<aside> <img src="/icons/home_gray.svg" alt="/icons/home_gray.svg" width="40px" /> Menu
</aside>
Note: Homa Team and our external studios can view all the results of their game, you will be able to see if the A/B result has encouraging results and/or if you want to create new tests. If not, you can deploy the new value in the game for new players.
Once the test is launched, new players will start to be randomly allocated to test groups, based on the % and criteria you selected. Each day statistical tests will be run to calculate the differences between the groups. Depending on the volume of installs and the scale of differences between groups, getting statistically significant results may happen as soon as the next day, or take a few days.
For a given A/B testing, you will be able to find a table and a graph with all the results from all your test configurations. By default, we display the column with the best result.
You have a graph to compare data into the differents variants
Value | Description | Guideline |
---|---|---|
Summary | For each configuration, you will have a status result that can have 5 possibles values: | - No result: no change has been logged during the A/B testing |
- Clear loss: Ad LTV has a bad value and decrease compared to the control | ||
- Stopped: all the data has decreased, this configuration should not be applied to the game | ||
- Possible win: Ad LTV has a good value and increase compared to the control | ||
- Clear win: Ad LTV has a excellent value and increase compared to the control | ||
Result / Variant | List of all the configuration that are available in your A/B Test | - If you want to roll out this configuration in your game, you can click on the button “roll out [name]”. |
A pop-up will display and you can choose the scope to roll out the parameters. |
PS: When you stop your experiment, data will still update until D30, but we will not add new users into the experiment