> On Jul 17, 2014, at 6:33 PM, Howie Fung <[email protected]> wrote:
> 
> Thanks for putting this together -- this is really interesting data.  It's 
> interesting that the 5+ activation rate for mobile web > desktop now -- just 
> a few months ago, mobile was about half of desktop.

Yeah, the jump happened in May, and we still don't know why. My theory is 
lastmodified, but it's a bit of a stretch for 5+.

> 
> It could be the case that mobile web editors make a smaller number of small 
> edits/spread a big edit out into smaller ones -- would be interesting at some 
> point to see whether these initial edits happen in rapid succession.  

I can play around with some visualizations for that :)

> It'll also be interesting to compare the (short-term) retention rates of 
> mobile vs. desktop.
> 
> The Activation rate for apps is weird.  We know on desktop that about 70% of 
> users create an account, but never edit -- this would imply that on mobile 
> apps a far higher percentage of users create accounts for reasons unrelated 
> to editing.  We should look a little closer to the data .  Something doesn't 
> seem right.

If it's weird, it's at least consistently weird -- I saw the same thing (a ~1% 
1+ conversion rate) in the early post-release data. This is coming from 
Wikimetrics, which is pretty reliable. But yeah, let's get an independent third 
party to verify...

In the meantime, some wild speculation: maybe people really are used to signing 
up on all apps as part of first time ux without necessarily getting anything 
out of it, more so than on other platforms.

> 
> Howie
> 
> 
> 
>> On Thu, Jul 17, 2014 at 5:40 PM, Maryana Pinchuk <[email protected]> 
>> wrote:
>> Greetings all, and happy 1 month anniversary of the tablet redirect :)
>> 
>> I've got some data to share on new tablet, phone, and app editors and how 
>> they stack up against new desktop editors. This is enwiki only for now; I 
>> await the glorious return of Dario to help us out with global data :)
>> 
>> New active mobile editors
>> 
>> Our metric for new active mobile editors is 5+ edits within 30 days of 
>> registration, so we don't quite have the full numbers on the first month of 
>> newly registered users yet (because people who signed up yesterday haven't 
>> had time to reach active status). But here's today's snapshot of the percent 
>> of newly registered users we've converted to active editors in the last 30 
>> days across all out different platforms:
>> 
>> <image (21).png>
>> Note that the trend Dario observed still holds – both newly registered 
>> tablet and mobile site users are being converted to 5+ editors at a higher 
>> rate than desktop. 
>> 
>> Slightly concerning is the conversion rate on the Android app – I'd have 
>> expected this to be higher, more in line with mobile site editors on phones. 
>> We'll need to figure out if this is due to users editing anonymously on the 
>> app instead of signing in and counting toward our new active user count.
>> 
>> New user productivity
>> 
>> I was also interested in the volume of contribution we're seeing across each 
>> of the platforms, to get a sense of how much new folks are contributing. So 
>> I ran the numbers for bytes changed (added, removed, and total changed) by 
>> newly registered users across all our different platforms:
>> 
>> <image (20).png>
>> Though mobile site editors (both tablet and phone) are reaching 5+ edits at 
>> a higher rate than desktop users, it looks like desktop editors make 
>> significantly bigger edits in terms of bytes. Not super surprising, given 
>> the smaller screen real-estate and greater difficulty in contributing 
>> longform text, but it's good to validate that with real data. Note again 
>> that Android app editors are on the lower end of bytes contributed.
>> 
>> To me, these two graphs raise the question of whether mobile site editors 
>> (on both tablets & phones) make a larger number of small edits as opposed to 
>> one big edit like they might on desktop. I don't know if we'll be able to 
>> answer that definitively anytime soon, but it's definitely something to 
>> think about and dig into more in the future...
>> 
>> -- 
>> Maryana Pinchuk
>> Product Manager, Wikimedia Foundation
>> wikimediafoundation.org
> 
> 
> 
> -- 
> _______________________
> Howie Fung
> Director of Product Development
> Wikimedia Foundation
> 
> [email protected]
> 
_______________________________________________
Mobile-l mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/mobile-l

Reply via email to