Google and Meta Both Claim the Sale: Who Really Earned It?

Run ads on Google and Meta for a month, then add up what each dashboard says it delivered. Google reports eighteen conversions. Meta reports twenty-two. Your lead sheet shows twenty-six bookings in total, and a good share of those came from referrals and repeat customers. The platforms are not lying, exactly. Each one is counting every sale it touched, and many sales were touched by both.

This matters because budget decisions get made from those dashboards. If you believe both numbers, you think ads produced forty bookings, the cost per booking looks wonderful, and you raise spend on the strength of sales that were counted twice or would have happened anyway.

Why both platforms claim the same customer

Think about how a real customer behaves. She sees your reel on Instagram on Monday and does nothing. On Thursday she searches for senior fares to Toronto, clicks your Google ad, and messages you. Meta counts that booking because she saw or clicked its ad within its attribution window, which by default reaches back several days for a click and a day for a view. Google counts it because she clicked its ad before converting, within a window that is typically weeks long. One customer, one booking, two platforms each reporting a conversion.

Neither platform can see the other's ads, and neither has any reason to share credit. Each reports the world as seen from inside its own walls.

The view-through problem

The most generous counting is the view-through conversion: someone was shown your ad, did not click, and later converted some other way. Meta includes one-day views in its default setting. For a brand people search for anyway, this inflates results considerably, because the ad is credited with customers who were already on their way to you. In Ads Manager you can compare attribution settings and look at click-only results. The number will be smaller and closer to the truth.

Your lead sheet is the referee

The platforms count conversions. Only you can count bookings, and you already have the tool: the lead sheet with its source column, filled in by asking every customer how they found you. That answer is imperfect, since people forget or say "online", but it is the only count in which each booking appears once. Use it as the total, and treat the platform figures as clues to how that total was produced.

Once a month, set three numbers side by side: bookings the sheet attributes to ads, conversions Google claims, conversions Meta claims. If the platforms together claim far more than the sheet shows, you know the size of the overlap, and you know not to add their figures together again.

Deciding where the next rupee goes

Attribution arguments are really budget arguments, so test the budget directly. The cleanest method for a small business is a holdout: reduce or pause one channel for two to three weeks, change nothing else, and watch total bookings on the sheet. If bookings hold steady, that channel was claiming credit it had not earned. If they fall by about what it claimed, it was earning its keep. Do this outside your busiest season, note the dates, and test one channel at a time.

In practice the two often work together: Meta creates the awareness and Google catches the search that follows. A holdout shows you that too, because pausing Meta will slowly thin out your branded searches on Google a few weeks later.

What not to conclude

Do not decide that tracking is pointless. Platform conversion data is still what trains the bidding, and the earlier tracking article explains why that matters. Do not pick one platform's numbers as the truth and dismiss the other's. And do not judge either channel on a single week. The point is narrower: platform reports are for steering each platform, and the lead sheet is for judging the business.

FAQ

Should I change Meta's attribution setting to click-only?

For reading results, comparing the click-only view is useful and more honest. Changing the setting on a live campaign also changes what it optimises towards and can restart learning, so use the comparison view first and change the setting only on a new campaign if you decide to.

Can Google Analytics settle the argument?

It gives one consistent view across channels, which helps, but it credits by its own rules and misses WhatsApp and phone conversations that never touch the website. It is a third opinion, not a verdict. The lead sheet still decides.

We only get a few bookings a month. Is a holdout test worth it?

With very small numbers a two-week test cannot tell signal from luck. Run it for a full month, or compare the same month a year apart, and rely more heavily on simply asking each customer how they found you.