ROI vs ROAS vs CPM: How to Tell If Your Ads Are Actually Making Money

Your ad dashboard says 3x return on ad spend, and it feels like a win. Then you check the bank account and the month ended flat. Nothing is wrong with the dashboard. ROAS, ROI and CPM answer three different questions, and mixing them up is the most common reason small businesses keep funding ads that quietly lose money.

Three numbers, three different questions

Advertising reports are full of acronyms, but only a handful decide whether you are making or losing money. Each one measures a different step between spending a dollar and keeping a dollar.

• CPM (cost per mille) is what you pay for every 1,000 times your ad is shown. It measures the price of attention.
• ROAS (return on ad spend) is the revenue you earned for each dollar spent on ads. It measures the efficiency of the ads in terms of sales.
• ROI (return on investment) is the profit you earned compared with what you spent. It measures whether the whole thing was worth doing.

The key difference is that ROAS uses revenue while ROI uses profit. A sale of $100 on a product that costs you $70 to buy, ship and handle is $100 of revenue but only $30 of gross profit. If you paid $35 in ads to get that sale, ROAS looks healthy at 2.86 while you actually lost $5.

ROAS: revenue for every dollar of ad spend

The formula is simple:

ROAS = revenue from ads / ad spend

If you spend $1,000 and the ads generate $3,000 in sales, ROAS is 3.0. Platforms often show it as 3.0x or as 300%. They mean the same thing: every dollar spent returned three dollars of revenue.

ROAS is useful for comparing campaigns, audiences and creatives against each other, because it needs only two numbers that your ad platform already tracks. Its weakness is that it knows nothing about your costs. Two businesses with the same 3.0 ROAS can have opposite results, one thriving on 70 percent margins and the other bleeding on 20 percent margins.

ROI: did you actually make money?

ROI compares profit with the amount you put in:

ROI = (profit – cost) / cost x 100

For advertising, the cleanest version uses gross profit (revenue minus the direct cost of the goods sold) and subtracts the ad spend:

Ad ROI = (gross profit – ad spend) / ad spend x 100

Take the same $1,000 ad spend and $3,000 of revenue. If your gross margin is 40 percent, gross profit is $1,200. Subtract the $1,000 of ads and you are left with $200. Divide by the $1,000 you invested and ad ROI is 20 percent. That is a modest but real profit, even though ROAS looked like an impressive 300 percent.

If you are not sure what your margin is, or you mix it up with markup, read our guide to markup vs margin first. The two are routinely confused, and the wrong one will wreck every calculation below.

Break-even ROAS: the number that matters most

Instead of asking whether a ROAS of 3 is good, ask what ROAS you need just to avoid losing money. The answer comes straight from your margin:

Break-even ROAS = 1 / gross margin

With a 40 percent margin, break-even ROAS is 1 / 0.40 = 2.5. At 2.5 you earn $2.50 of revenue per ad dollar, keep 40 percent of that ($1.00) and hand all of it back to the ad platform. Below 2.5, you lose. Above it, you profit.

Write your break-even ROAS on a sticky note beside your monitor. Every campaign decision then becomes a quick comparison instead of a debate.

One adjustment matters if customers buy again. If the average customer spends $150 over their lifetime but your first order is $50, you may accept a first-order ROAS below break-even because later purchases make up the difference. Do this only with real repeat-purchase data, not hope.

CPM, CTR, CPC, CPA: the funnel behind the result

ROAS is the outcome. The funnel numbers explain why it is high or low, and they tell you where to intervene.

1. CPM = ad spend / impressions x 1,000. How much 1,000 views cost.
2. CTR (click-through rate) = clicks / impressions. How many people found the ad interesting enough to click.
3. CPC (cost per click) = ad spend / clicks. Equals CPM / (1,000 x CTR).
4. Conversion rate = orders / clicks. How many visitors buy.
5. CPA (cost per acquisition) = ad spend / orders. Equals CPC / conversion rate.
6. AOV (average order value) = revenue / orders.

Notice how they chain together. Revenue per ad dollar, which is ROAS, equals AOV divided by CPA. To raise ROAS you can only do three things: raise AOV, raise conversion rate, or lower the cost of getting attention (CPM) or clicks (CPC, via a better CTR).

A full worked example

An online shop spends $1,000 on a campaign. Here is what the platform reports and what we can calculate from it.

Per order, the shop keeps $20 of gross profit (40 percent of $50) and pays $16.67 to acquire the customer. That leaves $3.33 per order, times 60 orders, which is the $200. The campaign works, but with a thin cushion: if CPM rises by just 20 percent, the $200 profit falls to zero. That fragility is invisible if you only watch ROAS.

