Attribution is how you answer the question: "Which marketing channel actually caused this sale?" Get it wrong, and you'll over-invest in the wrong channels and under-invest in what's actually working. Here's a practical guide to choosing the right model.
The Attribution Models
Last-Click Attribution
100% credit goes to the last touchpoint before conversion. Pros: simple, easy to understand. Cons: ignores everything that happened before the final click. It overvalues search and direct traffic while undervaluing awareness channels like social and display. This model is still the default in many platforms, which is why so many businesses over-invest in bottom-funnel tactics.
First-Click Attribution
100% credit goes to the first touchpoint. Pros: values awareness and discovery. Cons: ignores everything that nurtured and closed the deal. Useful for understanding which channels introduce new customers to your brand, but incomplete as a decision-making tool.
Linear Attribution
Equal credit distributed across all touchpoints. Pros: acknowledges every interaction. Cons: treats a random display impression the same as a high-intent search click. It's better than last-click but still naive.
Time-Decay Attribution
More credit to touchpoints closer to the conversion. Pros: reasonably models real buying behavior — recent interactions are usually more influential. Cons: still somewhat arbitrary in how it weights the decay. This is a good middle-ground model for most businesses.
Data-Driven Attribution (DDA)
Uses machine learning to assign credit based on actual conversion paths in your data. Pros: the most accurate model available. Cons: requires enough conversion data (Google recommends 600+ conversions per month) and a diverse media mix. If you qualify, this should be your default in Google Ads.
Our Recommendation
Use data-driven attribution if you have enough conversion data. If not, time-decay is the best practical alternative. But more importantly: don't rely on any single model. Look at multiple attribution views side by side to understand the full picture.
Beyond Platform Attribution
Platform-reported attribution is inherently biased — Google credits Google, Meta credits Meta. For the truth, use incrementality testing (geographic lift studies, holdout groups) and media mix modeling. These approaches answer "what would have happened without this channel?" rather than "who gets credit for this conversion?"
Good attribution is the foundation of smart budget allocation. We can help you build an attribution framework that works.