Google Ads and Analytics Attribution

Discrepancy Between Google Ads and Analytics

When recently asked to place a Google Ads conversion tag for purchases on a site, I suggested just importing the purchase event from the linked Google Analytics property. I got this Gish Gallop response from marketing instead:

... Instead of using GA4 data (70% is attributed to Direct, it is not reliable) can we use the Google Ads tag? That would be first party data, it's usually more accurate than importing the GA4 key event into the google ads account ...

I responded that both Google Ads and Analytics are first party data.

... but this has to do with cookieless policies and attribution accuracy, it is not normal to have that high share of direct in the attribution mix

There's often a tension between marketing and analytics. Though on the same team, one reports on the success of the other.

In this case I had to infer that the issue wasn't with a particular conversion event but a more general distrust of Google Analytics data for reporting on paid search campaigns.

After attempting to pull some simple ads data myself, I have a newfound empathy for marketers who have to spend time in the Google Ads UI.

Google Analytics is perennially attacked for reporting data less than that of various digital ads platforms such as Google or Meta. Google Analytics will typically report between 10-30% less sessions than Google Ads clicks for a given campaign and time period. Many users use ad blockers, and many bounce before the analytics tag has had a chance to load. Ads conversions are attributed to the date of click while analytics to the date of conversion. Some campaigns may not be tagged properly.

I wanted to rule out an attribution issue further upstream from conversions so I compared Google Ads clicks with Google Analytics sessions at the campaign level.

In this case I did find something more tangible to explain a discrepancy between Ads and Analytics. After pulling an Ads campaign report with campaign_type and clicks, and then aligning with Analytics sessions for the same date range, I noticed all campaigns showing sessions between 10-30% less than clicks, as expected.

Except on one campaign of type Performance Max, where I saw ~2,000 Ad clicks compared to ~100 Analytics Sessions.

Performance Max (Pmax) Campaigns

I did some research into Performance Max (PMAX) campaigns.

Performance Max is a goal-based campaign type that allows performance advertisers to access all of their Google Ads inventory from a single campaign. It's designed to complement your keyword-based Search campaigns to help you find more converting customers across all of Google's channels like YouTube, Display, Search, Discover, Gmail, and Maps.

On performance max campaigns, clicks can mean different interaction types. They are not necessarily clicks to the website, they could be Display, YouTube or Gmail ad expansion clicks.

I'm not sure how to verify this is the issue, I could not find a click interaction type dimension in the Google Ads report builder, but the discrepancy is unique to the PMAX ads campaign and the description fits what we are seeing.

I still added the Google Ads Conversion tag via Google Tag Manager. Hopefully I restored some confidence in Google Analytics data by explaining the main source of discrepancy in this case.