MER against ROAS: why they disagree, and which to trust.

Updated 16 October 2026 · 4 min read

Ads Manager says your ROAS is 4. Google Ads says 6. Your Shopify sales over all your ad spend come to 2.8. Nobody has made a mistake. The numbers measure different things, and once you know what each one is for, the disagreement becomes useful rather than confusing.

The short answer: trust MER to tell you whether your advertising as a whole is making money, because it uses the sales that actually reached your bank. Use each platform's ROAS to compare ads and campaigns inside that platform, where it is consistent with itself. Never add platform ROAS figures together.

Two numbers, two questions

Platform ROAS = the revenue a platform says its ads produced ÷ that platform's spend. It answers: by its own rules, which of this platform's ads look best?
MER = all your store's net sales ÷ all your ad spend. It answers: is the business getting enough sales for what it spends on ads?

Our guide to what MER is covers the sum and how to set a target.

A week that doesn't add up

Take one week for a store spending on two platforms:

Meta: £2,000 spend, claims £8,000 of purchases. ROAS 4.0.
Google Ads: £1,000 spend, claims £6,000 of conversions. ROAS 6.0.
Shopify: £8,400 of net sales for the week, from every source.

The platforms claim £14,000 between them, but the store only took £8,400. Spend was £3,000, so MER = £8,400 ÷ £3,000 = 2.8. If this store breaks even at 2.2, the ads are working, but nowhere near as well as either platform suggests.

Why platform ROAS and Shopify disagree

The same order is claimed twice. A customer clicks a Meta ad on Monday and searches your name on Thursday, clicking your Google ad before buying. Meta counts the sale (a click within its window) and Google counts it too (an ad click before the purchase). Shopify counts it once.
Views count as well as clicks. Meta's usual setting gives an ad credit for purchases within 7 days of a click or 1 day of a view. Someone who scrolled past an ad and bought the next day through your newsletter still counts for Meta.
Modelled conversions. Where tracking is blocked, platforms estimate the purchases they think they caused. Estimates can be generous.
Different revenue. The purchase value a platform reports is whatever your store's tracking sends, often the full order total with VAT and delivery. Shopify net sales leave those out. See ROAS and VAT.
Different days. Google puts a conversion on the day of the click, so last week's figures keep growing for weeks. Each platform also uses its own time zone.
Brand search takes credit for demand it didn't create. People who already know your name and search for it would mostly have found you anyway, yet branded search often shows the highest ROAS of all.

It can go the other way too. Platforms miss sales they helped with but couldn't track: the TikTok that made someone remember you, the ad seen on a phone and bought on a laptop. For upper funnel platforms, ROAS often understates their part.

Which to trust, and when

Setting budgets and judging the month: MER. It can't be double counted, and it is the number that matches your bank account.
Choosing between ads in one platform: platform ROAS and cost per purchase. Every ad in Meta is measured by the same rules, so the comparison is fair even if the level is inflated.
Comparing platforms with each other: neither number alone. Each platform's ROAS uses different rules, so Google's 6 and Meta's 4 aren't comparable. Watch what happens to MER when you move money between them.
Spotting trouble: both. If platform ROAS holds steady while MER falls, the platforms are probably claiming sales that were coming anyway.

How to check your ads really add sales

The question underneath all of this is whether the ads cause sales that wouldn't happen otherwise. You can get closer to the answer without a data science team:

Watch MER when spend changes. When you raise a budget by a clear step and hold it for a week or two, do total sales rise with it? If spend goes up 30% and sales barely move, the extra spend isn't adding much.
Log every change. Note budget changes, new campaigns and paused ones with the date, so you can read the effect later instead of guessing. Our guide to testing a change shows how.
Try a pause test. Turning off a campaign for a week, such as branded search, and watching total sales tells you more than its ROAS ever will. Pick a quiet week, not a sale.
Use the platforms' own lift tests when your spend is big enough. Meta and Google both offer tests that hold back a group from seeing your ads and compare the two.

How That Other Dashboard shows both

That Other Dashboard puts spend, MER, cost per order and each platform's own ROAS in one panel, so you can see the gap rather than wonder about it. MER is your Shopify net sales over the spend of every ad platform you've connected. Platform ROAS is the revenue each platform reports over its own spend, labelled as such, because it flatters. Every ad sits in one table underneath, where platform ROAS does its proper job: comparing one ad with another.

Common questions

Is a big gap between ROAS and MER bad?

Not by itself. Some gap is normal because platforms overlap and count views. What matters is that MER clears your break even, and that the gap doesn't keep widening.

Should I turn off ads with a low ROAS?

Compare them with other ads on the same platform first, over the same days. A low ROAS ad that reaches new people can still be lifting MER, so check total sales before and after you pause it.

Why is my Google Ads ROAS so much higher than Meta's?

Often because Google captures people who already meant to buy, especially on branded search and Shopping, while Meta finds people who hadn't heard of you. Each platform measures differently, so compare them through MER, not their own ROAS.