TopOn new A/B testing, your winning monetization strategy

What is monetization A/B Testing?

An A/B test is a controlled test that splits users into different groups, and exposes each group to different app configurations. This allows you to test different monetization strategies and compare results through data analysis, thus helping you understand which strategy performs better.

TopOn New A/B Testing

A/B test are ones of the most important parts during advertising revenue analysis. With TopOn’s A/B test tool, publishers could definitely get a lot of insights for monetization strategies.

Our new A/B test tool provides:

• Traffic split across multiple placements, which makes user groups unified.
• App level experiment data report as well as placement level, provides you better understanding of the overall performance of the different advertising strategies, while detailed data for each placement.
• Various metrics for the tests in full report page.
• Go with the strategy which has the best performance on one click at the end of the test.

Set up the tests

1.1 Create an A/B test

You can create and manage A/B tests via Advanced → A/B Test. If you do not get the option, please contact our operation team. After updating to new A/B testing, the former entrance: Mediation→ Advanced→ A/B Test will no longer be available.

When creating test, you could set up multiple A/B groups. Fill in the weight when adding a new group and we will calculate the percentage for you automatically.

Meanwhile, multiple placements in the same App could be covered under one test. This is to make sure that, users could be assigned to the same bucket for several placements simultaneously, in which way we could observe the result of the experiment on the App level.

• You could start the test immediately after setting up all the configurations, or choose a future time to get it activated.
• When the test is set up, the original waterfall configuration of the placements will be automatically synchronized to the group with the smallest ID. (Eg. if there are 3 groups - A, B and C, then A will have the mediation configuration of the placements.)
• You could quickly locate a test by status tabs or search function on the page.

1.2 Set up waterfall(s) for A/B Tests

You could configure Segment and Ad Source for each A/B Test on Mediation page.

To complete the configuration quickly, especially when you have many segments or waterfall with many ad sources, try the Copy Segment and Copy Ad Source funtions on Mediation page.

How to copy segments and ad sources in a batch way? Take duplicating ad sources as an example. Click Copy AD Source, then you'll open a window in which you can find ad sources of other A/B Test and Segments.

Check all ad sources you want to copy to the current Segment and click Submit, then the selected ad sources will appear in the waterfall of the current Segment.

1.3 Get the data insights of the tests

You could get the performance data of the A/B Tests on the A/B Test page and DETAIL page. Data will be displayed by App dimension or placement dimension. Customize metrics are provided.

• If you like to see the trends, to get some insights from charts of the data, please click the data trend icon by the side of the test and each placement. The charts will show you the data within the selected date range.
• Here you could also change metrics or reset date range to display the charts and details of the data you want.

Note: Please refresh the page if you could not see the icon on the page.

• For the data came before the test is set up, we provide historical traffic to collect and show it.
• Historical traffic could only provide data for its own ad sources as a whole, but can not split by any A/B Test nor segment. Still, you could find historical data related to A/B test and segment on full report page.
• You can not change waterfall configuration for historical traffic group.

2. Finish A/B Tests

You could choose to finish the test on all placements or any specific placement.

For wrapping up an A/B Test, you could choose the group with the best performance (thus has the best strategy) to proceed.

Note, if you choose to end a test on the mediation page of any placement, with the convenient entry by the side of AB Test, you will only end the test on that placement, but not the whole test.

After an A/B Test is wrapped up, HyperBid SDK will still cache the strategy for a period of time. The group(s) that are not selected will generate legacy data for a while. If you'd like to view the data, you could try:

• Go to full report page to query the placement data with the A/B test filter and dimension.
• Or go to A/B Test page to view the Finished A/B Test's data.

After an A/B Test is finished, on the Mediation page only the data for the best-performance group will be shown. Data for other groups could be found on the Full report page.

Historical A/B Test configuration and data

If you have experiments with our previous version of A/B Test, we'll automatically synchronize the tests for you to the current version of A/B Test. We'll create each placement a separate A/B test then.

Contact

Don’t hesitate to contact TopOn's business operation team, or email us via business@toponad.com to apply for permission to activate new A/B testing.