User blog:MisterWoodhouse/An update on Fandom and Gamepedia domain migrations

Hey gang!

I wanted to give you an update on how the Gamepedia to Fandom domain migration testing has gone and what our next steps are for this work.

Since the initial migration of wikis in the test group back in November, we have been monitoring the effect of the migration on inbound traffic to these wikis from Google. As a reminder for context, the initial migration set included wikis where Project Crossover had largely been completed and the affected wikis for archival with -archive.fandom.com had little traffic.

In monitoring the moves in the first five weeks, we noticed that established, active Gamepedia wikis moving to Fandom URLs held their traffic without much deviation, which is a very good sign for the initial move timeline. When we moved from Wikia URLs to Fandom URLs, we saw an initial drop followed by a recovery period, but that is not the case this time. A really nice finding among all of this is that less established wikis saw substantial increases in traffic as a result of the domain migration, so much so that they skewed the average percent change dramatically, to the tune of +74% by the fifth week.

We also saw that the Google algorithm change released during the first week of monitoring and fully deployed by the end of the third week was beneficial for the Fandom domain, with a positive trend of all aspects of the migration test data.

The one downside we observed was that wikis holding prime Fandom URLs that migrated to new archive URLs to make way for their more established Gamepedia counterparts lost significant traffic. This is a phenomenon that we are seeking to correct with a new process for running migrations with domain conflicts.

Starting this week, we are going to do a second test to see how we can retain as much traffic as possible for Fandom wikis moving into an archive state. Rather than make both URL changes simultaneously, we are going to do a redirect from the prime Fandom URL to the archive Fandom URL for a set duration and then move the Gamepedia URL to the prime Fandom URL. This will allow us to see if we can “teach” the Google algorithm that the Fandom content has moved to a new home and adjust accordingly before we move the Gamepedia URL.

The New Process:


 * 1) Fandom wiki gives up the prime Fandom URL with a redirect to the new archive URL
 * 2) A set amount of time passes to let Google adjust
 * 3) The Gamepedia wiki moves to the prime Fandom URL

As we said before, the inner workings of Google algorithm is a closely guarded secret and we understand the effects of changes we make by testing. We are quite pleased with how the active Gamepedia wikis reacted to the domain migration, as well as how the December algorithm change affected us positively across the board, but refining the approach will allow us to be more confident in our process and the results.

Please note that we are aware of the Community Central bug preventing you from making blog comments and we are working to resolve it.

pt:Blog_de_usuário:Eduaddad/Uma_atualização_sobre_as_migrações_de_domínio_Fandom_e_Gamepedia