This guide explains how Mida handles conversion attribution (including purchases, sign-ups, and other conversion events) across different scenarios in A/B testing.
Active Test Participation
Mida only tracks conversions from users who are actively participating in your A/B test:
Conversions are counted only after a user becomes part of the test group
Users who make purchases without being assigned to a variation aren't included
Previous actions before joining the test aren't counted in results
Delayed Conversions
When a user converts days after their initial visit:
Mida uses cookies to track users across multiple sessions
If a user sees Variation A on day 1 and converts on day 3, the conversion will be correctly attributed to Variation A
This ensures accurate tracking of conversion data even with longer customer decision cycles
Cross-Device and Cross-Browser Attribution
Cross-device tracking is a common challenge faced across the web analytics industry. Here's how Mida handles it:
Users are treated as new visitors on new devices/browsers since the original tracking cookie isn't present
They may see different variations on different devices
If a conversion occurs on a device/browser where no variation was shown, the conversion won't be attributed to any test variation
Mida only attributes conversions to variations that users have actually been exposed to
Best Practices
To ensure the most accurate test results, we recommend:
Running your tests for a sufficient duration to minimize the impact of edge cases
The larger your sample size, the more reliable your results will be
Being aware that ad blockers may affect tracking in some cases
Note: Conversions can include any measurable user action such as purchases, sign-ups, form submissions, or other defined goals in your testing setup.