ChatGPT
Claude
Gemini
Grok
Perplexity
When your ROAS says one thing and your MER says another, trust neither one on its own. Each answers a different question, so each has a different job: ROAS steers your campaigns, MER checks the whole business, and new-customer contribution margin settles whether your ads actually made money. This guide shows you which number to open for which decision, worked on one example month of a $50 product.
ROAS, return on ad spend, is the revenue one ad platform credits to itself divided by what you spent on that platform. MER, marketing efficiency ratio, is your total revenue divided by your total ad spend across every channel. When the two disagree, it is not a glitch. They are measuring different things, and knowing which to believe is the whole skill.
ROAS Is a Channel Number and MER Is a Business Number, Which Is Why They Disagree
ROAS is a channel metric. It lives inside one ad platform, it is that platform's own count of the sales it thinks it caused, and it is really a feedback signal for the algorithm. Give the campaign a clean, accurate ROAS and it optimizes better. That is what the number is for.
MER is a business metric. It divides your total revenue by your total ad spend, calculated from your Shopify revenue and your spend, outside any platform. It does not care which ad got the credit. It asks one thing: for every dollar you put into ads, how many dollars came back to the business.
So which to trust depends on the job in front of you. Trust ROAS to steer a campaign. Trust MER, alongside contribution margin, to judge whether the business is acquiring profitably.
| Metric | What it divides | Best for | What it hides |
|---|---|---|---|
| ROAS | Revenue one platform claims, over that platform's spend | Steering a campaign day to day | Double-counting, tracking errors, and all your costs |
| MER | Total business revenue, over total ad spend | A business-level efficiency check | Whether the ads caused the sales |
| New-customer contribution margin | New-customer profit after the cost of delivery, all ad spend on new | The profit verdict on acquisition | Nothing structural; it needs clean new-versus-repeat data |
ROAS Double-Counts Across Platforms, and MER Cannot
Buying is multi-touch now. A customer might see a Google ad, then a Facebook ad, run a search, ask an AI, and then buy. Every one of those platforms can claim the same sale. Add up the ROAS each platform reports and you have counted one order two or three times, so the total looks better than the business really did.
Google ad → Facebook ad → Google search → AI answer → Purchase
One sale. Every platform can claim it, so summed ROAS counts it more than once. MER counts it exactly once.
One measurement audit shows how far this runs: three platforms reporting 80, 65, and 30 conversions claim 175 sales between them, while the store's backend recorded only 95, an 84% overcount. Your own gap will differ, so reconcile the platforms against your Shopify orders rather than trusting the sum.
MER cannot double-count, because it divides all your revenue by all your spend. The same sale can only land in the numerator once. That is the core reason a business-level number is the safer one for the final verdict: it is arithmetic the platforms cannot inflate.
ROAS Is Only as Trustworthy as the Tracking Under It
ROAS is built from the pixel, and the pixel can be wrong. The formula is purchase revenue divided by ad spend, and the platform tracks two separate things that can each break: the purchase count and the purchase value.
MER is built from your Shopify revenue, the number you can reconcile against your bank. That makes it much harder to fool. So when your ROAS and your MER pull apart, the tracking is usually why, and the number you can tie to your payouts is the one to believe.
Neither Number Means Anything Until You Set Your Breakeven
Here is the trap that catches both metrics: a 4.0 ROAS or a 3.3 MER can be a losing number, and you cannot tell until you know your breakeven. Once COGS, shipping, returns, transaction fees, tools, and any agency retainer are in, a "good" 2x can be a losing 2x.
Your breakeven comes from your cost of delivery: everything it costs to fulfill one order. Start there, on a $50 product, a simple filter.
| Shipping | $5.00 |
| 3PL (fulfillment) | $2.00 |
| COGS | $10.00 |
| Transaction fees | $1.80 |
| Returns (at a 7% return rate) | $3.50 |
| Cost of delivery | $22.30 |
That leaves a contribution margin of $27.70 per order ($50.00 minus $22.30), which is also your breakeven CAC: the most you can pay to acquire a customer, or CAC, and still break even on the first order. Set a 20 percent margin goal and your target CAC is $17.70, the number to actually aim for.
Now watch the same month through all three metrics.
Illustrative example
| View | The number | What it says |
|---|---|---|
| Platform ROAS ($10K Meta spend, $40K reported) | 4.0x | Looks like a clear winner |
| Business MER ($50K revenue, $15K total spend) | 3.3x | Lower, the double-count removed |
| New-customer CAC (vs $27.70 breakeven) | $37.50 | Above breakeven, acquisition underwater |
- ROAS view (the platform): you spend $10,000 on Meta, Meta reports $40,000 in sales, so your platform ROAS is 4.0. It looks like a clear winner.
- MER view (the business): Shopify shows $50,000 in total revenue that month, and your total ad spend was $15,000 ($10,000 Meta plus $5,000 Google), so your MER is 3.3. Lower than 4.0, because the 4.0 was single-platform and double-counted.
- Verdict view (new customers): of that $50,000, about $30,000 came from repeat customers who likely would have bought anyway. New-customer revenue is roughly $20,000, about 400 new orders at the $50 price. Put all $15,000 of spend against acquisition and your CAC is $37.50, well above the $27.70 breakeven. Acquisition lost money on every new order, even though ROAS showed 4.0 and MER showed a healthy-looking 3.3.
The point the numbers make: ROAS steered the campaign, MER caught the double-count, and new-customer contribution margin delivered the verdict. Three numbers, three jobs.
MER Has Its Own Blind Spot: It Cannot Tell You the Ads Caused the Sales
MER is the safer top-line number, but do not trust it blindly either. MER divides all your revenue by your ad spend, and all your revenue includes organic and repeat orders the ads did not cause. A brand with strong repeat business can post a flattering MER while acquisition quietly loses money, exactly like the month above.
