THE SHORT ANSWER
Marketing attribution assigns conversion credit to observed interactions according to a chosen model and data scope. It supports reporting and optimization, but identity gaps, missing exposures, model rules and selection effects limit causal conclusions.
Every attribution answer contains rules
A report defines the eligible interactions, identity, lookback window, conversion time, channel rules and credit model. Change those rules and the reported credit can change without any customer doing something different.
Record the model beside the result. ‘Paid social drove 30%’ is incomplete without the system, conversion, window and credit rule.
Evidence & context: Google Analytics
Why reports disagree
| Difference | Effect |
|---|---|
| Observation | Platforms see their interactions; analytics sees tagged site activity |
| Identity | Browsers, devices and consent break or model journeys differently |
| Windows | Credit may include different days before or after interaction |
| Model | Click, view and data-driven rules distribute credit differently |
| Time | Reports may assign to interaction date or conversion date |
Attribution is not incrementality
Attribution asks how observed credit is assigned. Incrementality asks what changed because the marketing occurred. A channel can receive credit for converting customers who would have bought anyway, or influence customers without receiving the final credit.
Randomized experiments can estimate causal lift under their design and context. They still require adequate execution, power and interpretation; one result is not a universal channel truth.
Evidence & context: Google Research
Use a measurement stack
- Business ledger for completed outcomes and economics
- Analytics for cross-channel journeys and site behaviour
- Platform reports for delivery and in-platform optimization
- Experiments or credible comparisons for incremental questions
- Qualitative research for motives and unmet needs
When a campaign weakens, use the diagnostic sequence before changing attribution rules to improve the story.
Sources & further reading
- API dimensions and metrics
Google Analytics. Official GA4 reporting definitions checked 13 September 2026, including session source, medium, referral and landing-page dimensions. Attribution remains limited to observable interactions.
- Methods for Measuring Brand Lift of Online Ads
Google Research. Original research using randomised experiments to estimate advertising effects; no universal lift or ROI benchmark is inferred.
Examples and exercises are illustrative unless attributed to a source. No independent expert review is claimed.
A correction, a counterexample or an experience worth sharing?
Join the conversation ↗