Affiliate attribution
Affiliate attribution is the rule that assigns a sale (and the commission) to a referring partner, conventionally the last affiliate click within a tracking-cookie window. Because money changes hands, the model is contractual, not just analytical: it determines who is paid, invites last-click stuffing, and clashes with the brand's own multi-touch view of the same sale. This page explains the mechanics and the conflicts.
How affiliate credit is assigned
When a user clicks an affiliate link, a tracking cookie or click identifier records the partner. If a qualifying sale happens within the cookie window — often days to weeks — that partner is credited and paid a commission.
The near-universal default is last-affiliate-click: among competing affiliate touches, the most recent one inside the window wins the whole commission.
Why it causes disputes
Last-click cookie windows reward whoever sits closest to the sale. That incentivizes coupon and loyalty sites, toolbars, and ad injectors to drop a cookie at the last moment on demand other channels created — sometimes called last-click stuffing.
It also conflicts with the brand's multi-touch view: your analytics may credit search and email, while the network pays an affiliate who only appeared at the end. Reconciling the two requires comparing path data, not just totals.
- Default rule: last affiliate click within the cookie window
- Window length and de-duplication are contractual choices
- Last-click windows invite coupon/toolbar 'cookie stuffing'
How it appears in analytics and logs
When affiliate-reported sales exceed what your multi-touch model credits the channel, the affiliate's last-click window is capturing demand other channels created.
Diagnostic use case
Understand why an affiliate network credits a different source than your own analytics, and why last-click cookie windows drive coupon and toolbar abuse.
What WebmasterID can help detect
WebmasterID's first-party path data lets you see whether an affiliate touch was the genuine introducer or a late last-click rider on existing demand.
Common mistakes
- Assuming network-reported sales match your own attribution.
- Ignoring that long cookie windows over-credit late affiliates.
- Paying for sales other channels demonstrably introduced.
Privacy and accuracy notes
Affiliate tracking relies on cookies or click identifiers subject to consent and lifetime limits; modeled gaps appear as windows shrink. Educational, not legal advice.
Related pages
- Coupon code attribution
Coupon (promo) code attribution assigns a sale to the partner, creator, or campaign whose code the buyer entered at checkout. It is deterministic and cross-device by nature — the code is typed regardless of cookies — which makes it popular for influencers and affiliates. But it only captures buyers willing to use a code, and shared or leaked codes can be claimed by buyers a partner never reached.
- Last-click attribution: simple, and what it hides
Last-click attribution assigns 100% of a conversion's credit to the last touchpoint before it. It is simple, deterministic, and the historical default — which is exactly why it misleads: it ignores every earlier touch that created demand, systematically overrating bottom-funnel channels and underrating discovery.
- Influencer attribution measurement
Measuring influencer impact is hard because most of it is unattributable by clicks: audiences see content, remember the brand, and convert later through search or direct. Practical influencer attribution combines unique discount codes, dedicated tracking links, and self-reported 'how did you hear about us?' surveys, accepting that view-through and dark-social effects will always be undercounted by last-click models.
- Attribution analytics
See whether an affiliate truly introduced the demand.
Sources and verification notes
- Google Analytics Help — Default channel group (Affiliates)Documents how affiliate traffic is grouped as a channel.
Last reviewed 2026-06-24. Facts are checked against primary/official sources where available; uncertain specifics are marked “Data not yet verified” rather than guessed.