Leads & Tracking
Meta Ads Attribution -- How It Works and Why Numbers Look Inflated
Meta's attribution model is one of the most misunderstood parts of the platform. Understanding it does not just explain the discrepancy with your CRM -- it changes how you evaluate campaign performance.
What is actually happening
Meta attributes a conversion to an ad if the user clicked the ad in the last 7 days (default) or viewed it in the last 1 day (view-through attribution, enabled by default). This means: someone who clicked your ad on Monday and then found you via Google on Wednesday and converted -- that conversion is attributed to Meta. The click happened; it was in the attribution window; Meta gets credit. This is not fraud -- it is a different attribution model than last-click. It answers the question: 'did this ad play a role in this conversion?' rather than 'was this ad the last touchpoint?' A second inflation source: Meta's 7-day click window counts all clicks, including non-outbound clicks -- likes, comments, and profile taps. Someone who liked your ad does not necessarily visit your site, but they entered the attribution window. Post-iOS 14, Meta also uses modelled conversions (statistical projection of conversions it cannot observe directly) which adds an additional layer of inflation versus CRM.
⚑ Most common wrong move
Optimising campaigns based purely on Meta-reported ROAS without cross-referencing CRM or backend data. Meta ROAS will almost always be higher than actual ROAS. Using it as the sole optimisation metric leads to over-investing in channels that are less effective than Meta claims.
What to do
1Set your attribution window explicitly. Ads Manager → Campaign → Attribution Settings. Switch from default '7-day click, 1-day view' to '7-day click only' to remove view-through inflation.
2Switch reporting to 'First Conversion' rather than 'All Conversions.' This removes inflation from repeat purchases being attributed to the original ad click.
3Add UTM parameters to all Meta ads. Build a first-touch field in your CRM alongside last-touch. Compare Meta-reported conversions against CRM first-touch over 30 days to understand your normal gap.
4Expect a 20-50% gap between Meta-reported conversions and CRM last-click conversions. This is normal. A gap above 100% suggests deduplication problems or view-through attribution inflating heavily.
5Use Meta numbers for relative optimisation (which ad is performing better vs worse) and CRM numbers for absolute business decisions (total revenue, true CAC).
Go deeper in the tool
◈ Diagnose
Run a full diagnosis with ranked root causes and numbered actions.
Pre-select: Meta results vs CRM don't match
Run Full Diagnosis →
◈ Full Diagnosis
Not sure what is wrong? Select all your symptoms and get ranked root causes.
Works across performance, creative, budget, targeting, and tracking
Run Diagnosis →
Related situations
Take it further
30-Minute Account Review -- €199
If the tool does not solve it, book a call. We go through your specific account, identify what is actually wrong, and leave you with a prioritised action list. No pitch, no retainer.