Definition
Data-driven attribution is an attribution model that distributes credit for a conversion across the touchpoints in the journey based on patterns in your own data, rather than using a fixed rule like giving all the credit to the last click.
It is the current default model in Google Analytics 4, which retired last-click as its standard. Instead of assuming which touch mattered, it looks at the paths that did and did not lead to conversion and assigns credit accordingly, which makes it more realistic than the rule-based models it replaced.
How data-driven attribution works
Rule-based models apply a fixed formula: last-click gives everything to the final touch, linear splits evenly, position-based weights the first and last. Data-driven attribution instead learns from your conversion paths, comparing journeys that converted with those that did not, and estimates each touchpoint's contribution from that comparison.
The practical effect is that a channel is credited more in line with how much it actually appears to move conversions in your data, rather than by where it happens to sit in the sequence. This usually shifts credit away from the last click and toward the discovery and consideration touches that rule-based models ignore, which is why it tends to give a fairer picture. Our guide to revenue attribution covers how to read it, and it depends on clean tracking, so it also relies on correct GA4 ecommerce tracking.
Why it matters, and its limits
Data-driven attribution matters because the model you use decides where you invest. Last-click systematically underfunds the top and middle of the funnel; a data-driven model corrects much of that bias and gives a more balanced view of what drives revenue.
But it is still a model, not proof. It needs sufficient conversion volume to work well, it can be harder to explain than a simple rule, and, like every attribution model, it credits correlation rather than proven causation. It observes that touches preceded conversions; it cannot tell you what would have happened without them.
This is why a data-driven model should be your reporting baseline, but not your final word on a big decision. Before a major budget shift, validate with an incrementality test, which measures what a channel actually caused. Used this way, data-driven attribution allocates and incrementality verifies.
Related metrics
Data-driven attribution is one approach to the broader question of revenue attribution, and it pairs with incrementality, which checks its conclusions with real experiments. Together they move from a modelled view of revenue to a validated one.
In short
Data-driven attribution distributes conversion credit from your own data rather than a fixed rule, and it is the GA4 default. It is more realistic than last-click, but it is still a model that needs volume and credits correlation, so use it to allocate and incrementality to verify.