Attribution Is Broken — Here’s What to Do Instead

If you’re running paid acquisition and relying on last-click attribution, you’re making budget decisions based on incomplete data. This isn’t a controversial opinion anymore — it’s a widely acknowledged reality that most teams still haven’t addressed.

The problem is structural. Modern buying journeys are non-linear. A prospect might see a social ad, read a blog post two weeks later, click a retargeting ad the week after that, and finally convert through a branded search query. Last-click attribution gives 100% of the credit to that branded search — which is the channel that least needs the credit.

“Last-click attribution is a comforting lie. It tells you that the last thing someone clicked before converting is the thing that caused the conversion. That’s like crediting the final step of a marathon for the entire race.”

Google’s own data-driven attribution model in GA4 is a step in the right direction, but it’s still a black box. You don’t know the weights, you can’t audit the model, and it’s optimised for Google’s ecosystem. If you’re spending significant money on Meta, TikTok, or organic content, you need a model that gives those channels fair representation.

The practical solution is a blended model. Start by tracking every touchpoint with proper UTM parameters and custom events. Build a first-party data layer that captures the full journey — not just the conversion event. Then use that data to build your own weighted attribution model, even if it’s simple.

A basic position-based model (40% first touch, 40% last touch, 20% distributed across middle touches) is already more honest than last-click. It acknowledges that awareness and consideration matter, not just the final click.

The point isn’t perfection. The point is moving from a model that’s definitively wrong to one that’s approximately right. The budget decisions you make based on a better model will compound over time — and that compounding is where real growth lives.