Definition
Attribution is the method that distributes the credit for a conversion across the touchpoints the customer encountered along the way. The model decides which channel gets the credit, and therefore where budget flows.
Also called: Attribution modeling, Attribution model, CreditingSeveral channels can claim the same purchase.
All 100 purchases in this example are claimed by at least one channel. Adding channel reports counts shared purchases twice. MER uses store revenue and total marketing spend; it does not determine causal lift.
The models and their bias
Last-click gives all the credit to the final touchpoint — typically brand search or retargeting, which makes the upper-funnel channels look weak. First-click does the opposite. Data-driven and position-based models spread the credit, but still build on data that has grown patchier and patchier after iOS updates and cookie restrictions.
No model is “true”. They're lenses, not verdicts. The danger is treating the platform's attributed ROAS as reality and cutting the very channels that actually drive demand.
How we use attribution
We use attribution for direction, not for judgment. For the final budget decision we weigh it against MER (the blended total) and incrementality tests. Attribution says “something seems to be happening here” — incrementality says “something is genuinely happening here”.
A solid server-side setup makes attribution less patchy by rebuilding the signals the browser has shut down. But even perfect data doesn't change the fact that attribution and incrementality are two different questions.
Attribution windows: where the number gets inflated
The model itself is only half of it; the window is the other. Meta reports on 7-day click and 1-day view by default — so the platform takes the credit if someone clicks and buys within a week, or just sees the ad and buys within a day. View-through conversions are especially generous: many of them would have happened anyway. Google counts just as broadly. The longer and more inclusive the window, the more the attributed ROAS is inflated.
That's why two accounts with identical results can report wildly different ROAS, purely because they have different attribution settings. It's also why the sum of every channel's attributed revenue often exceeds what actually came in: each measures in its own silo with its own generous window.
Our approach: use one consistent window to compare over time (so trends are real), but never make budget decisions on the attributed number alone. The final verdict belongs to MER and incrementality tests — the only measures that can't be inflated by a window.
Frequently asked questions
Which attribution model is best?
There's no single best model — each has a built-in bias. Smarter than picking one is reading several together and validating with blended numbers (MER) and incrementality tests.
Why do Meta and Google show higher numbers than my store?
Because each platform credits itself with sales that other channels also saw — they double-count. So the sum of the platforms' reported revenue often exceeds the actual revenue in your store.
From insight to action
See how it applies in practice.
Website, channels and profit measurement
The Løgbutikken engagement covered the website, email, Meta, Google Ads and ProfitMetrics. The business grew and was subsequently acquired.
Read the caseMeasurement and customer economics
Discuss measurement and bidding
We review the values you measure and how they can inform budget decisions.
Choose a time30 minutes · No obligation
About the author
Nicklas Segatz Mortensen
Growth hacker and fractional CMO with 10+ years' experience and hundreds of millions in managed ad spend behind him. Background from larger Danish and international scale-ups, and from the agency world.
Meet the team