CEO tears into recent Flurry research on apps versus web to celebrate.
Smartphone browser Skyfire has now been downloaded more than 7.5 million times across iOS and Android, the company has announced.
That's a mixture of paid and free downloads, since the app has veered between £1.19 and £1.79 on iPhone and between £2.39 and £2.99 on iPad, but has been a free download on Android.
Skyfire says it has been making hay on the App Store and Android Market partly due to carriers ditching their unlimited data tariffs, and thus making their customers more conscious of their data usage. Skyfire claims it can compress web video by up to 75%, although the very fact that it can play Flash videos was a big selling point on iOS, given Apple's decision not to support Adobe's technology.
In separate news, Skyfire CEO Jeffrey Glueck has published an angry blog post criticising mobile analytics firm Flurry's recent claims that people are spending more time using mobile apps than using the web.
Glueck points out that Flurry's data compared comScore data for the entire US population with its own metrics from just mobile app users.
"Statistics 101. All the light or non-smartphone users in the ComScore data drag down the web per user minutes average. Comparing that to Flurry’s app devotee population is just sleight of hand from there," he writes. "It’s comparing apples and oranges. It's like conducting a survey at a comic book convention and announcing that time spent on comic book reading has passed all other books combined... It's rigged."
Glueck also criticises coverage of Flurry's research that has suggested it shows apps are 'winning' against HTML5.
"Flurry's definition of Apps can include html5-written, browser-based Apps, like the ones that Facebook's web-based Project Spartan would merchandise. So App growth does not mean a zero-sum loss for html5," he writes.
"I have moderated several panels in the last year on html5 vs native apps, and on the last panel, both CNN digital and Condenast Digital admitted that the page views they get on their mobile web site is an order of magnitude more than the page views on their apps (in part because of the feature phone traffic)."
Advertisement
Add a new comment
You need to be logged in to post comments. If you do not have an account then please register.
Comments
1 comment
It is great to see the comments from Jeffrey Glueck on the recent Flurry report. One reason I believe the original Flurry report has seen such uptake within mobile circles is because it is so obviously flawed. But how could a company as successful as Flurry produce such dubious data? I suspect the main problem is one of company focus. Naturally as a company only focused on apps it is in their interest to encourage the development of many more, rather than the creation of HTML 5 widgets and other web content. In addition, while they have a large percentage of the worlds mobile apps instrumented with their app analytics, they do not and cannot measure usage of web sites - as Jeffrey points out, they defaulted to comScore survey data and thus measured apples with elephants.
And even if Flurry did have mobile web measurement, recording accurate use of web sites by mobile devices is significantly more complex than measuring mobile app use, where access to a Unique Device ID (UDID) is easily available. For mobile web the only way to gain any form of accuracy is through a combination of smart server side technology and partnerships with carriers, device manufacturers and browser developers. Without this many of the mobile devices browsing today are simply not counted and easily contribute to artificially low web numbers.
It is very easy to show how inaccurate your current analytics provider is at measuring your mobile visitors by grabbing a free trial of Bango Analytics from http://bango.com/analytics and adding the small piece of tracking code to your website alongside your current provider. Bango provides analytics for both web and apps for some of the worlds largest brands, especially those that need to retain clear ownership of the data they record from their apps and websites. The results we see are very different to that of Flurry and we will share those with everyone shortly.
Andrew Bovingdon Jun 28th 2011 at 11:32AM
0 0