Search
Clear search
Close search
Google apps
Main menu

Differences in Optimize and Analytics reports

Why the results you see in Analytics may differ from what you see in Optimize.

Results reported in Optimize may differ from results reported in Google Analytics. This is because Analytics metrics are calculated differently than in Optimize. For example, you may notice that the Analytics metrics Sessions and Ecommerce Conversion Rate do not match what you see in Optimize.

These differences occur because:

  • Analytics reports actual conversion rates while Optimize reports modeled conversion rates.
  • Reporting delays differ between the two products (Analytics is faster than Optimize).
  • Sessions accrued just before ending an experiment will not be pushed into Optimize, but will show in Analytics.

Inconsistencies between Optimize and Analytics

Both the Optimize report and the Experiments report in Analytics report Experiment Sessions, which represents the most accurate session count. However, there are expected discrepancies between Optimize and the Analytics Experiments report, which are explained below.

Optimize Reporting Tab > Experiment Sessions

As in the Analytics Experiment report, Optimize increments an Experiment Sessions for any session that has seen your experiment's variation, and all future sessions for that user for the duration of the experiment (whether they see the variation again or not).

Due to data processing delays, Experiment Sessions reported here will be different than in the Analytics Experiments report. Data is reported faster in Analytics than in Optimize; Optimize has to fetch data from Analytics on a regular basis. Also, when you end an experiment, Experiment Sessions accrued between ending and the last data fetch will not be pushed to Optimize.

Optimize Reporting Tab > Conversion Rate

Optimize uses a modeled Conversion Rate based on statistical analysis. It is not an empirical calculation as it is in Analytics, even though the two metrics have the same name. It is expected that the Conversion Rate displayed in Optimize will be different than the one displayed in Analytics, even though experiment sessions are counted in the same way.

Was this article helpful?
How can we improve it?