Affiliate Last Click Attribution Explained: Who Gets the Sale?

How Affiliate Last Click Attribution Works

Here is affiliate last click attribution explained: last-click attribution is a model that assigns credit to the final click before a conversion ^[2]. Google Analytics says last-click models give 100% of credit to the final clicked channel ^[2]; for any affiliate program, check its terms, as Amazon’s Associates Program Operating Agreement illustrates ^[1].

You have probably met a quoted cookie window in a forum thread and wondered whether your work evaporates the moment a reader opens a second tab. That worry is reasonable. Attribution is not a moral system that rewards whoever did the most research; it is a set of rules for assigning credit ^[2], and for Amazon’s program those rules are set out in its Associates Program Operating Agreement ^[1].

The good news is that the rule is almost always findable in writing before you join, and a handful of questions separate a program that credits your readers’ purchases from one that quietly hands them to somebody else. This page walks through the mechanics, gives you a fill-in sheet to compare programs, and shows the same five clicks paying out two different ways.

What Counts as the "Last Click" — and What Gets Ignored

Google’s analytics documentation is the clearest public explanation of the mechanic, and its worked examples are worth copying into your notes ^[2]. Under the paid and organic last click model:

Three details from that page matter here. First, "paid and organic last click" and "last non-direct click" are two names for the same model ^[2]. Second, direct traffic is excluded from receiving credit unless the whole path is direct ^[2], so a reader who types your site name into the address bar and buys can look like nothing at all in your reports. Third, the first-click, linear, time-decay and position-based models were retired and are no longer available as of November 2023 ^[2], which means you cannot simply flip a report to first-click to prove an early-stage article did the work.

Now the part beginners mix up, so let us be blunt. Those models live inside Google Analytics, a reporting tool. They decide how your reports assign credit; they do not decide what an affiliate network pays. At Amazon Associates, for example, payout rules come from the Associates Program Operating Agreement you accept when you join ^[1]. Google’s own illustration of an attribution path even lists an affiliate exposure alongside paid search, social and search touchpoints ^[2] — a useful reminder that from a network’s side, your click is one event among several competing for the same 100%.

Five Rules That Decide Whether Your Click Survives

Nobody outside a program can tell you what it will pay in a specific case; the answer lives in its terms, and those terms change. What you can do is ask the same five questions of every program before you give it an evening of your writing time.

• Attribution window — “How long after a click does a purchase still count for me?” A short window expires while your reader is still comparing options. The click was yours; the sale is not.
• Override rule — “If another partner’s link is clicked after mine, whose sale is it?” This is where last click bites. Whether an earlier click can be replaced by a later one within that program’s window is something to confirm in its terms.
• Cross-device and logged-in tracking — “Does a click on a phone count if the purchase happens on a laptop, or inside an app?” Readers research on one device and buy on another all the time.
• Direct and returning customers — “If they type the brand name later and buy, how is credit handled?” Google’s reporting model ignores direct visits ^[2]; networks write their own version, so read theirs.
• Reversals — “What happens to a credited commission if the item is returned or cancelled?” Amazon, for example, may hold accrued unpaid commission income after termination to account for cancellations or returns ^[1].

The override rule costs beginners more than the commission rate

A high headline commission on a program that hands every late click to someone else is worth less than a modest rate you actually collect. Look for the sentence explaining what happens when two partners refer the same customer. If a later click replaces yours, plan your pages around the moment a last click happens: a review or comparison written for someone who is ready to choose sits closer to that moment than an explainer written for someone still deciding what to buy. Search intent for affiliate content shows how to tell those page types apart before you start writing.

Reversals are a separate risk that feels identical

You can win the attribution argument and still not get paid, because a commission is not money until it clears. Returns, cancellations and payout timing live in the terms too, and Amazon explicitly reserves the right to hold accrued unpaid commission income for a reasonable period after termination to account for cancellations or returns ^[1]. Pending, approved, reversed, paid explains the status labels in a dashboard and what each one means for your balance.

A Pre-Join Attribution Check Sheet (Copy This)

A spreadsheet row can hold a program’s details. Paste this into a spreadsheet before you write anything, and fill it from the program’s terms page rather than from a review — including ours.

When the terms page is open, use your browser’s find function on three groups of words: "last click" or "last referral"; "cookie" and "window"; and "reversal", "cancellation" and "return". If the document does not answer a question, email support and keep the reply in the same folder. A program that will not put its credit rule in writing is telling you something useful.

Want the fill-in version? Get the free affiliate marketing checklist. Get the free affiliate marketing checklist Disclosure: We may earn a commission if you use some links on this page.

For the wider set of clauses worth reading before you join, see 12 clauses to read in an affiliate program’s terms. For how window lengths differ and what they mean in practice, affiliate cookie duration and attribution windows goes deeper than this page can.

