Attribution Windows Explained: What the Setting Really Does
Seven day click, one day view, twenty eight days: the window setting quietly decides which ads get credit. What windows do and how to choose one.
The attribution window is the look back period in which a touchpoint can claim a sale. Change it and the same campaigns appear to perform differently overnight, with no change in a single customer or order. Few settings in a marketing account have that much quiet power.
Choosing a window should start from how customers actually buy: how long they consider, how often they return, and which channels start journeys versus close them. The default in the platform is a starting point, not an answer.
Longer windows flatter prospecting, shorter windows flatter closers
A long window lets early touches, video views, first visits, still claim credit weeks later, which lifts prospecting channels. A short window concentrates credit on the final touches before purchase, which lifts retargeting and branded search.
Neither is the truth. Each is a rule for dividing credit, and teams argue about channel performance when they are really arguing about the rule. Naming that openly improves every budget conversation.
Match the window to the consideration cycle
For considered purchases, furniture, appliances, high value clothing, shoppers research for weeks, and a one day window erases most of the journey. For impulse, low cost items, a twenty eight day window credits ads the customer barely remembers.
Use your own data: time from first visit to first order, and repeat visit patterns. Set the window to cover the typical consideration period, then leave it alone long enough for comparisons to mean something.
Report the window with every number
Every attributed figure shared internally or with partners should carry its window with it, because a return figure at twenty eight day click is not comparable to one at seven. Many cross team disputes dissolve once windows are printed beside the numbers.
When you do change a window, restate history under the new rule where the platform allows, and mark the change date in reports. Otherwise the team will celebrate or panic over a jump that was only a settings change.
Related reading
If this topic is useful, these earlier guides go deeper on connected questions: Attribution for Subscription Renewals and Reorders, Measuring the Offline and Retail Halo of Online Ads, and Attribution When Loyalty Points Pay for the Order.
Frequently asked questions
What is an attribution window in plain terms?
The number of days after an ad interaction during which a purchase can still be credited to that ad. It is a credit dividing rule, not a measurement of what a customer was thinking.
Does a longer window make campaigns perform better?
It makes them look better in reports by giving early touches more time to claim sales. Actual orders do not change; only the division of credit does, which is why window changes must be announced.
How should a team choose its window?
From its own consideration data: the typical time between first visit and first order for its products. Cover that period, then keep the setting stable so period comparisons stay meaningful.
Why print the window beside reported figures?
Because figures from different windows are not comparable, and much internal argument about channel performance is really unnoticed window mismatch between two reports.