How Meta's attribution works by default
By default, Meta uses a 7-day click + 1-day view attribution window. This means: if someone sees your ad (but doesn't click) and then buys within 24 hours, Meta takes credit. If someone clicks your ad and buys up to 7 days later — even after clicking a Google ad and typing your brand name directly — Meta takes full credit.
This isn't a bug. It's the default. And it systematically inflates your reported ROAS.
The attribution overlap problem
Suppose a customer sees your Meta ad on Tuesday, clicks a Google Shopping ad on Thursday, and buys on Saturday. Meta reports a conversion (7-day click window). Google reports a conversion (last-click). Your Shopify shows one order.
Your combined platform-reported revenue: 2× your actual revenue. Your actual ROAS: roughly half of what the platforms tell you.
This overlap is why your MER (total revenue ÷ total spend) is almost always lower than the average of your per-platform ROAS figures.
How to get closer to reality in Meta Ads Manager
Switch to 7-day click only attribution: Remove the 1-day view window. View-through conversions are notoriously unreliable and inflate ROAS significantly.
Compare to your 1-day click window: This is the most conservative and most accurate attribution window. Campaigns that look profitable on 7-day click but terrible on 1-day click are likely benefiting from organic intent.
Use the Meta Conversions API alongside the pixel: The CAPI reduces signal loss from iOS 14+ privacy changes and gives you more accurate event matching. Brands using CAPI report 15–30% more matched events.
Run incrementality tests: Meta's Conversion Lift studies measure the true incremental lift of your ads versus a holdout group. This is the gold standard for measuring real campaign impact.
The framework that actually works in 2026
Use MER as your primary decision metric for budget allocation. Use 1-day click ROAS for campaign-level creative and targeting decisions. Use Conversion Lift for major budget scaling decisions.
When MER is trending up as you increase Meta spend, your campaigns are generating incremental revenue. When MER stays flat or drops, you're moving the curve but not growing.
Dayla displays your MER trend alongside your per-platform ROAS, so you can instantly see when a platform's reported performance diverges from your actual business results.
Building a measurement stack that works in 2026
The honest answer is that perfect attribution no longer exists. Post-iOS 14, multi-device journeys, and cross-channel overlap mean you will never have a complete picture of every conversion path.
What you can have is a measurement stack that's consistent and directionally correct. Use MER as your weekly north star. Use 1-day-click ROAS for intra-platform creative decisions. Use holdout tests quarterly to validate channel contribution. And use first-party data (email open rates, repeat purchase cohorts) to validate whether your customers are actually engaging after acquisition.
Suis ton vrai profit avec Dayla
Arrête de piloter ton e-commerce à l'aveugle. Dayla connecte tes données et calcule ton profit net réel en temps réel.