Optimized models are automatically-generated SKAdNetwork revenue models calculated to give you the most accurate estimate possible of how much revenue you are earning through a SKAdNetwork campaign.
For background on SKAdNetwork and conversion models, see the SKAdNetwork Model Configuration FAQ.
SKAdNetwork revenue models are only as effective as the revenue buckets defined for them (see How does a revenue model work?).
Optimized models solve the dilemma of how to define the revenue buckets (how many buckets, how big/small, etc.) by generating the best revenue buckets for your app automatically.
Optimized models are calculated based on real user-level revenue event data for the specific app (as measured by the Singular tracker).
Currently, optimized models are only available for In-App Purchase Revenue models (not ad monetization or combined models).
Singular requires two months of data for an app as well as a minimum amount of tracker installs to be able to generate a statistically significant model.
If you are creating your first SKAdNetwork revenue model (but we have enough data about that app from the tracker already), we present the optimized model as an option in the Model Configuration panel.
If you have an active revenue model, but we can offer a significantly more accurate optimized model, we show a green button next to the model that says Optimized Model Available. You can click it to see the details. Then you can choose to use the optimized model (this overrides the existing revenue model you were using).
Singular recommends an optimized model if:
- At least a month has passed since you made changes to your revenue model;
- We are able to generate a model that is significantly better (more accurate) than the existing one.
Optimized models are calculated based on your known revenue events, meaning IDFV-based data for iOS installs that come through the tracker.
We take these known revenue events and divide them into buckets using proprietary methods so that the buckets most accurately represent the actual revenue.
Once we've scored our models on accuracy, we compare the best ones with your existing SKAdNetwork revenue model.
If the automatically-generated model is significantly more accurate than the existing model, we will recommend it to you in the SKAdNetwork Model Configuration page.