d spec­u­late that you need a gen­uine copy of Win­dows 7 to run it, and I believe that Site A’s glob­al pop­u­lar­i­ty would mean that it’s used on pirat­ed Win­dows which could­n’t install it.

Of course, this is all spec­u­la­tion, and with the lim­it­ed sam­ple base I have, espe­cial­ly on Site B, it could just be an aber­ra­tion in the data. As always, it pays to use many dif­fer­ent data sets.

1 comment on
“Some stats on OS and browser share”

  1. Could you also list Safari ver­sion num­bers? I find that I’m hav­ing more and more prob­lems with fea­tures that old ver­sions of Safari can­not han­dle (it is becom­ing a prob­lem at an alarm­ing rate, big­ger than Inter­net Explor­er ever was due to its dom­i­nance on mobile devices). I’ve seen a lot of users still on Safari 5.1 due to old­er iphones or jail­bro­ken ones.