Illustrative Example: Five Clicks, Two Different Payouts

Illustrative example. The price and commission rate below are invented to show the arithmetic. They are not typical earnings, and no result is promised.

Say a program sells a $100 product and pays 25% commission, so one qualifying sale is worth $25. Five readers click your link in a week.

Five clicks at $25 each looks like $125 of potential commission (5 × $25). Two of those clicks survived the rules, so $50 is credited (2 × $25). Nothing about readers B and C is unusual or unfair — they clicked a different link, or clicked too late. What changes the arithmetic is the program’s window and override rule, which is exactly why you check them first.

Notice reader D, though. In reader D’s case, the email click is the last click before the purchase, so it receives the credit under a last-click model ^[2], and it only works if the program allows affiliate links in email and your list is run properly; can you put affiliate links in emails covers the program rules and the CAN-SPAM basics.

What Your Own Reports Can and Cannot Tell You

You will never see another affiliate’s clicks. Your network dashboard shows what was credited to you — clicks, conversions and statuses — not why a particular reader’s purchase went elsewhere. That limit is permanent, so base decisions on the written rules rather than on guesses about other people’s numbers.

What you can see is the shape of your own traffic. In Google Analytics, the attribution settings page lets you review the reporting attribution model, which channels can receive credit, and the key event lookback window ^[2]. The data-driven model can also reattribute conversions for up to seven days after the event ^[2], so a number you screenshot today may shift slightly tomorrow. That is normal, and it is one reason to judge a page over months rather than days.

To connect a specific page to a specific commission, use sub-IDs or tracking parameters on your affiliate links; UTM tags and sub-IDs shows how to set them up so the network report can name the page that earned. Then compare your own earnings per click with the headline figure the network advertises, which is a different calculation — what EPC means in affiliate marketing explains why the two rarely match. For the underlying mechanics, our guide to affiliate tracking on the main site is the place to start.

So Is Your Work Being Stolen?

Usually, no — and it matters that you can tell the difference. Google Analytics, for example, offers a paid and organic last-click model that attributes 100% of key event value to the last channel clicked ^[2], and Amazon Associates operates under its own operating agreement ^[1], while other affiliate programs have their own separate agreements; splitting credit fairly between an ad, a video, an article and a coupon site is genuinely hard. When a reader clicks another link and buys there, they made a choice. That is not theft; it is a customer using the internet.

What deserves your caution is building a whole plan on traffic that structurally loses the last click. Three habits reduce the exposure, and none of them removes it:

• Write pages for the decision, not just the question. Honest reviews that name drawbacks, and comparisons that state a recommendation, are read by people about to buy. The anatomy of a useful product review and how to write a comparison post that actually helps readers choose both give you structures for that.
• Prefer programs whose window matches how long the purchase takes. If your readers typically research for weeks, a program with a very short window will keep crediting somebody else’s closing page.
• Earn a second click. If the program treats a later click as a new referral inside its window — and allows affiliate links in email — a reader who clicks again from your email is credited to your link; verify both points in the program terms. It is the most reliable way to shorten the gap between your recommendation and the purchase.

And about trusting a site like this one: we earn commissions from some links on this site, which is exactly why this page tells you to read each program’s own terms instead of taking our word for it. We cannot see your network reports, we have not audited every program’s current wording, and we are not going to pretend otherwise. What we can do is show you where the rules live and how to record them. Our FTC affiliate disclosure examples page covers how to disclose the same relationships on your own site.

Terms Change: How to Keep Your Notes Current

Amazon’s Associates Program Operating Agreement page, for example, displays "Updated: October 15, 2025" at the top ^[1]. A reminder that the document you read last spring may not be the one in force now. Amazon also reserves the right to change the nature, features, functions, scope or operation of its service offerings at any time ^[1], and it states that it may monitor and record information about your site’s users, including that a particular customer clicked through one of your links before buying ^[1]. If you write about Amazon, Amazon Associates rules beginners break most often covers the day-to-day compliance side.

Two habits keep you straight:

1. Record the last-updated date next to every rule you rely on, and recheck the page each quarter or before you refresh a post. Refreshing old affiliate posts gives you a checklist for that pass.
2. Separate what the program says from what you assumed. If you have not found the override rule, write "not stated on the terms page as of [date]" rather than guessing.

Commission income may have tax implications, and the treatment depends on your situation; affiliate income and US taxes is a plain-English starting point, and a qualified tax professional is the right person for your actual return.

Related guides: Keyword Research With Free Tools Only; Search Intent for Affiliate Content; 12 Clauses to Read in an Affiliate Program's Terms Before You Join; How to Judge a Product Before You Promote It.

This guide is general education. Program terms and tools change; check the linked official pages for the current version. How we write: editorial policy. This page has no affiliate links.