just one more geek in a sea of austin techies

October 24, 2012

Web Stats and Your Target Audience #AnalyticsGeek

General web stats don't tell 
the whole story and can 
sometimes severely mislead.
Web browsing stats are a regular news item: mobile device browsing gaining heavily on desktop browsing, IE use still outnumbers Firefox use, etc. We hear the general trends but rarely the fine details. IE users continuing to outnumber Firefox users? Not within tech circles. Chrome now more popular than either IE or Firefox? Not even close for web-based content related to health (as we'll see).

Because generalizations don't tell the whole story I regularly review statistics for sites I create and maintain, both business and personal. In that vein I thought I'd share some "targeted audience" stats and highlight a few trends...

Don't Make Decisions based on Generalizations
If you're planning how to best spend effort and dollars, don't simply Google up the latest browsing trends and call it a day. You need to dig down and find the trends within the specific population(s) you're targeting.  Take a look at the following "30-day Web Browser Use" comparison table:

30-day Web Browser Use ending 10/1/2012
General Web
(from StatCounter)
(public-funded US site)
30% Google Chrome 66% Internet Explorer
29% Internet Explorer 12% Google Chrome
20% Firefox 9% Safari
10% Safari 8% FireFox
4% Opera 2% Android
3% Android 1% Opera
If you follow general web browser news you'll probably recall what was widely-reported earlier this year: Google Chrome use overtook Internet Explorer use.  Months later the trend has held: Chrome use still outnumbers IE use.

This suggests that IE is on its way out and web developers should begin to target Chrome compatibility first and IE second. That generalization, however, doesn't take into account specific target audiences. Are you marketing towards workers within the enterprise or state or federal agencies? Many of those workers are still using IE because it's the browser already installed and/or it's the only browser allowed by their IT departments. The more sensitive the local network, the more likely that IE is still the browser of choice (due to convention, not better browser security).

From the comparison table above we see Internet Explorer is still far in the lead among visitors to a certain publicly-funded health-related website. The website is the launching pad for a number of other health websites so I believe it's a good general indicator of US health-related web traffic.

Many of the visitors to this health site work in hospitals which remain a stronghold of IE as those networks and the software in use is tightly regulated -- changes, even for security improvements, are not introduced unless absolutely necessary. So how much do hospital-related visitors skew the stats in favor of IE? Well I just happen to have the numbers from a site strictly devoted to use by hospital employees:
30-day Web Browser Use ending 10/1/2012
General Web
(from StatCounter)
(public-funded US site)
30% Google Chrome 79% Internet Explorer
29% Internet Explorer 6% Google Chrome
20% Firefox 5% Safari
10% Safari 5% FireFox
4% Opera 1% Opera
3% Android

Wow. 79% market share for Internet Explorer among hospital employees. It's been more than 5 years since IE had that kind of share in the general population (StatCounter data only goes back to June 30, 2008). Apple's share (desktop and mobile devices combined) is half the norm and Android hasn't even peeked its nose into regular hospital use.

With so very many visitors still using Internet Explorer, the inevitable next question is "which versions of IE are most used?" Here's the breakdown of that 79% versus the general population's use of IE:
IE Version Breakdown
General Web
(from StatCounter)
(public-funded US site)
56% IE 9 50% IE 8
41% IE 8 26% IE 7
3% IE 7 23% IE 9

1% IE 6

We see again what we already know: with regards to hospital environments, change is slow. The general population has all but dropped IE7 yet IE7 still accounts for a healthy 26% of web browsers in use by hospital employees. IE6 also still eeks out an existence somewhere out there though I'm not sure I'd care to seek treatment at a facility relying on web browsing technology from 2001 that has known security issues and which the browser developer is actively attempting "kill" the version and completely remove it from use (Microsoft estimates 6% use of IE6 worldwide compared to StatCounter's < 1% estimation).

What It All Means
In these examples we find generalizations about browser usage are way off base for the given target audience. Know your target audience and use data that's relevant. Even if you can't lay your hands on data specific to your audience it helps to know that generalizations don't tell the whole story. If you don't have good intel when launching your web project then it's extra-critical to collect good stats on your own and be prepared to act on them over time.

No comments:

Post a Comment