A good ROAS is any ROAS above your break-even ROAS, and break-even is set by your margins, not by industry folklore. Return on ad spend divides revenue attributed to ads by the ad spend that produced it. A ROAS of 4.0 means four dollars of attributed revenue per dollar spent, which sounds healthy until you learn the business runs on a 20% gross margin, at which point it is losing money on every order.
The only formula that decides "good"
At a 40% gross margin, break-even ROAS is 2.5: each advertising dollar must return 1.00) to repay the dollar spent. At a 25% margin the hurdle is 4.0; at 60% it is just 1.67. This is why a supplements brand and a software company can look at the same 3.0 ROAS and one is printing money while the other is subsidizing the ad platform. And gross margin is the generous version: fulfilment, payment processing, returns, and packaging all consume gross profit, so the honest hurdle uses contribution margin after those variable costs, which pushes break-even higher still.
Worked example
A store sells a product for 62, a gross margin of 38% and break-even ROAS of about 2.63. Fulfilment and payment fees take another 7 per $100 of ad-attributed revenue once the full variable cost is counted. Most "profitable" campaigns that disappoint at the bank are exactly this arithmetic error.
Why platform ROAS overstates your result
Ad platforms claim credit for conversions they influenced to any degree, use attribution windows that reach back days or weeks, and each platform claims the same sale independently, so summed platform revenue routinely exceeds actual store revenue. Platform ROAS is still useful for comparing campaigns inside one platform with one consistent window. For business decisions, calibrate against blended figures: total revenue divided by total ad spend (sometimes called MER or efficiency ratio), which cannot double-count and shows whether the whole programme is above break-even even when individual attributions are fuzzy.
Setting a target ROAS
Start at your true contribution break-even, then add the profit you require. If break-even is 3.45 and you want a 10% operating margin on ad-driven revenue, target roughly break-even divided by (1 - 0.10), about 3.83. Prospecting campaigns usually must be allowed a lower target than remarketing, because cold audiences convert worse but feed the funnel; judging prospecting by the remarketing hurdle kills growth, while judging remarketing by the prospecting hurdle overpays for demand that would have converted anyway. Segment the target by funnel stage, then review against blended results monthly.
ROAS is not profit, and not LTV
ROAS counts first-order revenue only. Where repeat purchase is common, a campaign can be rationally run below first-order break-even if customer lifetime value reliably covers the gap, but that is a financing decision that demands cohort data, not a hope. Conversely, a high ROAS on tiny spend proves little: efficiency usually falls as budgets scale and the best audiences saturate, so the right question is the ROAS at the spend level you actually need.
Limitations
Every ROAS figure embeds attribution choices: window length, click versus view credit, and cross-device matching. Change the window and "good" changes. Use one consistent definition for comparisons, verify against blended store-level efficiency, and never set targets from another company's margins.
How the answer changes by business model
Physical products live and die by the margin math above. SaaS and digital products, with gross margins of 70 to 85%, break even between roughly 1.2 and 1.45, so a ROAS of 2 can be strongly profitable and targets are set by payback period instead. Lead generation has no revenue at the click at all: translate through close rate and deal value first (a 2,000 sale is $200 of expected revenue per lead, and the campaign's effective economics follow from there). Marketplaces and aggregators often run deliberately below first-order break-even to buy liquidity. The constant across models is that the hurdle is derived from your own unit economics; the variable is only which margin and which time horizon you honestly use.
A two-week ROAS audit
Pull total store revenue and total ad spend for the last fourteen days and compute the blended ratio. Then list every campaign's platform-reported ROAS and mark each against your true contribution break-even. Campaigns below break-even on platform numbers are almost certainly below it in reality; campaigns above it need the blended check to confirm they are not merely claiming organic demand. Reallocating budget from the first group to the second, in small steps with three to four days between moves, is the highest-value hour most advertisers can spend this month.
Frequently Asked Questions
Is a ROAS of 4 good?
Only if your break-even is below 4. At a 40% gross margin (break-even 2.5) it is comfortably profitable before other costs; at a 20% margin (break-even 5.0) it loses money on every sale.
What ROAS should a new store aim for?
Aim first to measure your contribution break-even honestly, then target 10 to 25% above it while you gather data. Early campaigns often run below break-even during learning; the plan should say how long that is acceptable.
Should I optimize ROAS or profit?
Profit dollars, ultimately. A lower ROAS at ten times the spend usually produces more total profit. Use ROAS as the efficiency floor and scale spend while the floor holds.