Search...
Menu

A/B Test

1. Function overview

A/B Test is a comparative test that divides users into a control group and a test group in proportion and uses different configurations for different groups. You can use different test groups to use different advertising strategies to compare the results to help you verify which advertising strategy works better.

Hyperbid Tools A/B Test supports the following functions:

(1) Supports A/B Test in segment

(2) After the A/B Test is completed, you can select the optimal test group with one click.

(3) You can view the group data and historical advanced settings of the A/B Test.

 

2. Operating Instructions

2.1 A/B Test process

2.1.1 Create an A/B Test

(1) You can select the Mediation Unit for which you want to create an A/B Test from the [Application] - [Mediation Unit] list. Click the Mediation Unit to enter the Mediation Unit editing page. Click [Add A/B Test].

(2) After the A/B Test is created, the WaterFall configuration of the original Mediation Unit will be automatically synchronized to the group with the smallest ID in the A/B Test (for example, if three groups A\B\C are configured, it will be automatically copied to group A). ​​If other groups (B/C) want to synchronize the WaterFall configuration of the original Mediation Unit, you can use the copy existing WaterFall function.

(3) The effective time of the A/B Test can be selected immediately or at a specified time.

 

2.1.2 Setting WaterFall for A/B Test groups

After you create an A/B Test, you can configure AD Units for each A/B Test group in the Mediation Unit.

 

2.1.3 View A/B Test data in A/B Test

(1) Go to the [Application] - [A/B Test] page, select the A/B Test you want to observe, and view the comparative data of different groups in the current A/B Test.

(2) If you need to view detailed data of AD Units in different groups under the A/B Test, you can select the corresponding Mediation Unit in [Report] - [Waterfall Report] to view it. In [Report] - [Waterfall Report], you can view all historical data of the A/B Test.

(3) For an A/B Test that has ended, you can click [History Settings] to view the advanced settings of different groups under the A/B Test.
Note that the history settings only record the advanced settings of each group before the end of the A/B Test.

 

2.1.4 View A/B Test data in A/B Test

(1) You can view test data by segment on the [Application] - [A/B Test] page.

(2) A/B Test support data indicators:

  • DAU
  • DEU
  • Estimated revenue/DAU
  • Estimated revenue/DEU
  • Estimated revenue
  • Estimated eCPM
  • Mediation Unit request
  • Mediation Unit fill rate
  • Impression
  • Impression rate
  • Penetration rate
  • Impression/DAU
  • Impression/DEU
  • Click
  • Click rate
  • isReady call count
  • isReady success rate
  • Number of people cut

 

2.2 End the A/B Test

When you finish an A/B Test, you can select the optimal grouping strategy. After the test ends, the optimal grouping strategy is delivered by default.

 

3. FAQ

(1) How to allocate devices for A/B Test?

If the AB strategy remains unchanged, the device ID will be fixedly assigned to one of the groups. If the strategy changes, such as modifying the volume permissions, the device ID will be reassigned.

Previous
Mediation Platform
Next
Report
Last modified: 2025-03-25Powered by