First let me say I'm a huge fan of Business Insider. But today they published a graphic and failed to provided appropriate context which I feel guides readers to the inappropriate conclusion. I would like to set the record straight.
The article I took issue with is here: www.businessinsider.com/chart-of-the-day-microsofts-windows-compared-to-ios-and-android-2012-2
At a glance it concludes "Microsoft isn't going anywhere", alluding to the fact that mobile isn't even making a dent.
It made me scratch my head for a minute then I got really mad and felt like somebody was trying to pull a fast one. I hate when people play fast and loose with the data.
The graphs appear as two equal sized pie charts - but the pies aren't equal sized in reality. So the pie charts themselves should have been corrected for size of sessions.
Next the graphs should have pointed out that app usage is not counted as "digital traffic", it isn't measured as traffic at all. So that graph only counts mobile vs. desktop browser usage.
More and more things done on tablets or mobile are done inside apps, and more apps are being created everyday and those are actively stealing traffic from mobile browsers, and desktop apps - it's just wrong.
Just like most smartphone users - I find fewer and fewer reasons to use the browser on my phone each day. I'm frankly amazed that windows isn't gaining market share as a % on that graph. Surfing the web on a mobile device requires patience, and apps don't.
So it makes sense that popular items like Yelp are being accessed more and more frequently via mobile apps, than mobile web.
So all those graphs really tell me is that % of people leaving the desktop web to go mobile web, is roughly equivalent to the % of people leaving the mobile web to the mobile apps.
What should have been done - was compare the volume of sessions between mobile and desktop, and then adjust for the # of new mobile browsers. This would show that more people are using their phones for less mobile web browsing everyday -- and of course this could only be attributed to the transition to apps and RIA's which can't be tracked like standard web views.
Pay attention - user behavior is shifting.