ROAS measures how much revenue each peso of ad spend generates (revenue ÷ spend), ROI measures real profit after all costs ((profit − spend) ÷ spend), and CPA measures what it costs you to get a sale or customer (spend ÷ conversions). To know whether a campaign is profitable, compare your ROAS with your break-even ROAS, which is 1 divided by your contribution margin.
- ROAS = attributed revenue ÷ ad spend.
- ROI accounts for product, shipping, fees and operating costs; ROAS doesn't.
- CPA = spend ÷ number of conversions; compare it with what each customer brings in.
- Break-even ROAS = 1 ÷ contribution margin (as a decimal).
- Customer lifetime value changes how much you can afford to pay to acquire a customer.
Why these three metrics and not others
Ad platforms show you dozens of metrics: reach, impressions, CTR, CPM, frequency. They're all useful for diagnosis, but none of them answers the owner's question: did this money come back with a profit? That's what three numbers are for, and they're worth mastering.
- ROAS (return on ad spend): how much you sold for every peso spent on ads.
- ROI (return on investment): how much you really earned, after every cost is taken out.
- CPA (cost per acquisition): how much you paid for each sale, lead or customer.
If you want to run the numbers as you read, open the free ROI and ROAS calculator: plug in your own figures and see in seconds whether the campaign makes or loses money.
ROAS: the formula and how to read it
ROAS = revenue attributed to the campaign ÷ campaign spend.
A ROAS of 4 means every peso you spent brought back 4 in sales. It's expressed as a number (4) or a multiple (4x); some platforms show it as a percentage (400%).
The problem with ROAS is that it measures revenue, not profit. A ROAS of 4 is excellent for a high-margin cosmetics store and can be a loss for an electronics store that earns little per unit. That's why you need your break-even point.
Your break-even ROAS
Contribution margin is what you keep from each sale after variable costs: cost of goods, shipping, packaging, the payment processor's fee and the marketplace commission if there is one. With that margin as a decimal:
Break-even ROAS = 1 ÷ contribution margin.
| Contribution margin | Break-even ROAS | Typical fit |
|---|---|---|
| 60% | 1.67 | Digital products, courses, services with low variable costs |
| 45% | 2.22 | Apparel, accessories, private-label cosmetics |
| 35% | 2.86 | Specialty retail, gourmet food |
| 20% | 5.00 | Electronics, home appliances, low-margin resale |
Below that number, every sale the campaign generates costs you money. Above it, the campaign contributes to fixed costs and profit.
ROI: the number the business cares about
ROI = (attributable profit − spend) ÷ spend.
The difference from ROAS is that ROI starts from profit, not revenue. For an honest calculation, include everything the campaign caused: product, shipping, fees, creative design, agency management or your team's hours, and returns.
An example in ratios: a specialty coffee shop in Torreón has a 40% contribution margin and its Meta campaign delivered a ROAS of 3.5. For every peso spent it sold 3.5 and kept 1.4 in contribution (3.5 × 0.40). Subtract the peso it spent and it earns 0.4 per peso: a 40% ROI on ad spend. If it also pays for video production, that ROI goes down, and it's better to review it quarterly rather than weekly.
Rule of thumb: look at ROAS to optimize campaigns day to day; look at ROI to decide how much budget each channel deserves.
CPA: what each customer costs you
CPA = spend ÷ number of conversions. The conversion can be a purchase, a lead, a booked appointment or a WhatsApp conversation. That's why it helps to be specific: cost per purchase, cost per lead, cost per appointment.
CPA is the natural metric for businesses that don't sell online: clinics, construction companies, schools, professional firms. There are no "attributed sales" in the platform, but you do know how many leads came in and how many became customers.
From cost per lead to cost per customer
A low cost per lead is useless if the leads don't buy. Calculate the full funnel:
- Cost per lead = spend ÷ leads.
- Close rate = customers ÷ leads.
- Cost per customer = cost per lead ÷ close rate.
