Why ad platforms claim more sales than you made

Why ad platforms over-count conversions: overlapping attribution windows, view-through credit and double counting, plus how to check claims against real sales.

By the Matrix ROI team at Smith App StudioOctober 18, 20263 min read

The short answer

Ad platforms over-count because each one grades its own work. Google, Meta and TikTok each look back over their own attribution window and claim any sale that followed a click or even a view of their ad. When a buyer touched two platforms, both take full credit, so the claims add up to more than your real orders.

Each platform counts a sale if its ad came anywhere in the buyer's recent path, so two platforms can claim the same order.
Default windows differ: Google Ads looks back 30 days after a click, Meta 7 days after a click and 1 day after a view.
View-through credit and modeled conversions add sales that no one can trace to a click.
The fix is a single source of truth: compare every claim against payments you actually received.

Default Google Ads click-through conversion window for a new conversion action: 30 days; Default Google Ads click-through conversion window for a new conversion action; Google Ads Help
Meta's default attribution window for reported actions: 7d click + 1d view; Meta's default attribution window for reported actions; Meta for Developers
Facebook ad experiments in a 2019 study where common observational methods often failed to match the true effect: 15; Facebook ad experiments in a 2019 study where common observational methods often failed to match the true effect; Kellogg School of Management

Every platform grades its own homework

Add up the sales that Google Ads, Meta and TikTok report for last month. For most businesses that run on more than one platform, the total is bigger than the number of orders in your store. Nobody is lying, exactly. Each platform uses its own rules to decide which sales it helped cause, and none of them checks with the others before claiming credit.

The reporting tool inside each ad platform sees only its own ads. If a customer clicked a Google ad on Monday and a Meta ad on Thursday, then bought on Friday, both platforms see a click followed by a purchase. Both report one conversion. Your bank account sees one payment.

Overlapping attribution windows

An attribution window is the stretch of time after someone interacts with an ad during which a purchase still counts for that ad. The defaults are generous and they do not match. According to Google's help page on conversion windows, if you don't change the setting, a new Google Ads conversion uses a 30-day click window, and you can stretch it as long as 90 days depending on the conversion source. Meta's developer documentation lists its default reporting window as 7 days after a click plus 1 day after a view.

So a single buyer who saw ads on two platforms within a few weeks can easily sit inside both windows at once. The longer the windows, the more overlap. Our guide to attribution windows walks through how to pick lengths that fit how long your customers really take to decide.

One month of sales: what three ad platforms claimed vs. orders that happened

Google Ads claims62
Meta claims48
TikTok claims21
Total claimed131
Real orders (store)100

This is a worked example with made-up numbers: when buyers touch more than one platform, each claims the sale, so the claims add up to more than the real orders.

Source: Worked example

Credit for ads people only saw

Some platforms also count view-through conversions: sales from people who saw an ad but never clicked it. Google is careful here. Its conversion measurement guide says view-through conversions are kept out of the main Conversions column. Meta's default, by contrast, folds 1-day views into the main number.

A view is a weak signal. Someone may have scrolled past your ad for a second on the way to a friend's photo and then bought because a coworker recommended you. The platform still sees an impression followed by a purchase. We cover this in more depth in view-through conversions explained.

Sales that would have happened anyway

The deeper problem is that a click before a purchase does not prove the ad caused the purchase. People who were already going to buy are exactly the people most likely to search for you and click your ad. Two well-known studies show how big this gap can be:

These studies don't mean ads don't work. They mean the number a platform reports is a measure of ad contact, not a measure of extra sales your ads created.

A 2015 study of eBay's search ads, published in Econometrica, found that returns from paid search were a fraction of non-experimental estimates, and that brand keyword ads had no measurable short-term benefit.
A 2019 Marketing Science paper used 15 US advertising experiments at Facebook and found that common observational methods often fail to produce the same effects as randomized experiments.

How to see the real number

You can't make the platforms stop over-counting, but you can stop relying on their totals. A few habits help:

Matrix ROI, built by Smith App Studio, does this side-by-side view for you: what each platform claims next to what your Stripe payments or store exports verify, day by day. You can try the live demo to see how a claimed-versus-verified gap looks in practice.

Pick one source of truth for revenue, such as your payment processor or store, and treat platform numbers as claims to be checked.
Put each platform's claimed conversions beside your real orders for the same days. A rising gap is a warning sign even if each platform's report looks healthy.
Use the same attribution window across platforms when you compare them, so you are not mixing 30-day and 7-day numbers.
Tag every ad link with campaign parameters so your own analytics can see where buyers came from.
Run a holdout test now and then, where you pause ads for a region or a group, to measure what changes when the ads are gone.

Frequently asked questions

Why do Google and Meta both claim the same sale?

Each platform only sees its own ads. If a buyer clicked a Google ad and a Meta ad within each platform's attribution window and then bought, both platforms record a conversion. Neither checks with the other, so one order shows up twice. The only way to catch it is to compare the platforms' totals with the orders in your own store or payment records.

Is over-counting the same as ad fraud?

No. Over-counting usually comes from honest but generous rules: long attribution windows, credit for ads people only viewed, and modeled estimates for sales the platform can't track directly. Fraud means fake clicks or fake traffic. Both inflate results, but over-counting happens even in a perfectly clean account.

Should I shorten my attribution windows?

Shorter windows reduce overlap but can also undercount real sales if your customers take weeks to decide. A better approach is to choose windows that match your actual buying cycle, use the same window when comparing platforms, and always check totals against your real revenue.

Which number should I use to judge my ads?

Use revenue you can verify, such as payments in Stripe or orders in your store, as the base. Platform numbers are useful for comparing campaigns inside that platform, but they should not decide your total budget on their own. Check the two side by side every week.

Found this useful? Pass it on.

X LinkedIn Facebook Email

Sources

General information only, not legal, tax or financial advice. Figures belong to the sources linked above and were checked on the date shown. See our disclaimers.

Keep reading

AttributionAttribution windows explained (and how they change your numbers)
AttributionView-through conversions: what they are and why they inflate results
Agencies & reportingHow to reconcile ad platform data with your Stripe or store sales

See what your ad platforms claim next to what your records prove.

14-day free trial. No card needed to start.

Start free trial Live demo