What is a good ROAS? How to work out break-even and target ROAS from your margins

There is no good ROAS, only yours. Break-even ROAS is 1 ÷ gross margin; the target, which also pays for overhead and profit, is 1 ÷ the margin left after them. Judge it on closed sales from new customers, not platform conversions, and on lifetime value where repeat business is real. The highest ROAS is often not the best.

Search for a good ROAS and you will find a number: 4 to 1, 3 to 1, 400%. Every one of them is right for some business and wrong for most. Return on ad spend is revenue divided by ad spend, and revenue is not profit. The same $4 of sales per ad dollar leaves a dental practice comfortably ahead and a used-car lot underwater, because one keeps 60 cents of each sales dollar and the other keeps 12.

This post is for businesses whose sales close on the phone or in person, such as roofers, dentists and car dealers, with an online store for comparison. It works out the two numbers that define a good ROAS for you, explains why the ROAS on the ad platform's screen is usually not that number, and works through four businesses whose ranking flips once margin is counted.

Why there is no universal good ROAS

A good ROAS is the return at which advertising pays for itself and then for its share of the business. That point depends on gross margin, overhead and the profit you want, so it differs by industry and by company. A benchmark ROAS compares businesses whose margins can differ sevenfold.

ROAS has one cost in it, the ads. Everything else a sale costs, from the materials and the installer's day to the rent and the receptionist, sits outside the ratio. A business with expensive work to deliver needs much more revenue per ad dollar just to stand still, and no single number can serve a software company and a car dealer at once.

The arithmetic behind that range: 1 ÷ 0.7172 = 1.39 for software, 1 ÷ 0.3318 = 3.01 for retail, 1 ÷ 0.1546 = 6.47 for construction and 1 ÷ 0.1041 = 9.61 for vehicles. A ROAS of 4 is a 2.9-times cushion for one and a loss of more than half the ad spend for the other. Industry benchmarks also average in businesses with different baselines, as our post on blended ROAS explains, so even a benchmark for your own industry tells you what others report, not what you need.

Break-even ROAS: 1 ÷ gross margin

Break-even ROAS is the return at which the gross profit from ad-driven sales exactly pays for the ads. It is 1 divided by gross margin, as a decimal: 2.5 at 40%, 4 at 25%, 10 at 10%. Below it, every ad dollar loses money before overhead; above it, the ads add gross profit.

Gross margin here means revenue minus the costs that come with each sale, divided by revenue. For a store that is goods, shipping, packing and payment fees. For a service business it is materials, the crew's or clinician's time on the job, subcontractors, sales commission and card fees. Rent, office staff and the owner's salary do not belong here; they come in at the next step.

The third column is ROAS × margin − 1: at a 30% margin, 4 × 0.30 − 1 = $0.20 of gross profit left per ad dollar after the ad is paid for. The same ROAS swings from a 60-cent loss to a two-dollar gain down the table, which is the whole case against a universal number.

Target ROAS: add overhead and profit

Target ROAS is the return at which advertising also pays its share of overhead and leaves the profit you want. Subtract overhead and profit, each as a share of revenue, from gross margin; target ROAS is 1 divided by what remains. At 40% margin, 15% overhead and 10% profit, it is 1 ÷ 0.15 = 6.7.

What remains is the share of each sales dollar you can afford to spend on ads. If gross margin is 40% and the business needs 15 cents of every dollar for rent, office wages, insurance and vehicles, and wants 10 cents of profit, 15 cents is left for advertising. Spending 15 cents per dollar of sales is a ROAS of 1 ÷ 0.15 = 6.67.

Take overhead and profit from your own accounts: last year's overhead divided by last year's revenue, and the net margin you are aiming for. If the result is zero or negative, no ROAS works at your prices, and the fix is pricing or costs, not marketing.

Which line to judge which decision by

The two lines answer different questions. Overhead is mostly fixed: the rent is the same whether you take one more job or not. So for the next dollar, break-even is the real floor; a campaign above it adds money the business would not otherwise have. The target is for the program as a whole, the question of whether marketing, on average, is carrying its weight. Our post on how to allocate a marketing budget covers judging the next dollar rather than the average one.

That warning is the trade-off in one line. A higher target buys fewer, cheaper sales; a lower one buys more sales at a thinner return. A business that sets its target at break-even grows fastest and earns nothing on the margin; one that sets it far above its target earns well on a shrinking volume. The target formula gives a defensible middle, and the bidder can only aim at it if the conversion values it sees are real.

