On Mon, Feb 10, 2014 at 2:53 PM, Steven Walling swalling@wikimedia.org wrote:
On Mon, Feb 10, 2014 at 2:35 PM, Dario Taraborelli dtaraborelli@wikimedia.org wrote:
Given that background research on article creation is not limited to data analysis or testing for features developed within Growth it's important that this reaches the latter type of audience and I believe a little bit of redundancy doesn't hurt.
Yeah I don't mind redundancy on principle, however, I do want us to be very careful about viewing R&D as separate from our actual team work. For instance, we might learn things about article creation in general that are of broad interest, but we did it in order to provide a solid background for product changes by Growth. The primary purpose of the analysis is informing change, with educating the community or organization at large a tertiary goal.
If there is analysis work, such as traffic analysis, that doesn't fit within a single team's scope but impacts the entire organization... well obviously that makes sense to put in an overall Analytics report.
Can't comment on the specific issue being discussed here, but it may be worth pointing out that the monthly engineering reports have contained an Analytics section for quite a while. (For those who are not familiar with these reports, they can be found at e.g. https://blog.wikimedia.org/c/technology/wmf-engineering-reports/ . They are usually published at the day of the metrics meeting; and I believe Dario is working to fill out the gap at https://www.mediawiki.org/wiki/Wikimedia_engineering_report/2014/January#Ana... .)
Like the general monthly WMF reports, the engineering reports are generally organized along team / department lines. And it's quite frequent that inter-team work is reported both in the section of the client team and of the supporting team (e.g. if, say, the Wikipedia Zero team launches a new partnership and we in the Communications team put out a press release about that, then it's going to be mentioned in both sections).
But just like the UX team does not have a separate report, but instead reports what it does through and for our cross-functional teams, I feel very strongly that reporting about analysis work should be done through normal team reporting whenever possible. This keeps our analysis conceptually tied to measuring outcomes and trends relevant to product work, rather than just generalized R&D.
-- Steven Walling, Product Manager https://wikimediafoundation.org/
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics