The cost of keeping multi-touch attribution in the EU
Running a path-based model on a European customer base carries four costs. Three are quiet, one arrives all at once, and none of them appear on the invoice.
Three platforms, three answers, one budget to defend. Your GA4 export already holds the answer, and the first finding is free and never leaves your browser. Get your first finding free
By Joris van Huët, Founder & CEOUpdated 3 min read
Run the numbers for your store: the free customer journey credit calculator.
Keeping a path-based model on a European customer base carries four costs, and only one of them appears on the invoice. The other three are paid in decisions, disclosure and eventual scramble.
The four
| Cost | How it is paid | When you notice |
|---|---|---|
| Licence | Invoice | Monthly, and it is the smallest |
| Eroding coverage | Credit concentrating on observable channels | Never, from inside the tool |
| Disclosure surface | Sub-processors, transfer basis, retention | At an audit or a specific complaint |
| Misallocated budget | Over-funding proximity, under-funding upstream | In margin, attributed to something else |
The coverage cost, quantified on your own data
Pull your unassigned or direct share of revenue for this quarter and for the same quarter two years ago. That gap is the share of your business your path model has stopped explaining, and it is being redistributed across the channels that remain observable rather than being reported as missing.
The practical effect is a drift toward funding whatever is easiest to observe, which correlates with proximity to the conversion rather than with causing it. The unassigned traffic problem covers how to size it.
The misallocation cost
This is the expensive one and it is the hardest to see, because it shows up as margin rather than as a reporting error. Channels that appear close to the conversion get credited, get funded, and keep appearing close to the conversion. Channels that create demand upstream get no proximity credit and get cut in the next efficiency review.
The correction is not a better allocation rule, it is a different question. Allocation asks who to thank; a counterfactual asks what to fund, as set out in what multi-touch attribution never measured.
The disclosure cost
Every collecting vendor is a sub-processor. The cost is not the paperwork, it is the day a specific question arrives and the honest answer requires assembling a list that does not exist. Keeping the list short is easier than keeping it accurate, and a method that collects nothing is not on the list at all.
The European specifics are in GDPR-compliant attribution for European ecommerce.
What switching does not cost
It does not cost you your history, provided you export the old method's per-channel numbers by period before cancelling. It does not require a migration project, because an aggregate read runs on a window you already have. And it does not require removing your platform conversion pixels, which do a different job and should stay.
The migration order that avoids a reporting gap is in replacing multi-touch attribution in four steps.
The arithmetic worth running
Take your monthly ad spend. Take the share of it going to channels credited primarily on proximity. If a causal read moved even a modest fraction of that to better use, compare the figure against the licence you are paying for the model that credited it.
Related answers
Frequently asked questions
What does multi-touch attribution actually cost beyond the licence?
Eroding coverage that concentrates credit on whatever remains observable, a disclosure surface of collecting sub-processors, and misallocated budget from crediting proximity rather than causation. The last is usually far larger than the licence and shows up in margin instead of reporting.Do I have to remove my Meta and Google pixels?
No. Platform conversion pixels feed ad optimisation as well as reporting, and removing them degrades how the platform spends your money. Changing your allocation method is a separate decision from changing what the platforms can optimise against.Will switching attribution methods lose my history?
Not if you export the old method's per-channel numbers by period before cancelling. That export stops being available when the contract does, so it is worth doing as the first step of any migration rather than the last.
Go deeper: Incrementality testing, explained.
Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.
Have an audience of ecommerce operators? Earn 20% recurring commission by recommending the €99 read.
Keep reading
Terms in this article
- Ad SpendAd Spend is the total amount invested in advertising campaigns. It is measured against Return on Ad Spend (ROAS) to evaluate campaign effectiveness.
- AttributionAttribution identifies user actions that contribute to a desired outcome and assigns value to each. It reveals which marketing touchpoints drive conversions.
- ConversionConversion is a specific, desired action a user takes in response to a marketing message, such as a purchase or a sign-up.
- CounterfactualA counterfactual is what would have happened under conditions that did not occur, such as a customer's outcome had they not received the campaign they did receive. It is never observed, so every effect estimate has to stand in for it with a control group, comparable units or a forecast.
- GDPRGDPR is a European Union regulation governing data protection and privacy, impacting how businesses collect and process customer data.
- Multi-Touch AttributionMulti-Touch Attribution assigns credit to multiple marketing touchpoints across the customer journey. It provides a complete view of channel impact on conversions.
- RevenueRevenue is the total income generated by the sale of goods or services related to a company's primary operations.