Reporting

This guide provides an overview of the AI-powered ad optimization process following Airflux integration and how to access the initial report.


1. Model training

Once the integration with Airflux is complete, your game data will be fed into the Airflux AI for model training, which will take approximately 2 weeks.

Why does it take 2 weeks for model training?

Airflux requires a sufficient amount of data for player segmentation to develop optimized advertising strategies tailored to each profile. This initial training period is an essential process for personalized ad display optimization.

The following data sets are key to model training:

  • Player behavior: Play patterns, such as game playtime, play frequency, purchase history, and ad viewing history.

  • Game progress: Player’s current level, in-game status, etc.

  • Ad attributes: Ad types, frequency, rewards players receive after watching rewarded ads, etc.

Using anonymized data in adherence to privacy regulations, Airflux effectively trains its models to determine the best timing for ad displays, enhancing LTV and retention without affecting the gameplay experience.

To learn more about how inference works in Airflux, click here.

2. Accessing the initial report

After sufficient model training, the first optimization report will be emailed within 4 weeks of app deployment.

The report will include the key metric changes resulting from optimization, including shifts in LTV and retention, and details of the Airflux settings customized for your game.

Have questions?

If you have any questions regarding model training and reporting, contact us at [email protected].

Last updated