Why platform ROAS is not ROAS on closed sales

An ad platform's ROAS is the value of the conversions it counted divided by spend. For a business that sells by phone or in person, those conversions are calls and forms carrying values someone assigned, not sales. ROAS on closed sales divides invoiced revenue, matched to the ad's leads, by the same spend. The two often differ several times over.

An online store's platform sees the checkout, so its ROAS is close to real apart from refunds and other channels claiming the same order. A roofer's platform sees a phone call. It does not hear whether the caller wanted a quote, a warranty visit or a wrong number, and it never learns that the job closed six weeks later at $14,000.

So the platform's ROAS for a phone business is roughly (qualifying calls × the value you typed in) ÷ spend. The gap from real ROAS has several sources, and they push in both directions:

• Assigned values. A flat $300 per call is a guess. If your average job is $14,000 and one call in twenty-five buys, the true value of a call is $560; if one in fifty buys, $280.
• Calls are not sales. Repeat callers, existing customers, job seekers and vendors all pass a 60-second bar.
• Each platform counts alone. Two platforms that both touched one customer both claim the sale, as our post on channel conflict shows.
• Offline leads are invisible to it. A customer who saw the ad and called the number on your truck shows up nowhere in the platform.

The fix is to compute ROAS yourself on closed sales: match each invoice to the lead that started it on phone or email, total the matched revenue by channel, and divide by that channel's spend. You can also send the sales back to the platform so its bidder learns from them, as our post on offline conversion imports describes.

ROAS vs ROI

ROAS is revenue per ad dollar and contains no costs other than the ads. ROI is profit per dollar of all the marketing's costs. With ad spend as the only cost, ROI on gross margin = ROAS × gross margin − 1. ROAS is the steering number for campaigns; ROI on margin is the verdict on whether they made money.

Break-even ROAS is simply the ROAS at which ROI on margin is zero, and target ROAS is the ROAS at which ROI covers overhead and profit too. That is why the two lines in this post are worth more than a benchmark: they turn a revenue ratio into a profit statement for your business. When agency fees, call-handling staff and software also belong in the cost, move to ROI; our post on how to calculate marketing ROI sets out which costs count.

New-customer revenue vs repeat revenue

A ROAS meant to judge acquisition should count only revenue from new customers. Repeat customers who click an ad or call from it were already yours; counting their purchases lifts the ratio without the ads having won anyone. Report new-customer ROAS alongside total ROAS, and set the target on the new-customer figure.

This matters most where customers come back on their own: a dental practice's existing patients booking cleanings, a dealer's service customers buying their next car, a roofer's past clients calling about a leak. Each may find you through an ad because it is the first result, and each shows up as ad revenue. Mark every sale new or returning from the sales system, which usually knows. Our post on customer acquisition cost by channel covers the same split from the cost side.

That is the platform's version of the split, and it is only as good as the customer list behind it, which is one more reason to keep a clean record of who bought before.

How lifetime value changes the target

When customers buy again, the first sale understates what an ad bought. Judged on lifetime revenue, a target ROAS applied to the first sale can fall in proportion: first-sale target = lifetime target × first-sale revenue ÷ lifetime revenue. It only holds for repeat revenue you have measured, not hoped for.

Take a dental practice whose new patients spend $700 in their first year and $1,900 over three, at a 60% margin. Its target ROAS on lifetime revenue is 5.56. On first-year revenue, a ROAS of 5.56 × $700 ÷ $1,900 = 2.05 meets the same target, because the remaining $1,200 a patient will spend arrives later. A practice that demands 5.56 on the first visit will turn away campaigns that are profitable over three years.

Three cautions keep lifetime value honest. Use the horizon you have actually observed: if your records cover two years, do not plan on five. Use margin, not revenue, since later visits may carry different margins than the first. And measure it per channel, because customers from different sources come back at different rates; our post on customer lifetime value by channel shows how from a sales export.

A worked example: four businesses, four good ROAS numbers

An online kitchenware store, a roofing contractor, a dental practice and a used-car dealer each spent on one ad channel for a quarter. Each has its own break-even and target ROAS, and each has a platform-reported ROAS and one on closed sales. The figures are illustrative. The highest ROAS on the page misses its target.