The fix is to read new-customer contribution margin right next to it. Split your customers new versus repeat, pull that data from Shopify, and attribute all of your ad spend to new customers.
| Contribution margin | New customers | Repeat customers |
|---|---|---|
| Revenue | From Shopify, new only | From Shopify, returning |
| Minus cost of delivery | Per order | Per order |
| Minus all ad spend | All of it here | None |
| Equals contribution | the number that matters | supporting only |
If the ads are mostly capturing rebuys, that is not the ads working. That is the ads taking credit for revenue you would likely have gotten anyway. Trust MER for the efficiency picture; trust new-customer contribution margin for whether acquisition actually pays. This is one part of the full check on whether your Facebook ads are working, and it rests on the contribution-margin math worked in full here.
Why the Platform and Some Agencies Show You ROAS Instead
The ad platform leads with ROAS because a big in-account number makes the platform look effective, which is the outcome it is built to optimize for. Some agencies lead with the same number for a related reason: the channel is what they are paid to manage, and ROAS is the metric that most flatters that channel. A reported 4x can still be losing money once the retainer, the tools, and all the ad spend are in.
None of that requires bad intent; it is the incentive structure. And the honest read cuts both ways: when you finally run MER and new-customer contribution margin, good work sometimes turns out to be paying after all. Ask your agency, or your own reporting, for the blended net, not just platform ROAS. Set that business-level number as the goal from the start, and verify your tracking yourself rather than assuming it is clean, the same reason your ROAS can drift from your bank balance in the first place.
Which to Trust, in Practice: Steer on ROAS, Check on MER, Judge on New Customers
Put the three numbers in their lanes and the decision gets simple.
Steer on ROAS
Use it as the day-to-day optimization signal inside the platform, and feed the algorithm clean data so it works.
Check on MER
Use it as the business-level efficiency read, and frame your goal as a blended breakeven CPA (cost per acquisition), not a ROAS target.
Judge on new-customer CM
Use new-customer contribution margin against your breakeven as the final word on whether acquisition pays.
How hard you split new from repeat depends on your spend.
Under $20K/month per channel
Keep it simple. Run one campaign with prospecting and retargeting together, judge on blended MER and contribution margin, and still calculate new-customer contribution margin as a conservative read with all spend on new. Splitting campaigns this early just fragments your data and slows learning.
Over $20K/month per channel
Separating new from existing earns its place, with new-customer conversion events, audiences, and exclusions.
A metric read too early tells you almost nothing either way, so give a campaign enough time and volume before you judge it on any of these numbers.
Do This Now to Put All Three Numbers in Their Lanes
- 01
Write down all three numbers for last month.
Your platform ROAS, your MER (total Shopify revenue divided by total ad spend), and your rough new-versus-repeat revenue split.
- 02
Build your breakeven.
Add COGS, shipping, the 3PL fee, transaction fees, and your return rate for one order to get your cost of delivery, subtract it from your price for your contribution margin, and that is your breakeven CAC.
- 03
Run the verdict.
Put all your ad spend against new customers only, divide by new orders for your real CAC, and compare it to your breakeven. That comparison, not ROAS, tells you if acquisition paid.
- 04
Set the business goal.
Make blended breakeven CPA the target you manage to, and check MER against your profit and loss statement every month.
Or hand these steps to an AI tool to run your own numbers:
AI Prompt
Copy this and paste it into ChatGPT or Claude. It will walk you through your own numbers.
You are helping me decide whether my ecommerce ads are profitable, comparing ROAS, MER, and new-customer contribution margin. Ask me, one at a time: my total revenue and total ad spend last month; my platform ROAS; my new-versus-repeat revenue split; my product price; and my COGS, shipping, 3PL fee, transaction fees, and return rate. Then calculate my MER, my cost of delivery, my contribution margin per order, my breakeven CAC, and my new-customer CAC with all spend attributed to new customers. Show each number and tell me whether acquisition paid.
See more profit-first playbooks in our insights library.
Common Questions
ROAS vs MER FAQ
Should I trust ROAS or MER?
Trust both, for different jobs. ROAS is a channel signal you use to steer a campaign day to day; MER is a business-level check on whether your whole ad program is efficient. Neither is the final word on profit. For that, read new-customer contribution margin against your breakeven.
Why is my MER lower than my ROAS?
Because they count different things. Your platform ROAS only sees the sales one platform claims, and platforms often claim the same order across channels, which inflates it. MER divides all your revenue by all your ad spend, so nothing gets double-counted and the number comes in lower and more honest. If the gap is large, weak tracking is usually part of it too.
What is a good MER for an ecommerce brand?
There is no universal number, and chasing one is how brands get burned. A good MER is one that clears your blended breakeven once COGS, shipping, returns, fees, tools, and any retainer are in. Work it from your own numbers, because a high-looking MER can still lose money while a lower one pays, depending on your cost of delivery. Set your target as a blended breakeven CPA, not a round-number MER goal.
Can I just use MER and ignore ROAS?
Not quite. MER is the safer business-level number, but it cannot tell you the ads caused the sales, because it counts organic and repeat revenue too. ROAS still earns its keep as the day-to-day signal that helps the platform's algorithm optimize, so you feed it clean data and steer with it. Use MER to check the business, ROAS to steer the campaign, and new-customer contribution margin to judge whether acquisition actually pays.
The Bottom Line on ROAS vs MER
Steer on ROAS, check on MER, and judge on new-customer contribution margin against your breakeven. ROAS is a channel signal that can double-count and drift with your tracking; MER is the safer business read but cannot tell you the ads caused the sales; new-customer contribution margin is the verdict. When the numbers disagree, that is not noise, it is each one doing its own job.
