Geo experiments versus A/B tests, and how to tell which one you need
- Ran
- Time and length
- 7:30 PM GMT+3 · 60 minutes
- Where
- Online
Attribution models are fast and always on, but correlational by construction - they cannot say what would have happened without the campaign. The session covers what a geo experiment actually is, when it replaces a standard A/B test, and how treatment and control regions get chosen.
What we cover
Welcome
What actually is a geo experiment?
When do you need one instead of a standard A/B test?
How do treatment and control regions get chosen
what happens when they aren't comparable?
Q&A
