Attribution window: what it is and how window length changes credit

How does window length change who gets the credit?

A shopper sees a creator’s video on the 2nd, thinks about it, searches the brand on the 12th, and buys. Under a 7-day window the creator receives nothing and search takes the credit. Under a 30-day window the creator takes the sale and search looks considerably weaker.

Nothing about the customer changed between those two reports. The only thing that changed was a setting in a dashboard.

That is why comparing channel performance across two systems with different windows produces an argument rather than an answer. The wider set of failure modes, including the ones that survive any window choice, is in why your attribution is wrong.

Window length also interacts with which touch you credit, and the two decisions are usually made by different people. A long window combined with last-touch crediting systematically flatters whatever sits closest to checkout, normally branded search and retargeting, which is the comparison worked through in first-touch versus last-touch for D2C.

How do you choose an attribution window instead of inheriting one?

Most brands run whatever their ad platform shipped with. That is a decision, just an unexamined one, and it is usually wrong in a predictable direction.

Platform defaults tend to be generous to the platform reporting them. A window is a lever controlling how much credit a channel can claim, and in the default configuration the channel owns the lever.

The defensible method starts with your own order data rather than a benchmark. Measure the distribution of time between first identified contact and purchase, then set the window to cover the bulk of that distribution rather than its tail.

Model the cost of getting it wrong before changing anything. On stated assumptions where 35% of orders convert after the seventh day, a 7-day window moves that share of credit away from whatever created the demand, and budget follows the report within a quarter. The practical steps are in matching your attribution window to your buying cycle.

Write the chosen window down somewhere outside the ad platform. A setting that lives only inside a tool one team administers will eventually be changed by somebody optimising that tool. The report moves, and nobody records why.

Different channels can justify different windows, because decision lag genuinely differs between a retargeting click and a long-form creator video. That is defensible as long as the choice is written down and held stable, because changing windows and channel budgets in the same month makes the result unreadable.

Why do two dashboards report different numbers for the same week?

Because they are answering different questions with different rules. One system uses a 7-day click window, another attributes on last touch with a 30-day cookie, and a third counts only the orders it can see inside the store’s own database.

Summed platform-reported revenue can exceed real revenue for exactly this reason. Two platforms with different windows can each claim the same order, and neither of them is lying about what it observed.

Reconciliation is therefore not a matter of finding the correct number. It is a matter of writing down which question each report answers and refusing to compare across them, which is the approach in when Shopify, Meta and GA4 disagree.

Impression windows compound the problem. A platform that credits a view rather than a click can attribute an order to a channel the shopper never interacted with. View-through credit is rarely comparable across two vendors, so adding it to click-based numbers is worse than useless.

A window also cannot recover what was never observed in the first place. Cross-device journeys, cleared cookies, and links shared in private messages sit outside every window length you could choose, and lengthening the window does not surface any of them.

Where Korant fits

Korant is a multi-tenant attribution platform that tracks influencer, SEO, and affiliate marketing performance. Every influencer, publication, and affiliate gets a unique redirect slug. Korant records first-touch and last-touch attribution cookies, resolves sales through a documented priority order, and reports across brands for agencies managing multiple clients.

Because both cookies are recorded, the same order set can be scored first-touch and last-touch without re-instrumenting the site, which turns the window argument into something testable rather than theoretical. Ambiguous sales resolve through a fixed priority order in which a discount code attributed to an influencer outranks a last-touch cookie, so a code beats a click whenever both are present.

Korant is not the right tool for stores with a single paid channel, or for brands that only need Shopify’s native reports. If every order comes from one source, no window setting will tell you anything you did not already know, and the honest answer is to spend the setup time somewhere it changes a decision.