Meta Ads credits you with sales it didn't generate: how to measure real attribution

What is attribution and why does Meta inflate it?

Attribution is the process of giving a channel credit for a sale. If someone sees your ad on Meta on Monday and buys on Friday, who "caused" that sale? That's the question Meta always answers in its own favor.

By default, Meta uses a 7-day click and 1-day view attribution window. That means: if someone clicks your ad (or simply sees it) and buys within 7 days, Meta takes full credit. It doesn't matter if that person also searched for your brand on Google, came in through email, or went straight to your store because someone recommended it.

Result: Meta claims between 30% and 50% of sales that would have happened anyway — organic, direct, word of mouth. Your real ROAS can be 2x when Meta shows you 4x.

The attribution types Meta uses (and the ones it doesn't show you)

Inside Meta Ads Manager, there are several ways to measure:

• Last-click on Meta: the default model. It inflates ROAS because it takes credit for any click, even unintentional ones.
• View-through (1 day): Meta "gives you credit" if someone saw the ad and bought within 24 hours, even if they never interacted with it.
• Engage-through attribution: since 2026, it groups likes, shares and saves. Inflation in its purest form.
• Real incrementality: how many additional sales your ads generated vs. running no ads at all. Meta has a tool for this (Conversion Lift), but few people use it well.

Meta's main dashboard always shows the model that makes you feel good. Not the one that tells you the truth.

How do you calculate Meta Ads' real attribution?

To know how much Meta really sells, you need to cross Meta's data with your store's real data. The practical path:

• Blended ROAS: divide your total sales (for the whole store, not just what Meta attributes) by your total Meta spend. If you spend $200,000 and sell $800,000 in total, your Blended ROAS is 4x. If Meta shows you 7x, the difference is the overestimation.
• UTMs + GA4: tag all your campaigns with UTMs and check sessions and conversions in Google Analytics. The GA4 number will always be lower than Meta's — the difference is double counting.
• Incrementality test: pause a campaign for 2 weeks and see how much real sales dropped. If they barely drop, the campaign wasn't generating much.
• Compare against periods with no ads: if you have a history of organic sales, use it as your baseline.

Why doesn't your agency talk to you about incremental attribution?

Because its reports are Meta's data, and that data makes it look good. An agency that measures real incrementality has to be willing to show that 40% of what Meta attributes isn't to its credit — very few do that.

That doesn't mean your agency is dishonest. It means Meta's measurement system incentivizes it to show you what Meta wants you to see. Asking for real data is on you.

If you want to know whether your agency is generating real profit for you, here's how to audit it.

How Escalafy shows Meta's real attribution

Escalafy crosses Meta's data with your store's data (Tiendanube or Shopify) in real time. Instead of trusting what Meta says you sold, it looks at the real orders on your platform, matches them to campaigns, and calculates:

• Your real ROAS by campaign — real sales ÷ spend, not Meta's ROAS
• Your business's total Blended ROAS
• Which campaigns generate incremental sales vs. which ones just claim sales that would have happened anyway
• The real acquisition cost per customer, including every cost

With that, you can decide which campaigns to turn off, which to scale, and whether your Meta spend makes sense for your real margin. See how the real Meta Ads ROAS module in Escalafy works.

Frequently asked questions about Meta Ads attribution

Does Meta always overestimate conversions?

Not always, but it's the most common case. For e-commerce stores with a lot of organic or referral traffic, the overestimation can reach 50%. For stores that depend almost exclusively on ads, the problem is smaller because there are fewer organic sales for Meta to claim.

Which attribution window should I use in Meta?

For most e-commerce stores, "1-day click" is more representative than the default "7-day click." But even that overestimates. Blended ROAS is still your best source of truth.

How does attribution affect investment decisions?

If you think your ROAS is 5x when it's really 2.5x, you'll invest more in advertising that doesn't generate real profit. Many e-commerce stores scale their Meta spend and watch profitability drop — for exactly this reason.

Is the problem the same with Google Ads?

Google has attribution problems too, but the overestimation tends to be smaller because Google Search captures active intent (someone searched for your product), while Meta interrupts. The double-counting problem between Meta and Google does exist — both can claim the same sale.

Does Blended ROAS replace Meta's ROAS?

For business decisions, yes. Blended ROAS tells you whether your total ad spend is profitable for your real business. Meta's ROAS is useful for optimizing inside the platform. Use both: Meta's to tune campaigns, Blended to decide how much to spend in total.

Meta isn't going to fix this for you

Its business model is showing you numbers that make you spend more. Your job is to cross that data with the reality of your store.

If you want to understand your real ROAS in e-commerce, or know which campaigns actually generate profit, Escalafy does that cross-check automatically. No Excel, in real time.

Sources: Meta attribution changes in 2026 (Dataslayer) · Incremental attribution in Meta Ads (Dobuss)