Most ad dashboards hand you a ROAS number and invite you to feel good or bad about it. But ROAS on its own can't tell you whether you're making money, and it can't tell you whether you're actually growing.
Four numbers together give you a much clearer picture. Here's what each one means, how to calculate it and which decisions it should drive. Every example uses a fictional skincare brand, Fernway, and made-up round numbers for a single month.
ROAS: return on ad spend
ROAS = revenue attributed to ads ÷ ad spend.
Fernway spends $10,000 on Meta, and Meta attributes $30,000 in revenue to those ads. ROAS is 3.0.
ROAS is quick, and it's useful for comparing campaigns, ad sets and ads inside one platform. Its weak spot is the word "attributed." That revenue comes from the platform's own attribution, which can include sales that would have happened anyway and sales another channel is also claiming. The post on what to trust when Meta, Shopify and GA4 disagree covers why.
MER: marketing efficiency ratio
MER = total revenue ÷ total marketing spend, for the same period.
Fernway's total store revenue for the month is $80,000 across every channel. Total marketing spend, including Meta, other ad platforms and any other paid marketing, is $20,000. MER is 4.0.
MER ignores attribution entirely, which is its strength. It's a blended view of whether your marketing as a whole is paying for itself. The trade-off is that it includes revenue from returning customers, email and organic traffic, so it can look healthy while new-customer acquisition is struggling. Decide what counts as "marketing spend" (ad spend only, or also creative and agency costs) and keep that definition consistent.
CAC: customer acquisition cost
CAC = marketing spend ÷ new customers acquired.
Fernway acquires 250 new customers this month on $20,000 of marketing spend. Blended CAC is $80.
The key word is new. If you divide spend by all orders, including repeat buyers, you get a cost per order, which flatters you. Shopify's reports can separate first-time and returning customers, so you can count new customers directly.
A useful companion is MER calculated on new-customer revenue only. Say Fernway's 250 new customers each spent $100 on their first order. That's $25,000 of new-customer revenue against $20,000 of spend, or a new-customer MER of 1.25. It answers a blunt question: how much of your spend is being paid back by first orders alone?
Contribution margin
Contribution margin is what's left from an order after the costs that come with it: product cost, shipping, fulfillment, payment processing and returns. It's measured before fixed costs like salaries, software and rent.
Here's Fernway's contribution on a typical $100 order, before ad spend:
| Per order | Amount |
|---|---|
| Order revenue | $100 |
| Product cost | −$25 |
| Shipping and fulfillment | −$15 |
| Payment processing | −$3 |
| Returns and refunds allowance | −$7 |
| Contribution margin | $50 (50%) |
You'll also hear "contribution margin after marketing," which subtracts ad spend too. To keep the example simple, assume every Fernway order carries the same 50% margin. Then $80,000 of revenue produces $40,000 of contribution, and subtracting $20,000 of marketing spend leaves $20,000 of contribution after marketing. That's the money available to cover fixed costs and, after that, profit.
Break-even ROAS
Break-even ROAS = 1 ÷ contribution margin ratio (the ratio before ad spend).
For Fernway: 1 ÷ 0.50 = 2.0. At a ROAS of 2.0, each $1 of ad spend brings in $2 of revenue, which carries $1 of contribution margin. That exactly covers the dollar you spent. Below 2.0, those orders lose money on the first purchase. Above 2.0, they contribute.
Back to Fernway's Meta results: $30,000 of attributed revenue at a 50% margin is $15,000 of contribution. Minus $10,000 of ad spend, that's $5,000 contributed, assuming the attribution is accurate.
Now picture a different fictional brand with the same 3.0 ROAS but a 25% contribution margin. Its break-even ROAS is 1 ÷ 0.25 = 4.0. The same $30,000 of revenue carries only $7,500 of contribution against $10,000 of spend, a $2,500 loss. Same ROAS, opposite outcome. That's why comparing ROAS across brands tells you very little.
Why ROAS alone misleads
- It ignores margin. A 3.0 ROAS is comfortable at a 50% margin and a loss at 25%.
- It mixes new and returning customers. Retargeting past buyers can post high ROAS because many of those people were likely to buy anyway. That can hide weak acquisition.
- It depends on attribution. Platform ROAS can climb while total sales stay flat.
- It ignores what happens next. A low first-order ROAS can still be worth it if customers come back.
- It's easy to improve by shrinking. Cutting spend down to only the most efficient audiences often raises ROAS while lowering total contribution.
New vs. returning customers
Split every number you can by new and returning. Acquisition spend should mostly be judged on new customers: new-customer CAC, new-customer revenue against spend, and first-order contribution.
In Fernway's case, CAC is $80 and a first order contributes $50, so each new customer starts $30 in the hole. Across 250 customers, that's a $7,500 first-order gap. If 150 of them come back for a second $100 order at the same margin, that's 150 × $50 = $7,500, which closes the gap. Whether that happens, and how fast, is what decides if an $80 CAC is affordable.
Returning-customer revenue matters, but much of it comes from email, SMS and the product itself. Don't use it to justify acquisition spend that isn't working on its own terms.
Which number for which decision
- Which ad or ad set wins inside Meta? Platform ROAS or CPA. The same attribution rules apply to both sides, so the comparison is fair. For tests, see choosing the metric that decides a test.
- Is our marketing working overall? MER, tracked as a trend over weeks and months.
- Can we afford customers at this cost? CAC compared with first-order contribution and how often customers buy again.
- Should we increase spend? Contribution after marketing. If total contribution rises when spend rises, keep going.
- What's the lowest ROAS we can accept? Your break-even ROAS, plus whatever cushion you want for profit.
The short version
- ROAS is attributed revenue ÷ ad spend. Use it to compare ads within a platform.
- MER is total revenue ÷ total marketing spend. Use it as a blended check on the whole picture.
- CAC is marketing spend ÷ new customers. Compare it with what a new customer contributes over time.
- Contribution margin is what's left after variable costs, and it sets your break-even ROAS: 1 ÷ margin ratio.
- Judge acquisition on new customers and contribution, not ROAS alone.