Facebook ads attribution,
explained for 2026.

Someone sees your ad on Monday. They click it on Tuesday, leave, and buy on Friday. Did the ad cause that sale? Attribution is the rule that answers yes or no.

Meta calls the time limit a window. A 7-day click window means Meta counts a purchase if it happens within 7 days of someone clicking your ad. A 1-day view window means Meta counts a purchase made within a day of someone seeing the ad, even without a click.

You choose the setting for each ad set. It changes two things. It changes which results you see in Ads Manager. It also changes what Meta learns from, because Meta tries to get you more of the results it can count.

Two changes landed this year, and both can make your numbers look different from 2025.

On 12 January 2026, Meta’s reporting API stopped returning the 7-day view and 28-day view windows. Meta said this matches what Ads Manager already shows. If a dashboard or spreadsheet still asks for those windows, its view columns go blank or change.

On 3 March 2026, Meta said it would change what a “click” means. Before, a like, a share or a save counted as a click. Now click-through counts only link clicks. The other actions moved into engage-through, which Meta renamed from engaged-view. The engaged view for video also dropped from 10 seconds to 5.

Meta said there would be no change to how you are billed. But your click-through results can drop, while engage-through results rise. So if your ROAS moved in spring 2026, check the setting before you blame the ads.

Enter what Ads Manager shows on your normal setting and on 7-day click alone, then what your shop recorded for the same dates. The checker shows three ROAS figures side by side, how much of Meta’s count comes from views and engagement, and what to check next. Nothing you enter leaves this page.

Almost every business we audit has the same question. Why does Meta show 100 sales when the shop shows fewer from Facebook? There is usually no single bug. Several small reasons add up.

Consent is a big reason in Malta. Under regulation 5 of S.L. 586.01, a website may only store tracking cookies after the visitor agrees. The IDPC’s guidance adds that pre-ticked boxes, scrolling and cookie walls do not count as consent. So the Meta Pixel and your analytics miss people who say no. That is the law working, not a fault.

The Conversions API sends sales from your server as well as the browser. It closes part of the gap, inside what your consent setup allows. Our Meta Pixel and Conversions API guide shows how to set both up and remove duplicates.

For most Malta online shops, we optimise on 7-day click and 1-day view, as Meta suggests by default. Then we judge results three ways: Meta’s figure, the click-only figure and the shop’s own orders. Our guide to what a good ROAS is shows how to find your break-even point.

For lead campaigns, the real check sits outside Meta. We count enquiries that turn into quotes or bookings in your CRM, or in a simple sheet. Tagging every ad link with UTM parameters makes that easy.

1. Fix the tracking first. Pixel, Conversions API and a consent banner that works.
2. Pick one setting per account. Keep it the same, so this month compares with last month.
3. Report three numbers. Meta’s figure, click-only and your shop or CRM.
4. Test before you scale. When the numbers disagree, we hold back part of the audience and compare.

Malta Digital manages Facebook and Instagram ads from €1,249 a month plus ad spend, paid to Meta in your own ad account. Tracking setup and monthly reports are part of the service.