Where attribution debt is actually hiding
Everyone looks for it in the ad platforms. Four places it actually accumulates, three of them inside decisions rather than inside any dataset.
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
The four
| Location | What accumulates there |
|---|---|
| Unassigned traffic | Revenue credited to direct that something else caused |
| Legacy thresholds | Rules calibrated years ago against numbers that have drifted |
| Channels below the floor | Cut on absent evidence rather than negative evidence |
| Undated definition changes | A series that is discontinuous and looks continuous |
Unassigned traffic
Some share of orders arrives with no usable source: referrers stripped, consent refused, journeys crossing apps. That revenue does not disappear; it usually lands in direct, which flatters direct and starves whatever actually caused the visit.
Detail in the unassigned traffic problem.
Legacy thresholds
A rule that says scale above a certain return was calibrated at some point against some number. If nobody can say when or against what, it is now an arbitrary constant governing real budget.
This is the cheapest debt to retire: recalibrate deliberately, express the rule as a condition on the confidence interval rather than the midpoint, write down the date. Method in setting thresholds worth alerting on.
Channels below the floor
A channel whose spend is too small for any method to separate its effect shows up with no number, and in a review no number reads as no value. It gets cut. That is a decision taken on absence of evidence rather than evidence of absence, and it is how small experimental channels die.
The fix is a three-way report: measured and above threshold, measured and below threshold, not measurable at this spend. The arithmetic is in the measurability floor.
Undated definition changes
Someone regroups channels or changes a lookback and does not record it. The series is now discontinuous, looks continuous, and every comparison across the break is wrong in an invisible way.
This one produces the longest investigations, because there is no trace of it in the marketing data. A dated definitions document costs five minutes and prevents all of it.
Why these hide better than platform overclaiming
Because platform overclaiming is arithmetic you can check in twenty minutes, and these four are not visible in any single report. They surface in comparison across time, which is why keeping a dated series matters more than any individual read.
A read costs €99, once, excluding VAT, and is refundable within 30 days, no questions asked.
The audit to run once
Four questions. What is our coverage. When was each threshold set and against what. Which channels have we cut on no number. And when did any definition last change.
Related answers
Frequently asked questions
Where does attribution debt actually accumulate?
In unassigned traffic credited to direct, in legacy thresholds nobody can date, in channels cut for having no number rather than a bad one, and in undated definition changes that make a series discontinuous while it looks continuous.Why is a channel with no number dangerous?
Because in a review no number reads as no value, so it gets cut.What audit surfaces hidden attribution debt?
Four questions: what is our coverage, when was each threshold set and against what, which channels have we cut on no number, and when did a definition last change.
Go deeper: Causal attribution, 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
- AttributionAttribution identifies user actions that contribute to a desired outcome and assigns value to each. It reveals which marketing touchpoints drive conversions.
- Attribution DebtAttribution debt is the gap between what your ad platforms claim drove revenue and what actually caused it, carried quarter after quarter into the budget. It is how marketing debt accrues: allocate on claimed conversions long enough and the plan itself becomes the liability.
- Confidence IntervalConfidence Interval is a statistical range of values that likely contains the true value of a metric. In marketing analytics, it quantifies uncertainty around estimates, indicating the precision of an outcome or causal effect.
- DashboardA dashboard is a visual display of key information required to achieve specific objectives. It consolidates data onto a single screen for quick review.
- DashboardsDashboards are graphical user interfaces that provide at-a-glance views of key performance indicators (KPIs). They monitor campaign performance and visualize attribution insights.
- IncrementalityIncrementality is the part of an outcome, such as orders or revenue, that a marketing activity caused: the extra results compared with what would have happened without it. It is a causal effect, not the number of conversions credited to a channel.
- RevenueRevenue is the total income generated by the sale of goods or services related to a company's primary operations.