Meta Attribution Settings, and Which One to Use
Changing the attribution window changes your reported results overnight. Here is what each setting means for your decisions.
Change one dropdown in Meta Ads Manager and your reported results change by thirty percent. Nothing about your business changed. Just the dropdown.
That dropdown is the attribution setting, and understanding it is the difference between reading your Meta reports and being led around by them.
What the setting does
It defines the window in which Meta will claim credit for a conversion after someone interacted with your ad.
Two kinds of interaction. A click, and a view, meaning the ad was served and the person did not click.
The options are combinations of the two. Seven day click. One day view. Seven day click and one day view. One day click. And so on.
The default and why it matters
The current default is seven day click and one day view.
So a conversion counts if the person clicked within seven days, or saw the ad within one day and converted without clicking.
That second half is where the arguments start. View through conversions are, by definition, people who did not click your ad. Meta says the ad influenced them. Sometimes true, sometimes the person was going to buy anyway and happened to scroll past a post.
Comparing settings honestly
Here is what typically happens to reported conversions as you tighten the window, on a normal ecommerce account.
Seven day click, one day view. The default. The most generous commonly used setting.
Seven day click. Removes view throughs. Reported conversions usually drop by ten to thirty percent.
One day click. Only same day clickers. Drops further, sometimes a lot. Closest to what a naive last click model would show.
None of these is correct in an absolute sense. They are different definitions, and your job is to pick one, use it consistently, and understand what it includes.
The comparison window feature
Meta lets you view up to three attribution settings side by side in the same report, as comparison columns.
Use this. It is the fastest way to understand your own account.
If seven day click and one day click show similar numbers, your customers decide quickly and the setting barely matters. If they differ by half, the setting matters enormously and you should think carefully about which one you report on.
Attribution setting versus bidding
An important distinction. The attribution setting affects both reporting and optimisation, but not equally.
Changing it at ad set level changes what the delivery system optimises toward. Changing it as a reporting comparison column only changes what you see.
So if you change the ad set attribution setting, expect a learning period. If you only change the comparison column, nothing about delivery changes.
People conflate these constantly and then panic about a performance change they did not cause.
Why Meta and your store will never agree
Because they count differently, in at least four ways.
Meta uses its own attribution, crediting itself for view throughs. Your store’s analytics almost certainly uses last click.
Meta credits the conversion to the day of the ad interaction. Your store records it on the day of the sale.
Meta deduplicates across devices using its own identity graph. Your analytics uses cookies.
Meta models some conversions it cannot directly observe.
So the correct expectation is that they will not match, and the useful question is whether the gap is stable. A stable gap you can work with. A gap that changes month to month means something broke.
The equivalent problem on the Google side: why GA4 and Google Ads conversions do not match.
What I would actually do
Leave ad sets on the default unless you have a specific reason to change, because the default is what Meta’s delivery system is tuned around.
For internal reporting, look at seven day click as your headline number, with view throughs visible but separate.
Track your blended figures alongside it. Total revenue divided by total ad spend is immune to attribution arguments.
Run an incrementality or geo test once or twice a year. That is the only thing that answers the underlying question.
That fourth point is the one that ends the debate, and it is more accessible than most people assume.
Do not change it every month
The worst outcome is an account where the attribution setting has been changed three times this year, because now no two periods are comparable and every trend line is meaningless.
Pick a setting, write down the date you picked it, and leave it alone. If you must change it, note the date prominently so future you knows why the chart has a step in it.
The underlying point
Attribution settings are a reporting convention, not a measurement of truth. Every setting is a defensible answer to a slightly different question.
What matters is that everyone reading the numbers knows which question is being answered, and that the answer stays the same from month to month.
Need this built properly?
Attribution settings decide what your reports say, which is not the same as what happened. I set up and repair tracking for ecommerce stores and lead generation businesses, so the numbers in your ad accounts match the money in your bank account.
Have a look at what I do, or book a call and we can go through your setup together.
