1. How do I start collecting data?
Start the experiment.
Once you click Start, traffic begins splitting automatically wherever the Control video is embedded.
2. Do I need a new embed code?
No.
The experiment uses the Control video’s existing embed.
If the Control video is already embedded on your site, you can start the experiment immediately without updating any code.
3. Can I run the test on only one page?
The experiment runs wherever the Control embed is installed.
If the Control video is embedded on multiple pages, the experiment will run on all of them.
To limit the scope, embed the Control video only on the page(s) you want included.
4. What happens when I pause the experiment?
When you pause:
Traffic splitting stops.
100% of visitors see the Control.
Data collected so far remains available.
You can resume at any time.
5. What happens when I resume the experiment?
When you resume:
Traffic splitting restarts.
Data collection continues from where it left off.
It may take a few minutes for changes to propagate across embeds and cache refreshes.
6. Can I change the variants after the experiment starts?
No.
Variants are locked once the experiment starts.
If you want to test a different setup, duplicate the experiment and configure new variants in the new test.
7. What does “Make winner the default experience” do?
It rolls out the winning variant to 100% of traffic.
However, there is important embed behavior:
Pages using the Control embed will now show the winning variant.
Any placements where the winner video was previously embedded will now show the Control.
This embed swap behavior is expected and ensures the Control embed remains the primary routing mechanism.
8. Why can’t I delete this video?
If a video is used in a Running or Paused experiment, it cannot be deleted.
This restriction protects the integrity of the test results.
Finish the experiment first. Once finished, you can delete the video normally.
9. Why don’t experiment results match Google Analytics or Shopify exactly?
Different tools use different attribution models and tracking logic.
Experiments rely on:
Vidalytics exposure data
The selected goal metric inside Vidalytics
External tools may:
Attribute events differently
Miss events due to browser privacy settings or blockers
Use different session windows
Small discrepancies are expected.
10. What happens if I delete a video that’s only used in a Draft experiment?
The video deletes normally.
However, the Draft experiment will no longer be startable until you select a replacement variant.
No data is lost because the experiment has not started yet.
If you have questions, feedback, or need assistance, contact our Customer Happiness Team at [email protected].