Probabilistic Attribution
What is probabilistic attribution?
Probabilistic attribution is a form of attribution based on probabilities, not ID or device matching. It relies on machine learning and statistical modeling techniques to identify probable conversions across various marketing touchpoints.
How does probabilistic attribution work?
When users click on an ad, data is shared with their MMP. Then, if they download and open the app, that information is also shared. Within the MMP’s servers, they will match these two actions using multiple, non-deterministic data points to attribute installs to an ad campaign.
Is probabilistic attribution accurate?
It will never be as accurate as matching devices using a unique identifier such as IDFA, as it relies on probability to inform on campaign effectiveness.
For users who opt-out of IDFA with the new ATT consent flow, this is a method fully compliant with Apple’s guidelines to manage attribution to a high degree of probability. This helps provide more confidence that the data is accurate enough to inform marketing decisions and assess the overall performance of ad campaigns and creatives for LAT traffic.