If a real estate agency in Torreón pays little per lead but closes one out of every 80, its cost per customer is 80 times its cost per lead. Better follow-up (replying in minutes, not hours) usually lowers cost per customer more than any campaign tweak; we explain how in automating lead follow-up with n8n and WhatsApp.
Comparison table: when to use each metric
| Metric | Formula | Use it to | Common trap |
|---|---|---|---|
| ROAS | Revenue ÷ spend | Compare campaigns and ads in e-commerce | Ignoring margin and celebrating a ROAS that loses money |
| ROI | (Profit − spend) ÷ spend | Decide budget across channels | Forgetting production, agency or return costs |
| CPA | Spend ÷ conversions | Services, leads, appointments, messages | Optimizing for cheap leads that never buy |
| Lifetime value (LTV) | Margin per purchase × purchases over the relationship | Know how much you can pay for a customer | Assuming repeat purchases your business doesn't actually get |
Lifetime value changes the rules
A gym, a neighborhood pharmacy or a pet food store doesn't make money on the first purchase: it makes money on the fifth. If a customer buys several times a year on average, you can accept a higher CPA on the first sale because the margin comes later.
To use lifetime value (LTV) seriously you need data: who bought, when and how much. A CRM, a loyalty program or your online store gives you that. Microapps like Monkey Lealtad track repeat purchases at brick-and-mortar businesses, and email marketing is the cheapest way to trigger them.
Attribution mistakes that inflate or sink your numbers
- Adding up what each platform reports: Meta and Google can both claim the same sale. The total shows more sales than actually hit the register.
- Different attribution windows: Meta typically uses 7 days after a click and 1 day after viewing the ad; Google Ads has its own settings. Compare apples to apples.
- Sales over WhatsApp or in store that never get recorded: the campaign looks bad when it's actually selling. Use UTM-tagged links, discount codes per channel and the question "How did you hear about us?"
- Measuring too soon: considered purchases (furniture, cars, medical services) take weeks to close.
To make GA4 show the real source of each sale, follow our guide to UTMs for measuring campaigns.
A complete example: a boutique in Torreón
Let's walk through a case using ratios, not amounts. A clothing boutique in the Torreón Jardín neighborhood sells in its physical store and online. After the cost of the garment, shipping, packaging and payment fees, its average order leaves a 45% contribution margin. Its break-even ROAS is 1 ÷ 0.45 = 2.22.
For one month it runs two Meta campaigns:
- Campaign A, seasonal catalog: reported ROAS of 3.1. It's above break-even; for every peso spent it leaves 0.4 in net contribution (3.1 × 0.45 − 1).
- Campaign B, luxury collection: reported ROAS of 2.0. It looks acceptable, but it's below 2.22: every sale costs money.
However, the boutique notices that many shoppers from campaign B later come into the physical store and buy there. By reviewing the discount codes redeemed at the register, it uncovers sales Meta couldn't see. With those sales included, campaign B's blended ROAS rises above break-even. The conclusion isn't to turn it off, but to measure it better and decide with register data.
This example sums up the logic: the platform's ROAS is a starting point, margin sets the threshold, and offline sales can change the decision. And if the boutique also knows that a new customer buys three times a year on average, it can accept a higher initial CPA on acquisition campaigns and make it up with repeat purchases via email or WhatsApp.
How to use these numbers every month
- Calculate your contribution margin and break-even ROAS once, by product line.
- Set a maximum CPA per conversion type based on margin and lifetime value.
- Review ROAS and CPA by campaign every week; pause whatever is below break-even after a fair learning period.
- Review ROI by channel every month against actual register sales.
- Shift budget toward what generates the most profit, not what brings the most clicks.
If you want reports that tie ads to real sales, our digital marketing team builds them with GA4, Meta and your CRM; if you sell online, our e-commerce service sets up your store to record purchase value, margin and the source of every order. For businesses in La Laguna, we review your numbers in person through digital marketing in Torreón.
