Google says it will no longer use last-click attribution as the default conversion model in Google Ads, and will default to “data-driven attribution” instead
“Because of how we've been improving and training our data-driven attribution models, we've eliminated [that] previously existing requirement,” she said.
Context & Ripple Effects
This move lands on top of a decade of Google quietly reshaping how its ad performance gets measured and disclosed. In early 2020 it stopped publishing cost-per-click and paid-click counts even as it opened up other financial metrics for the first time (with CPC steadily declining before the disclosure was pulled), and back in 2015 it had already shown willingness to redefine what counts as a conversion event by shrinking clickable ad areas and citing a double-digit conversion-rate lift (the accidental-click reduction).
Defaulting Google Ads to data-driven attribution completes that arc on the measurement side: the model that decides which touchpoint gets credit for a sale is now Google's own trained system rather than the auditable last-click rule. It also rhymes with the direction of travel since — AdSense's later shift from per-click to per-impression payment (splitting revenue share by buy-side and sell-side rates) shows the company steadily migrating its ad economics off simple click-based primitives.
First-order effects
- Advertisers and agencies see their conversion reports change overnight: credit moves off the final click and onto earlier touches in the path, so which campaigns look like they drive results — and where budgets get reallocated — shifts without any bid or spend change.
Second-order effects
- Independent attribution and analytics vendors lose the comfortable default they benchmarked against, forcing them to justify paid multi-touch measurement against a free model only Google can train on its full cross-channel signal.
Third-order effects
- If platforms keep making their own learned models the default lens on performance, attribution migrates from an open methodology advertisers can audit into infrastructure controlled by whoever sells the ads — concentrating both measurement power and the ability to frame one's own effectiveness.
The trend: Ad platforms are absorbing conversion attribution into proprietary machine-learned defaults, replacing transparent rules like last-click with models only the seller can compute.