Between a user’s click and the commission in your dashboard there is an invisible chain: tracking. If that chain breaks, the sale still happens but it does not end up in your name, and the money simply never shows. Understanding how your traffic gets attributed stops being a technicality the moment you realize it decides how much you get paid.
This article explains how attribution works in affiliate marketing: what cookies do, what a postback is, what the attribution window is, why cross-device makes attribution harder, and how to recognize tracking you can trust.
What affiliate tracking is
Tracking is the system that connects one of your clicks with the action that user takes afterward. When someone clicks your link, the system records that this visitor arrived through you, usually with a unique identifier. If they later sign up or buy, the system retrieves that identifier and attributes the conversion to you. Without that thread, the platform has no way of knowing whom to pay.
Your entire commission depends on that record being created correctly, surviving long enough, and being read at the moment of conversion. When any of that fails, the conversion is lost with no visible trace.
Cookies: what they are and what they store
A cookie is a small file the site saves in the user’s browser when they click your link. It does not store your commission or sensitive data: it stores an identifier that says, in essence, that this visitor arrived through a specific affiliate, along with the date. When the user converts, the site reads that cookie and knows the sale is yours.
Cookies live in the specific browser and device where the click happened. If the user clears them, switches browsers, or uses private mode, that cookie can disappear, and with it any attribution based on cookies alone.
The attribution window
Almost nobody buys on the first click. The attribution window is the period during which, after that click, a conversion still counts as yours. If it is thirty days and the user buys on day twenty, you get paid; if they buy on day forty, you do not.
The longer the window, the more late conversions get attributed to you. It is one of the figures most worth checking in a program, because a short window penalizes traffic that needs time to decide.
Postback and server-to-server (S2S)
The postback, or server-to-server (S2S) tracking, is a form of attribution that does not depend on the user’s browser. Instead of reading a cookie, the platform’s server notifies the tracking system directly that a conversion has happened, sending the identifier that was stored on the click.
Because the communication runs server to server, it is not affected by cookie clearing, blockers, or private mode. That is why the postback is considered the most reliable method today, and many serious programs offer it alongside cookies.
Client vs server: why it matters
Cookie tracking is client-side: it lives in the browser. The postback is server-side: it lives out of the user’s reach. Browsers have spent years cutting the lifespan of third-party cookies and tightening privacy, so tracking that depends on cookies alone loses conversions over time.
The ideal is a combination: cookies for the normal case and a postback as a safety net and source of truth. When both agree, attribution is solid.
- Where it lives — Cookies: in the user’s browser. Postback: on the server, outside the browser.
- Affected by clearing cookies — Cookies: yes. Postback: no.
- Private mode and blockers — Cookies: can fail. Postback: barely affected.
- Crosses devices — Cookies: not on their own. Postback: only with a shared login.
- Reliability today — Cookies: medium and declining. Postback: high.
The cross-device problem
A user sees your link on their phone but buys at night from a laptop. To classic tracking these are two different devices, with different cookies, and the sale can be lost even though you generated the click. That is cross-device.
Solving it fully is hard unless the user logs into the same account on both devices. Some systems reconstruct the journey when there is a login; others do not. It is one of the reasons no attribution is perfect, and one of the reasons tracking that combines signals is worth having.
Attribution models
When several clicks are involved, someone has to decide which one earns the commission. The most common model in affiliate marketing is last click: whoever sent the last click before the conversion gets paid. Others exist, such as first click or splitting credit across several touchpoints, but last click is still the standard for its simplicity.
Knowing which model a program uses tells you how ties are resolved when the user passed through several sources before buying.
Why reliable tracking hits your pocket
Every unattributed conversion is money you made the platform earn and never collected. Weak tracking throws no visible errors: it simply pays you less, and it is nearly impossible to spot from the outside because you never see the sales you lost. That is why tracking reliability weighs as much as the commission percentage.
One extra point of commission is worth nothing if the system loses one in five conversions along the way. So before you compare percentages across programs, it pays to understand how each one counts what you send it and what assurances it gives that the count is correct.
When your numbers do not match the program’s
It is normal for your stats and the program’s not to match to the cent. You count clicks from your side; the program counts conversions from its own, after filtering fraud, duplicates, and actions that do not meet the requirements. A small gap is expected. A large, constant gap, on the other hand, is a sign that something in the attribution is not working.
When the difference is striking, check your side first: malformed links, redirects that drop parameters, or subIDs that do not propagate. If your side is clean and the gap remains, that is the moment to ask the program how it counts and what it is discarding.
How to recognize reliable tracking
- It offers postback / S2S alongside cookies, not cookies alone.
- It states the attribution window and the model clearly, usually last click.
- It has a dashboard with stats in something close to real time, not figures that appear days later with no detail.
- It allows your own parameters (subIDs) so you can identify campaign, source, or creative.
- It is transparent about what counts as a valid conversion and when it confirms.
- Its numbers add up when you run a controlled test: one click of yours and a test conversion should appear.
Common mistakes
- Looking only at the commission percentage and not at how attribution works.
- Relying on cookie-only tracking and never noticing the conversions you lose.
- Not using subIDs and being left without knowing which campaign performs.
- Assuming a long attribution window without checking it.
- Never running a test conversion to confirm the system attributes to you.
FAQ
If the user clears cookies, do I lose the commission?
With cookie-only tracking, it is possible. With postback / S2S, the conversion can still be attributed because it does not depend on the browser. That is why a program that combines both is worth having.
What is a reasonable attribution window?
There is no single number; it depends on the vertical and the buying cycle. What matters is that the program states it clearly and that it fits how long your traffic takes to convert.
Does incognito mode break tracking?
It can affect cookies, because they are cleared when the session closes. Server-side postback is far less sensitive to it.
What are subIDs and what are they for?
They are parameters you add to your link to tag traffic by campaign, source, or creative. They let you see what works without lumping everything into a single figure.
Can I check the tracking myself?
Yes. Run a click with your own parameters and a test conversion, then check whether they show up correctly attributed in the dashboard. It is the simplest way to build confidence before scaling.
Want to be sure every click ends up attributed, and paid, the way it should? Tell us how you move your traffic and we’ll point you to the right fit. We run RevShare, PPL and PPS so you pick the model that suits you. Reach us on Telegram @amateurtvcash or drop by amateur.cash/en.