Re: [Wikitech-l] [Analytics] Readership metrics for the timespan until December 4, 2016

[Olga Vasileva]

Thanks Zareen, it's great to see so much in-depth data!  Also, it answers
so many questions and curiosities from the past few months.

- Olga

On Wed, Dec 14, 2016 at 1:15 PM Caitlin Cogdill 
wrote:

> Yes, thanks so much for sharing this! It's fascinating and encouraging to
> see the November impressions bump.
>
> On Wed, Dec 14, 2016 at 11:12 AM, Toby Negrin 
> wrote:
>
> Thank you so much for this Zareen! It's really great to see this report --
> so much interesting data to think about!
>
> -Toby
>
> On Mon, Dec 12, 2016 at 5:45 PM, Zareen Farooqui 
> wrote:
>
> Hi all,
>
> This resumes the usual look
>  at
> our most important readership metrics. This time we can report that daily
> pageviews are up 4.8% (since the last report), with an interesting recent
> peak which meant that November’s pageviews surpassed those of November 2015
> (coincidentally also) by 4.8%, after October had already seen a 2.1%
> year-over-year increase. The iOS Wikipedia app saw increased downloads,
> while the Android app’s install base has stopped its previous downward
> trend.
>
> As laid out earlier
> ,
> the main purpose is to raise awareness about how these metrics are
> developing, call out the impact of any unusual events, and facilitate
> thinking about core metrics in general. As always; feedback and discussion
> welcome. Week-over-week and month-over-month changes are now being recorded on
> the Product page
>  at
> MediaWiki.org. This edition of the report covers a timespan of eighteen
> weeks.
>
> Some other recent items of interest, in case they didn’t already catch
> your attention:
>
>-
>
>The WMF Reading team published its quarterly review presentation
>
> 
>for Q1 2016-17 (July-September), which includes lots of traffic and usage
>data.
>-
>
>At the Foundation’s August metrics meeting
>
> ,
>the Reading team gave an update on longer-term traffic trends since 2013.
>(TL;DR: Overall pageviews have been flat to slightly declining, mobile has
>been steadily rising but recently slowed down, desktop has declining during
>these three years. However, total pageviews have been slightly increasing
>year-over-year in the last few months.) See the chart below, updated with
>data until November:
>
> [image: Wikimedia monthly pageviews (desktop+mobile), 2013-2016 (version
> December 2016).png]
>
> In particular, as mentioned, the number of total pageview saw
> year-over-year increases of +2.1% for October and +4.8% in November, in
> contrast to e.g. the -10.5% we had for May 2015-May 2016.
>
> Now to the usual data. (All numbers below are averages for August
> 1-December 4, 2016 unless otherwise noted.)
>
> Pageviews
>
> Total: 529 million/day (+4.76% from the previous report timeframe, with
> corrected numbers for anomalously high traffic on some main pages
> )
>
>
> Context (April 2015-December 2016):
>
>
>
> See also the Vital Signs dashboard
> 
>
> (Small caveats: iOS app’s pageviews were undercounted by about 1.6
> million/day from mid September to early November due to a bug
> .)
>
> Overall pageviews increased steadily during the timespan of this report
> aside from the week ending October 30th (right before Halloween) and the
> end of November. There appears to be a peak in pageviews in November.
>
> To facilitate our understanding of which traffic movements are seasonal
> and which may indicate lasting changes, here is a chart overlaying the
> total pageview numbers back to May 2013 (the earliest time for which we
> have data according to the current pageview definition):
>
> The blue line indicates a non-seasonal rise peaking around November 12. We
> checked whether this peak came from a particular country and were able to
> exclude that possibility.
>
> Wikimedia Daily Pageviews from US
>
>
> Pageviews in US do not show any drastic changes (even around the time of
> the US elections).
>
>
> Wikimedia Daily Pageviews from Mexico
>
>
> In Mexico, there seems to been a huge drop starting October 27th (perhaps
> some sort of local outage).
>
>
> Wikimedia Daily Pageviews from Ecuador
>
>
> Ecuador shows a huge spike on October 30th, followed by a drop for several
> days.
>
> Desktop: 54.1% ​(previous report: ​54.1%)
>
> Mobile web: 44.8% ​(previous report: 44.6%)
>
> Apps: 1.1% ​(previous report: ​1.3%) (missing some iOS pageviews, cf.
> 

Re: [Wikitech-l] [Analytics] Readership metrics for the timespan until December 4, 2016

[Toby Negrin]

Thank you so much for this Zareen! It's really great to see this report --
so much interesting data to think about!

