Why Meta, Google, and TikTok ROAS Never Agree (and What Actually Tells You the Truth)

Open three ad accounts on the same day and you get three different stories. Meta says 4x. Google says 5x. TikTok says 1.4x.

None of these numbers measures the same thing, because none of these platforms plays by the same rules, and in 2026 all three rulebooks changed within a few months of each other.

This is not an old, settled measurement quirk. Meta removed its longest view-through windows in January.

It redefined what counts as a click in March. Google shifted its default attribution model in April. TikTok keeps splitting its reporting between Shop orders and website conversions.

Every one of those changes moves the number on your dashboard without moving a single real dollar of revenue.

Meta’s ROAS keeps shifting because Meta keeps changing what counts as a conversion

On January 12, 2026, Meta removed the 7-day view and 28-day view attribution windows from Ads Manager, leaving 1-day view as the longest window most advertisers can still select, according to Jetfuel’s 2026 attribution breakdown.

If view-through conversions made up more than a quarter of your reported total before that change, your old ROAS was almost certainly inflated.

View-through credits a purchase to an ad someone scrolled past without clicking, and a lot of that credit belongs to organic, email, or direct traffic instead.

Then in March, Meta redefined what counts as a click and rolled out engage-through attribution for video and social interactions on top of it, per GrowWithSakib’s guide to Meta’s 2026 changes.

Reported ROAS dropped for some accounts and rose for others, and in both directions the change had nothing to do with how the campaigns actually performed.

On average, Meta’s reported ROAS still runs about 28 percent above what independent measurement shows, according to Mako Metrics.

Google’s window is longer, and it just got recalibrated too

Google Ads defaults to a 30-day click and 1-day view attribution window, four times longer than Meta’s default click window, notes Improvado’s 2026 PPC ROAS guide.

The same campaign can show 2x ROAS under a 1-day window, 5x under a 7-day window, and 8x under a 30-day window, with zero change in actual sales. A longer window simply gives Google more time to claim a sale that a customer might have made anyway.

Google added another layer of movement in April 2026, when a GA4 update reweighted how credit gets distributed across upper-funnel touchpoints in the default attribution model, as Groas.ai documented at the time.

Reported conversion values and ROAS can shift even when nothing in the campaigns themselves has changed. If you are benchmarking this quarter against last quarter, you are partly benchmarking two different measurement systems.

TikTok’s number tends to run the opposite direction

Meta and Google mostly overstate. TikTok often does the reverse.

TikTok for Business research, cited by TikAdTools, found the platform drove 788 percent more conversions than last-click attribution actually captured, because TikTok sits earlier in the funnel: someone watches a product video, gets interested, then searches for the brand later or buys through another channel entirely.

Last-click measurement misses that path completely, so TikTok’s reported ROAS routinely looks weaker than its real contribution.

At the same time, TikTok’s default 7-day click plus 1-day view window can inflate ROAS for brands with a large repeat-purchase base, since a customer who saw a retargeting ad and then made an unrelated organic repeat purchase within 24 hours still gets counted as a TikTok conversion, according to Cartlyzer’s TikTok attribution breakdown.

So the same platform can be quietly overstating results for a loyalty-heavy brand and understating results for a prospecting-heavy one, depending entirely on account settings and customer mix.

Stacking three broken numbers doesn’t produce a true one

Add up the revenue implied by Meta’s ROAS, Google’s ROAS, and TikTok’s ROAS, and the total usually runs higher than the revenue that actually landed in the bank.

Each platform claims credit under its own rulebook, and those rulebooks overlap. If Meta reports 5x and your blended MER is 2.5x, your channels are claiming credit for sales they did not cause on their own.

That gap is not a spreadsheet error, it is three different measurement systems disagreeing with each other and with reality at the same time.

What to actually do about it: track blended MER, not stacked ROAS

Marketing Efficiency Ratio sidesteps the credit fight entirely.

MER is total revenue divided by total marketing spend across every channel, calculated off your own backend numbers rather than any platform’s pixel.

It does not ask which platform deserves the credit. It asks whether the money going out is coming back. Read a full breakdown of the formula and how to track it in our MER guide.

Healthy blended MER is stage-dependent, not a single universal target. Brands doing $1 million to $5 million a year typically run a blended MER of 1.5x to 2.5x and often break even or lose money on the first order.

$5 million to $10 million brands run 2.5x to 3.5x. $10 million to $25 million brands run 3x to 4.5x. Brands above $25 million typically run 3.5x to 6x or higher, according to Eightx’s 2026 DTC benchmarks.

Track your own MER weekly for four to eight weeks before you compare it to any of these ranges, since a single week of noisy data will not tell you much.

None of this means channel ROAS is useless.

It is still the right tool for same-account, same-window tactical calls: which ad set to kill, which creative to scale, which audience is fatiguing. Use it inside a channel where the window and the platform’s own rules stay constant.

Use MER for the decision that actually matters at budget time: whether the whole engine is worth what you are spending. For a deeper look at reconciling channel-level numbers before they roll up into MER, see our guide to stitching Meta, Google, and TikTok into one source of truth.

Where AdBeacon fits into this

AdBeacon exists because platforms grading their own homework is not a fringe problem. On one real AdBeacon account, Meta reported 3.23x ROAS.

Independently measured, first-party, click-only attribution put the true number at 0.93x. That is the gap this article has been describing, made concrete on a single account. AdBeacon does not claim to be the one absolute truth either, since no attribution methodology is perfect.

It claims to be closer to it than a platform reporting on its own performance, because it tracks every click, every conversion, and every product sold against your own store data instead of a platform’s self-interested model.

For more on where Meta’s own numbers specifically drift, see our breakdown of Meta’s 2026 attribution changes, and for the Google side, our guide to where Google Ads and GA4 lose your conversion data.

The takeaway is not that Meta, Google, and TikTok are lying to you. It is that each one is measuring a different, narrower thing than “did this make me money,” and none of them can see what the other two are doing.

Blended MER, checked against your bank account, is the number that can. If you want to see what independent, cross-channel attribution actually looks like on your own account, book a live AdBeacon demo.

FAQ

What is the ROAS inflation gap?

It is the difference between the ROAS a platform reports and the ROAS your business actually achieved, once you strip out modeled conversions, view-through credit, and cross-channel double counting.

Why does Meta’s ROAS look different from Google’s ROAS on the same spend?

Meta and Google use different attribution windows, different click definitions, and different modeling for unverifiable users, so the same customer journey gets measured two different ways.

Does TikTok underreport its ROAS?

Often, yes. TikTok sits earlier in the funnel, so a lot of its influence shows up later as a Google search or a direct visit that last-click attribution credits to a different channel.

What is blended MER and how is it different from ROAS?

MER is total revenue divided by total marketing spend across every channel, based on your own backend data. ROAS is channel-specific and depends on that channel’s attribution rules.

What is a healthy MER for an ecommerce brand?

It depends on revenue stage and margin. Most established DTC brands run somewhere between 2.5x and 6x, with larger, higher-margin brands sitting at the top of that range.