When running an experiment, it is common to see conversions appear on only one variant, especially when total conversions are very low.
This is expected behavior and usually caused by normal result variation.
What variance means (in simple terms)
Variance means that when numbers are small, results can look uneven just by chance.
If only a few conversions have happened, it is completely normal for:
All conversions to land on one variant
Another variant to show zero conversions
Results to look “wrong” or unbalanced early on
This does not mean anything is broken.
Why single-digit conversions are especially misleading
When conversions are in the single digits, each individual conversion has a large impact on the results.
For example:
With 5-10 conversions total, it is very easy for all of them to land on one variant
This can happen even if both variants are performing the same
As more conversions happen, results usually begin to balance out based on real performance.
What this does not indicate
Seeing zero conversions on a variant does not mean:
The variant is not receiving traffic
The conversion goal is not firing
Tracking is misconfigured
If visitors are reaching the conversion page, tracking is working.
What to do next
Allow the experiment to run longer and collect more data
Avoid judging results based on a small number of conversions
Re-check results once conversions are no longer in single digits
Important notes
Early experiment results are often noisy and unstable. It is normal for one variant to temporarily show all or none of the conversions when traffic and conversions are low.
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article