What is referral attribution?
Referral attribution is deciding which referrer gets credit for a customer. It is straightforward when one person made one introduction and easy to verify. It becomes contested when several people influenced the same customer, when the customer arrived through a channel as well as a referral, or when the conversion happened long after the introduction.
The common models
First touch credits whoever introduced the customer first. It rewards the person who did the hardest part — creating awareness — and is the most natural fit for business referrals, where the introduction is usually the decisive event.
Last touch credits whoever was most recent before conversion. It is simple to implement and easy to game, because the most recent touch is the easiest one to manufacture.
Split credit divides between multiple contributors. It is fairer in principle and harder to explain in practice, and any rewards attached to it become correspondingly harder to calculate and to defend.
Attribution windows
A window is how long after the introduction a conversion still counts. Set it too short and genuine referrals go uncredited; set it too long and unrelated purchases get attributed to referrers who had nothing to do with them.
Windows should follow the actual sales cycle. Thirty days is reasonable for a consumer signup and absurd for a commercial construction referral that takes six months to reach contract.
Handling disputes
Two people will occasionally claim the same referral, and sometimes both will be telling the truth. The mechanism that resolves this is not a better algorithm — it is a record of who logged the referral first, with a timestamp, plus a person empowered to decide.
Publish the rule in advance. Attribution disputes are only corrosive when the rule appears to have been invented after the fact.
Log every override. If an administrator reassigns credit, the reason and the decision-maker should be recorded, both to keep the decision reviewable and because reward money may follow it.
MyReferral.systems is a referral network where local businesses hold a category seat, pass referrals to each other, and track every one from introduction through to closed business.
Browse referral chaptersRelated questions
Last reviewed . Next review due 5 March 2027. Tell us if something here is wrong or out of date.