-Toby

On Mon, Dec 12, 2016 at 5:45 PM, Zareen Farooqui  wrote:

> Hi all,
>
> This resumes the usual look
>  at
> our most important readership metrics. This time we can report that daily
> pageviews are up 4.8% (since the last report), with an interesting recent
> peak which meant that November’s pageviews surpassed those of November 2015
> (coincidentally also) by 4.8%, after October had already seen a 2.1%
> year-over-year increase. The iOS Wikipedia app saw increased downloads,
> while the Android app’s install base has stopped its previous downward
> trend.
>
> As laid out earlier
> ,
> the main purpose is to raise awareness about how these metrics are
> developing, call out the impact of any unusual events, and facilitate
> thinking about core metrics in general. As always; feedback and discussion
> welcome. Week-over-week and month-over-month changes are now being recorded on
> the Product page
>  at
> MediaWiki.org. This edition of the report covers a timespan of eighteen
> weeks.
>
> Some other recent items of interest, in case they didn’t already catch
> your attention:
>
>-
>
>The WMF Reading team published its quarterly review presentation
>
> 
>for Q1 2016-17 (July-September), which includes lots of traffic and usage
>data.
>-
>
>At the Foundation’s August metrics meeting
>
> ,
>the Reading team gave an update on longer-term traffic trends since 2013.
>(TL;DR: Overall pageviews have been flat to slightly declining, mobile has
>been steadily rising but recently slowed down, desktop has declining during
>these three years. However, total pageviews have been slightly increasing
>year-over-year in the last few months.) See the chart below, updated with
>data until November:
>
> [image: Wikimedia monthly pageviews (desktop+mobile), 2013-2016 (version
> December 2016).png]
>
> In particular, as mentioned, the number of total pageview saw
> year-over-year increases of +2.1% for October and +4.8% in November, in
> contrast to e.g. the -10.5% we had for May 2015-May 2016.
>
> Now to the usual data. (All numbers below are averages for August
> 1-December 4, 2016 unless otherwise noted.)
>
> Pageviews
>
> Total: 529 million/day (+4.76% from the previous report timeframe, with
> corrected numbers for anomalously high traffic on some main pages
> )
>
>
> Context (April 2015-December 2016):
>
>
>
> See also the Vital Signs dashboard
> 
>
> (Small caveats: iOS app’s pageviews were undercounted by about 1.6
> million/day from mid September to early November due to a bug
> .)
>
> Overall pageviews increased steadily during the timespan of this report
> aside from the week ending October 30th (right before Halloween) and the
> end of November. There appears to be a peak in pageviews in November.
>
> To facilitate our understanding of which traffic movements are seasonal
> and which may indicate lasting changes, here is a chart overlaying the
> total pageview numbers back to May 2013 (the earliest time for which we
> have data according to the current pageview definition):
>
> The blue line indicates a non-seasonal rise peaking around November 12. We
> checked whether this peak came from a particular country and were able to
> exclude that possibility.
>
> Wikimedia Daily Pageviews from US
>
>
> Pageviews in US do not show any drastic changes (even around the time of
> the US elections).
>
>
> Wikimedia Daily Pageviews from Mexico
>
>
> In Mexico, there seems to been a huge drop starting October 27th (perhaps
> some sort of local outage).
>
>
> Wikimedia Daily Pageviews from Ecuador
>
>
> Ecuador shows a huge spike on October 30th, followed by a drop for several
> days.
>
> Desktop: 54.1% ​(previous report: ​54.1%)
>
> Mobile web: 44.8% ​(previous report: 44.6%)
>
> Apps: 1.1% ​(previous report: ​1.3%) (missing some iOS pageviews, cf.
> above)
>
> Context (December 2015 - December 2016):
>
> Overall mobile percentage is similar to the last report, but we did see a
> small increase (besides one week in August) until late September.  As a
> reminder, mobile already has a solid majority in terms of unique devices,
> cf. below.
>
>
>
> Global North ratio: 75.3% of total pageviews (previous report: 75.5%)
>
> Context (January - December 2016):
>
>
>
> There is a slight