fivetran / dbt_google_ads

Fivetran data transformations for Google Ads built using dbt.
https://fivetran.github.io/dbt_google_ads/
Apache License 2.0
15 stars 29 forks source link

Retain Campaign History for Joining to CRM Data #45

Open SamtheDeveloper91 opened 1 year ago

SamtheDeveloper91 commented 1 year ago

Is there an existing feature request for this?

Describe the Feature

In some of the models such as the campaign_report, the CTEs are filtered for is_most_recent_record = True. This means that each campaign name reflects the most up-to-date name, which may not match the name on a lead, contact, etc from a CRM like Salesforce. This makes it more difficult to calculate metrics like Cost per X, where X is some event that takes place in the Sales funnel.

Is it possible to update models to not filter for the most recent record, so we can get a full history of campaign performance?

Describe alternatives you've considered

I've tried copy/pasting the code from this package into custom models to build reports that can relate spend to conversion events, but that seems duplicative. If we can do it all in this package, that would be much better =)

Are you interested in contributing this feature?

Anything else?

N/A

elanfivetran commented 1 year ago

Hey @SamtheDeveloper91, thank you for your engagement with our Ads Reporting packages and recommending this feature expansion. Including the ability to view historical campaign metrics is an improvement that we've considered in the past and would definitely be open to exploring. That being said, we would like to be thoughtful about the implementation to ensure we can avoid issues that can arise such as extended run times, duplicity, customer confusion, etc. I don't believe the solution would be as simple as removing the filter since then our grouping for the campaign in our campaign_report output model would be sorted by every variety of our dimensions instead of just the most recent campaign.

In order to dive into this a little deeper, I was curious if you'd be open to joining our office hours to talk to our team more about your use case(s) and what changes you have already implemented? This would give us some more insight into how your thinking about implementing this feature and allow us to brainstorm a potential solution that would work for most of our customers.