Scenario table: what each ROAS really means

Using the same 40 percent margin and $1,000 of spend, here is how profit changes across different results:

The relationship is not linear in the way people expect. Going from 3 to 4 triples your profit (from $200 to $600), because the extra revenue costs you nothing more in ad spend. Small improvements in conversion or order value have an outsized effect once you are past break-even.

The hidden costs that shrink your margin

The margin you plug into the break-even formula must be the real one, after every cost that scales with a sale. Gross margin that ignores these will make a losing campaign look profitable:

• Payment processing. Card and wallet fees take a percentage plus a fixed amount per order. See our comparison of PayPal, Stripe, Etsy and eBay fees on a $100 sale.
• Marketplace fees. Selling through a platform adds listing, transaction and sometimes advertising fees. Use the Etsy, eBay or Amazon FBA fee calculators.
• Shipping and packaging that you absorb.
• Returns and refunds. A 10 percent return rate removes a tenth of your revenue and often the cost of shipping both ways.
• Discounts and coupons that reduce AOV.
• Inflation in supplier prices, which erodes margin over time. The Inflation Calculator shows how much your costs have drifted.

Rerun your break-even ROAS whenever any of these change. A supplier price increase of a few percent can move the break-even from 2.5 to 3.0 without you noticing.

How to fix an unprofitable campaign

Start from the break-even gap and walk back up the funnel. Change one thing at a time so you know what worked.

1. Check tracking first. If ROAS looks wrong, confirm that purchases are being counted once and that refunds are removed. Bad data produces confident bad decisions.
2. Fix the landing page before the ad. If CTR is fine but conversion is below 1 percent, the problem is after the click: slow page, unclear price, surprise shipping fees or a clumsy checkout.
3. Raise AOV. Bundles, free-shipping thresholds and add-ons increase revenue per order without increasing the cost of acquiring it. Moving AOV from $50 to $60 in our example raises ROAS from 3.0 to 3.6 with no change to the ads.
4. Improve creative to lift CTR. A higher CTR lowers CPC, because the platform rewards ads people engage with.
5. Narrow the audience. A cheap CPM with the wrong people is expensive. A higher CPM with buyers is often the better deal.
6. Cut losers on a schedule. Pause ad sets that stay below break-even after a meaningful number of conversions, and move the budget to the ones above it.
7. Raise prices. If your product is competitive, a 10 percent price increase can lift margin from 40 to about 45 percent and cut break-even ROAS from 2.5 to 2.2.

Common mistakes

• Judging by ROAS alone. It ignores margin, so it cannot say whether you are profitable.
• Using markup as margin. A 50 percent markup is only a 33 percent margin. That error alone can flip a profitable campaign into a loss.
• Ignoring attribution windows. Platforms may credit sales that would have happened anyway. Compare platform revenue with your actual store totals.
• Scaling too fast. Raising budgets sharply often lifts CPM and lowers ROAS because the platform reaches less interested people. Increase gradually.
• Calling it too early. Ten clicks and one sale is not a result. Wait for enough orders for the numbers to settle.
• Forgetting fixed costs. Break-even ROAS covers only variable costs. Rent, salaries and software still need to be paid from the profit.

Frequently asked questions

What is a good ROAS?

There is no universal good number. It depends on your margin. A ROAS of 2 is excellent at an 80 percent margin and a loss at 30 percent. Calculate your own break-even ROAS (1 divided by margin) and aim above it with a safety buffer.

Is ROI the same as ROAS?

No. ROAS divides revenue by ad spend. ROI divides profit by cost. ROAS tells you how hard the ads work; ROI tells you whether the business made money.

Can ROI be negative?

Yes. If profit after ad spend is below zero, ROI is negative. In the table above, a ROAS of 2.0 at a 40 percent margin gives an ROI of minus 20 percent.

What is a normal CPM?

It varies widely by platform, country, season and audience. Rather than chasing a benchmark, track your own CPM over time and watch for sudden jumps, which usually signal audience fatigue or more competition.

Should I include shipping in revenue?

Include shipping you charge as revenue only if you also include the shipping you pay as a cost. The simplest approach is to measure product revenue and product margin, and treat net shipping separately.

How do I know if the ads caused the sale?

Platform reports credit ads for sales that may have happened anyway. A simple check is to pause ads for a short period or compare regions and see how total sales move, not only platform-reported sales.

Related guides and tools

References

1. Investopedia, “Return on Investment (ROI)”: definition and formula. investopedia.com
2. Investopedia, “Return on Ad Spend (ROAS)”. investopedia.com
3. Wikipedia, “Cost per mille”. en.wikipedia.org

The figures in the examples are illustrative, not benchmarks for your industry. This article is general information, not financial advice; check your own margins and costs before changing ad budgets.