Looker Connect Training

Help Center

Documentation

Community

Cloud Certifications

Question

SEO dashboard for multilanguage website

  • 31 January 2023
  • 3 replies
  • 40 views
Flo

Hello. I want to build a dashboard on looker to keep track of my SEO impressions, users, (etc.) by languages. My website is built like this:

xxx.com/ => FR urls

xxx.com/en => EN urls (the homepage is .com/en and the other pages are .com/en/...)

I would like to know what kind of filters or whatever should I use to create separate dashboards based on the URL structure (.com/ and .com/en), using different types of data (GSC, Analytics). I am sure there are formulas available however I can’t find them

For instance if I want to create a dashboard to keep track of SEO impressions I will select:

  • data source: search console
  • dimension: year month
  • metric: for instance “impressions”

And then create separate dashboards:

  • one filtered on URLs starting with “xxx.com/” without the URLs starting with “xxx.com/en”
  • one filtered on URLs starting with “xxx.com/en”

Thanks

3 replies

jcdufault
Userlevel 1

Hello Flo, 

I think the easiest way to do that would be to create a new dimension to track if you url refers to an english page or not. In LookML it would look like this:

dimension: url_language {
    label: "Page URL Language"
    type: string
    sql: CASE WHEN ${url} ilike '%/en/%'
              THEN 'English' ELSE 'French' END ;;
  }

With ${url} referring to your URL dimension already declared in your view file. Then you can use this dimension as a filter to select one or both languages or use it as a dimension to group your measures by.

 

Hope this helps!

 

Jonathan

Flo

Thanks a lot, it helps but unfortunately I cannot find by what I have to replace url in ${url}. I have connected GSC, I have data because I have a dashboard that shows the impressions from the entire website, however I cannot validate the formula. These are the data I have access to on GSC.

 

jcdufault
Userlevel 1

Sorry this looks like it’s from Looker Studio and not Looker. My best guess would be to leverage the landing page (if it’s a URL) but I can’t speak to Looker Studio’s functionalities to extract or parse a string. There is a separate community forum for Looker Studio and might get more luck there: https://support.google.com/looker-studio/community?hl=en

Reply


Terms of Service