Analysis updates: clustered experiments and last value aggregation
Eppo is introducing more tools to analyze your experiments.
Cluster analysis
Eppo now supports clustered analysis for cases where there is a need to compute metrics for entities that are different than the randomization entity.
Common cases include but are not limited to:
- B2B companies that randomize at the company level, but wants to look at user level metrics
- Social media companies that randomize at the user level (for a consistent user experience), but want to look at session-level metrics
Configuring clustered analysis is easy in Eppo. It requires two steps: mapping a new subentity column in the Assignment SQL and choosing clustered analysis as the analysis type when creating an experiment. Metric creation and all other configuration remains the same.
Learn more about configuring a clustered analysis here
Last value aggregation
Last value Is a new aggregation type which takes the most recent value of a fact.
It's useful in a number of cases:
- Recording changes to the current status of a user, such as if a user has mobile notifications enabled
- Using ML models to create surrogate or proxy metrics. Each user has a prediction, the prediction changes over time, and the metric should use the most recent prediction for each user
- Data that is already aggregated at the subject level
Last value is now available in Beta and will be rolled out to all customers shortly. Reach out to support@geteppo.com if you wish to be included in the Beta.
