Page MenuHomePhabricator

Google Search Console access for Search Platform team
Closed, InvalidPublic

Description

Being able to look at search queries and stats broken by query – especially for different languages – would be super useful for that team and the work they do.

Event Timeline

Restricted Application added a subscriber: Aklapper. · View Herald Transcript
EBjune triaged this task as Medium priority.Mar 1 2018, 6:16 PM

I'm providing a proactive approval, if that's what's needed, and I can certainly echo the usefulness for the Search Platform team.

Just to confirm the list of names to bring up in the next SRE meeting; This means the access is requested for Erika Bjune, Erik Bernhardson, David Causse, Trey Jones and Stas Malyshev, right? (Guillaume already has access to the password)

Yes @MoritzMuehlenhoff that is the correct list of names, thanks!

Unfortunately we've had an issue with managing access to the admin account for the Google search console for a long time. Currently the only option is sharing the single password to this account with multiple people, which besides the obvious security problems this entails, also causes Google to detect abuse all the time.

We can delegate access to individual Google accounts, but unfortunately only on a per-subdomain-basis. And we have many hundreds of those, each needing to be done manually. Clickety clickety click, clickety clickety click. Not really feasible to do for all our subdomains in practice. :(

Is it an option for your purposes to get access to only a few sub domains, so we can actually delegate a few domains those to a few accounts?

Dzahn changed the task status from Open to Stalled.Apr 17 2018, 5:40 PM

Is it an option for your purposes to get access to only a few sub domains, so we can actually delegate a few domains those to a few accounts?

Yes...and also probably no. The Yes part would be to request the few big ones that we work the most with (enwiki, commons, wikidata, etc...), but then things get a bit more complicated beyond that when we're doing things like testing Serbian language analyzers and we'd like to see how different ones compare against different sets of queries. In that case, it might be helpful to include data from various sub domains that use Serbian or a variant when doing analyses (and please weigh in @mpopov).

While it would be very helpful, this is not a huge priority for us, and I certainly don't want to add unnecessary administrative burdens, but if we can get access to a small pool of sub domains, that would be great.

I recently had to add another user to google search console, so I'll share my findings.

Please note directions on how to add/remove users from google search console have been added to wikitech: https://wikitech.wikimedia.org/wiki/Google_Search_Console_access

Also note that each project tends to have up to 4 entries: http/https and the m and non m versions. examples: http://en.wikipedia.org https://en.wikipedia.org http://en.m.wikipedia.org https://en.m.wikipedia.org

Please keep that in mind when asking to add users to sub-domains, as each sub-domain is actually 4 different sub-domain entries on the google search console.

@EBjune: This seems to be pending your input on which sub-domains are needed. Your last answer seems to indicate a few projects, but is a bit too vague for an actual checklist of what to add. Can you list out every project you want these accounts to have access do? Please include if they need the http version, and if they need the m versions (http and https).

Once we have a full listing of domains, the scope of the work will be more readily apparent and can be tasked.

@EBjune @RobH: BTW @JKatzWMF and I are going to be changing which properties are tracked in GSC (namely getting rid of all but 1-2 HTTP variants in favor of HTTPS) as well as which project classes and languages are tracked. We expect to be done with this "sprint cleaning" endeavor sometime in the next two weeks so I'll post an update here when it's ready.

herron subscribed.

Since this task has been inactionable in the access request queue for some time I'm going to remove the SRE-Access-Requests tag. Please re-add project tag SRE-Access-Requests when ready to proceed. Thanks!

Aklapper changed the task status from Stalled to Open.Nov 27 2020, 12:59 PM

T132373#2201301 states that Discovery / (Search Platform) already "has its own account for Google Webmaster Tools [...] searchteam+gwt@wikimedia.org".
I have no idea if that's the same as "Google Search Console", but all links on https://wikitech.wikimedia.org/wiki/Google_Search_Console_access go to https://www.google.com/webmasters/tools so I imagine that it is.

Thus I guess Search Platform could use Google Search Console if wanted. If I'm correct, then please close this task which has been lingering for 2.5 years. Thanks! :)