How the attribution setting works
When someone sees or clicks a Meta ad and later buys or enquires, something has to decide whether that outcome belongs to the ad. The attribution setting is that rule. It is built from two questions: how long after a click a conversion still counts, which is the attribution window, and whether a view — an ad shown but never clicked — counts at all, and for how long afterwards.
Meta’s default has for a long time paired a click window of several days with a view window of a single day, and the alternatives shorten or remove one or the other. The setting is chosen per ad set and it does two jobs at once. It decides what appears in your reports, and it defines the outcome the delivery system is trying to produce. A conversion falling outside the window is not credited, so as far as optimisation is concerned it never happened.
Why the attribution setting matters
It explains most arguments about whether Meta ads work. Ads Manager reports more conversions on a long click window with views included than on a short click-only window, from identical spend and identical real sales. Neither figure is dishonest; they answer different questions.
It matters again because your other systems use different rules. GA4 credits by its own model over its own window and generally ignores views altogether, so its Meta figure sits lower almost by construction. Expecting the two to agree wastes a great deal of meeting time; knowing why they differ is part of any honest analytics and tracking setup.
Common mistakes with the attribution setting
Changing it mid-campaign and then comparing months is the classic one. Performance appears to jump or collapse when nothing about the advertising moved at all. If a change is genuinely needed, record the date and treat the results as two separate series rather than one trend.
Reading view-through credit as if it were a click is the second. A view means the ad was on screen; it does not mean it caused anything. For a business with strong existing demand, a generous view window quietly claims sales that would have arrived anyway. The third is comparing ad sets that sit on different settings and drawing a conclusion from the gap between them.
How to act on it
Pick one setting, write it down and keep it. Judge every change in advertising against the same rule, rather than against whichever rule makes this month look better. If you want a cautious view of what the ads produced, read a click-only window. If you want the fuller picture, read the wider one. Holding both in mind is more useful than switching between them when the numbers disappoint.
Then check the thing that matters more: whether Meta’s count and your own record of sales or enquiries move in the same direction over time. If the platform reports growth your accounts cannot see, the attribution setting is usually where the gap begins, and that kind of data discrepancy deserves an explanation before anyone commits more budget.