When comparing A/B test results in Mida with metrics from Google Analytics (GA4) or other analytics tools, you may notice discrepancies—especially in single-day reports.
Here’s why that happens.
How Mida Tracks Conversions
Mida is purpose-built for A/B testing, not general analytics. It tracks conversions based on:
The day a user first enters a test (i.e., when they are assigned to a variant)
One conversion per user, regardless of how many times they convert later
So, in Mida, all conversions are tied back to the day a user first entered the test—even if they purchase days later.
Example: Why April 8 Might Look Off
Let’s say a user’s journey looks like this:
Date | Action |
April 7 | First visit → enters Variant A |
April 8 | Returns and makes a purchase |
In GA4: The purchase will show under April 8.
In Mida: That user’s conversion will be attributed to April 7, because that’s when they were first bucketed into the test.
So when you look at Mida’s April 8 data, it’s only showing:
Users who first entered the test on April 8
Their eventual conversions (even if they buy days later)
How to Compare More Accurately
To make a fair comparison:
Use Mida’s Event Log to view raw conversion events with timestamps.
Match GA4 events with Mida's first exposure date rather than the event date.
For day-by-day breakdowns, aggregate both tools by cohort start date if possible.
Summary
Tool | Counts Conversion On |
GA4 | The date the purchase happened |
Mida | The date the user first entered the test |
If you're seeing fewer conversions in Mida for a specific day, it doesn’t mean they’re missing—it’s likely those users entered the test earlier.