The arithmetic. Break-even is 1 ÷ margin: 1 ÷ 0.40 = 2.50, 1 ÷ 0.32 = 3.13, 1 ÷ 0.60 = 1.67, 1 ÷ 0.12 = 8.33. Left for ads is margin − overhead − profit: 40 − 15 − 10 = 15%, 32 − 14 − 8 = 10%, 60 − 30 − 12 = 18%, 12 − 5 − 3 = 4%. Target is 1 ÷ that share: 1 ÷ 0.15 = 6.67, 1 ÷ 0.10 = 10.00, 1 ÷ 0.18 = 5.56, 1 ÷ 0.04 = 25.00.

The arithmetic. The roofer's platform value is 160 calls over 60 seconds × $300 = $48,000, and its closed sales are 6 jobs × $14,000 = $84,000. The dentist's is 90 form leads × $100 = $9,000, against 26 new patients × $700 of first-year revenue = $18,200. The dealer's is 400 leads × $50 = $20,000, against 9 cars × $22,000 = $198,000. The store's platform counted $36,600 of orders; $31,200 survived refunds and orders another channel had also claimed. Each ROAS is revenue ÷ spend, for example $84,000 ÷ $8,000 = 10.50. Gross profit per ad dollar is ROAS × margin − 1: 5.20 × 0.40 − 1 = 1.08, 10.50 × 0.32 − 1 = 2.36, 6.07 × 0.60 − 1 = 2.64, 19.80 × 0.12 − 1 = 1.38.

What the four numbers say

• The store is the only one whose platform ROAS overstates: 6.10 on screen looks close to target, 5.20 on kept orders is 1.5 short. Still well above break-even, so it adds money, but it does not yet pay its share of overhead.
• The roofer's platform said 6.00, far short of its 10.00 target. Closed sales say 10.50, just above it. A roofer reading the platform would have judged its best channel a failure.
• The dentist has the second-lowest ROAS on closed sales and the best return per ad dollar, $2.64, because it keeps 60 cents of each sales dollar. On three-year revenue it would look better still.
• The dealer has the highest ROAS on the page, 19.80, and is about a fifth short of its target of 25. Three of the nine buyers were existing service customers; on the six new ones, 6 × $22,000 ÷ $10,000 = 13.20.

Here is the arguable part. Ranked by platform ROAS the order is store, roofer, dentist, dealer. Ranked by closed-sales ROAS it is dealer, roofer, dentist, store. Ranked by what each ad dollar left behind, it is dentist, roofer, dealer, store. Only the last ranking says where advertising made money, and it is the one no ad platform can show you, because it needs your sales and your margins.

CloseRev produces the closed-sales revenue in the second table, not the targets. It reads two CSV files you export, closed sales and leads, matches each sale to leads on normalized phone and email, and by default credits it to the source of the earliest matched lead; last touch is an option. Only high-confidence matches count automatically; medium- and low-confidence ones stay reviewable, and unmatched sales stay in Direct / Unknown. If you enter monthly spend per channel, the report shows ROAS for those channels. It does not know your margins or overhead, so break-even and target ROAS stay a spreadsheet step.

How to set your own good ROAS

Work out break-even from gross margin, the target from overhead and profit, and compare both with ROAS on closed sales from new customers, by the month the lead arrived. It takes an afternoon with the bookkeeper's figures and two exports, and the two lines then hold until your prices or costs change.

1. Get gross margin from last year's accounts: revenue minus the costs that come with each sale, divided by revenue. Split it into two or three bands if your work varies, as replacement and repair do.
2. Compute break-even ROAS, 1 ÷ margin, and target ROAS, 1 ÷ (margin − overhead share − profit share).
3. Match a quarter of closed sales to the leads that started them, mark each new or returning, and total new-customer revenue by channel and lead month.
4. Divide by each channel's spend for the same lead months, and set each channel against both lines: below break-even, fix or cut; between the lines, improve; above target, consider more.
5. Where repeat business is real and measured, restate the first-sale target with the lifetime ratio, and recheck it once a year.

If most of your revenue sits in an unmatched pile, fix the lead records before trusting any channel's ROAS; our post on unattributed revenue covers where those sales usually come from.

A good ROAS is not a number you look up; it is two lines you draw from your own books. Below 1 ÷ gross margin, the ads lose money. Above the target, they pay their way. Measure the space between on closed sales from new customers, and expect the highest ROAS on the page to be beaten by the business that keeps more of each dollar.