Glossary: Attribution Model

What is an attribution model?

An attribution model is a rule for assigning credit for a conversion to one or more eligible touchpoints in a user's journey. It determines how reports distribute credit; it does not prove that a credited channel caused the outcome.

Common attribution models

First-touch assigns credit to the earliest eligible touchpoint. Last-touch assigns it to the latest eligible touchpoint. Multi-touch models distribute credit across several touches according to a rule, such as equal weighting or position-based weighting. Formo's documented single-touch models use first touch or last touch within a lookback window.

Attribution across web and onchain activity

A blockchain-based apps conversion may occur after visits, referrals, wallet connections, signatures, and blockchain transactions. The model can only credit touchpoints that were captured and connected to the conversion under the chosen identity rules. Report the share of conversions with a matched acquisition source and preserve transaction success criteria.

Choosing and comparing models

Choose a model that matches the question: discovery, the interaction immediately before conversion, or contribution across a journey. Keep the conversion event, identity unit, direct-traffic treatment, channel taxonomy, and attribution window explicit. Compare models side by side to understand how the crediting rule changes channel rankings.

Common mistakes

Do not treat attribution as causal measurement or compare channels when tracking coverage differs materially. Do not switch models or windows mid-comparison without labeling the change. Complement attribution with experiments, retention, and unit economics.

Related reading

Explore First-touch Attribution, Last-touch Attribution, Multi-touch Attribution, Touchpoint, and Attribution Window.

Crypto and DeFi example

A DeFi app can compare first-touch, last-touch, and position-based attribution for wallets that complete a first swap within 30 days of a campaign click. The report should show each model, the wallet matching rate, and whether the swap was confirmed, so differences in credit are not mistaken for changes in actual activity.

FAQs

What is an attribution model?

An attribution model is a rule for assigning credit for a conversion to eligible touchpoints, such as a first click, last click, or several interactions. It describes how credit is allocated, not whether a touchpoint caused the outcome.

What is the difference between first-touch and last-touch attribution?

First-touch gives credit to the earliest eligible touchpoint in the lookback window; last-touch gives it to the latest eligible touchpoint before conversion. Both simplify a journey and can favor different channels.

Does an attribution model prove a campaign caused a conversion?

No. Attribution allocates credit among observed touchpoints. Causal lift requires a suitable experiment or other causal design, and results still depend on captured data, identity matching, and the model’s assumptions.