You can edit anything on an online experiment. However, we do not recommend it!
Indeed, any change, from the variation edition to the traffic allocation or the segment targeted, will disrupt the interpretation of your results. In this case, your results might not be reliable.
If you do decide to adjust the traffic allocation of an experiment, here’s what you need to know:
- Allocating 100% of the traffic to a single variation: In this case, all visitors to your site, including returning ones, will be directed to that variation. This will alter the experience for both new and returning users.
- Changing the allocation among two or more variations: In this scenario, the changes will only affect new visitors. Returning visitors will continue to see the same variation they were originally exposed to.
If you want to avoid these complexities and still make changes, we recommend the following approach:
- Stop the experiment (if you want to)
- Duplicate the experiment
- Edit it as you wish
- Launch this new experiment
This will result in applying the new allocation to both new and returning visitors.
