What is ROAS and how do you calculate it?
ROAS measures the gross revenue an advertising campaign returns for each unit of currency spent. The formula is simple:
ROAS = Revenue attributed to ads / Ad spend
If you spend $10,000 on Meta ads and those ads drive $50,000 in tracked revenue, your ROAS is 5.0 — often written as 5:1 or 500%. A ROAS of 1.0 means you earned back exactly what you spent, before accounting for product cost, shipping, or overhead.
Two things quietly distort this number, so define them before you optimize. First, attribution: a 7-day-click, 1-day-view window on Meta reports more revenue than last-click in Google Analytics, so the same campaign can show two different ROAS figures. Second, revenue basis: decide whether you're counting gross revenue or net of refunds and discounts. Pick one attribution model and one revenue definition, then hold them constant so your before-and-after comparisons are honest.
- Gross revenue basis: revenue including discounts and pre-refund — inflates ROAS
- Net revenue basis: after refunds and discount codes — closer to reality
- Blended ROAS: total store revenue / total ad spend across all channels
- Platform ROAS: revenue the ad platform claims via its own pixel and attribution window
What is a good ROAS?
A good ROAS is any ROAS comfortably above your break-even point with room left for profit — for most e-commerce brands that lands between 3:1 and 5:1, with 4:1 (400%) cited as a common average target. But a 4:1 target is meaningless without your margins.
A brand selling supplements at 80% gross margin can be wildly profitable at a 2.5:1 ROAS. A retailer reselling electronics at 15% margin will lose money at 4:1. The percentage that's 'good' is entirely a function of your unit economics, not an industry average you read somewhere.
Context also shifts the target by funnel stage. Prospecting to cold audiences naturally runs a lower ROAS (say 1.5:1 to 2.5:1) because you're paying to acquire first-time buyers, while retargeting and branded search often post 6:1 or higher because you're harvesting existing demand. Judge each campaign against its job, and judge the account on blended ROAS.
What is break-even ROAS and how do you calculate it?
Break-even ROAS is the point where the gross profit from a sale exactly equals the ad spend that produced it — spend more efficiently than this and you make money, less and you lose it. It's derived entirely from your gross margin:
Break-even ROAS = 1 / Gross margin
If your gross margin is 40%, your break-even ROAS is 1 / 0.40 = 2.5. Every dollar of ad spend must return at least $2.50 in revenue just to cover the product cost on that sale. To hit a target net margin, add it in: to keep 20% profit on a 40%-margin product, you need revenue to cover cost plus profit, pushing your target ROAS to roughly 1 / (0.40 − 0.20) = 5.0.
This single calculation reframes the whole optimization problem. You're not chasing the highest possible ROAS — you're trying to maximize total profit while staying above break-even. Knowing your break-even number tells you exactly how aggressively you can bid and how much you can afford to spend acquiring a customer.
- Find gross margin: (Selling price − COGS − shipping − payment fees) / Selling price
- Break-even ROAS = 1 / gross margin
- Target ROAS for X% net margin = 1 / (gross margin − target net margin)
- Example: 50% margin → 2.0 break-even; 25% margin → 4.0 break-even; 15% margin → 6.7 break-even
What's the difference between ROAS and profit (POAS)?
ROAS measures revenue per ad dollar; POAS — Profit on Ad Spend — measures actual profit per ad dollar, and it's the number that pays your bills. The difference matters because you can grow ROAS while shrinking profit.
POAS = Gross profit attributed to ads / Ad spend
Here's the trap. Imagine two campaigns at an identical 4:1 ROAS. Campaign A pushes a full-price product at 55% margin; Campaign B pushes a discounted bundle at 20% margin. Same ROAS, but Campaign A's POAS is more than double Campaign B's. If you optimize purely to ROAS, the algorithm happily scales the low-margin bundle because revenue looks fine — while your bank balance quietly erodes.
This is why I feed profit data, not raw revenue, into reporting wherever the platform allows it — passing margin-adjusted conversion values back through the Conversions API so the bidding algorithm optimizes toward profit. ROAS is the steering wheel most platforms hand you; POAS is where you actually want to end up.
How do you improve your ROAS?
To improve ROAS you pull five levers, roughly in order of impact: creative, offer, targeting, landing page/CRO, and post-click. Here's a mental model that shows why. For a paid-traffic funnel:
ROAS ≈ (Conversion rate × Average order value) / Cost per click
That equation means every ROAS gain comes from one of three moves: raise conversion rate, raise average order value (AOV), or lower cost per click (CPC). The five levers below each attack one or more of those variables. Start at the top — creative and offer move the needle fastest and cost the least to test.
- Creative (lowers CPC, lifts CTR): On Meta, creative is the #1 driver. Test hooks in the first 3 seconds, use native UGC and problem-solution formats, and run enough volume — I aim for 5 to 10 new concepts per week. A higher click-through rate and hook rate lowers CPM and CPC, so the same budget buys more qualified clicks.
- Offer (lifts conversion rate and AOV): The offer often beats the ad. Test free-shipping thresholds, bundles, a stronger guarantee, first-order discounts, or urgency. Raising AOV with a bundle or an order-bump directly multiplies ROAS at the same ad cost.
- Targeting (lowers wasted spend): Go broad and let the algorithm find buyers, but feed it clean signal — server-side conversions via CAPI, accurate value events, and exclusions for recent purchasers. On Google, cut wasted spend with negative keywords, tighter match types, and device/geo bid adjustments.
- Landing page & CRO (lifts conversion rate): This is the most under-invested lever. Ensure message match between ad and page, load under 2.5 seconds on mobile, put the value proposition and primary CTA above the fold, and add trust signals (reviews, guarantees, payment badges). Moving conversion rate from 2% to 3% is a 50% ROAS lift with zero extra ad spend.
- Post-click & retention (lifts LTV and blended ROAS): Capture emails and phone numbers, then recover abandoned carts and drive repeat purchases through email/SMS flows and post-purchase upsells. Higher lifetime value lets you afford a lower first-purchase ROAS and still win — which is how you outbid competitors on cold traffic.
How do you diagnose where your ROAS is leaking?
Before you change anything, find the weak link. Low ROAS is a symptom; the cause lives at a specific stage of the funnel. Walk the numbers in order and compare each against a healthy benchmark.
Run this diagnostic on any underperforming campaign. The first stage that's clearly below benchmark is your bottleneck — fix that before touching anything downstream, because optimizing a later stage can't compensate for a broken earlier one.
- Impressions to clicks (CTR): Below ~1% on Meta or a weak CTR on Google Search means the creative or ad copy is the problem — the offer isn't landing or the hook is weak.
- Clicks to sessions (bounce/load): A big gap between reported clicks and landing-page sessions points to slow load speed or a broken/mismatched page.
- Sessions to add-to-cart: A low add-to-cart rate signals a product, price, or page-clarity problem — the visitor isn't convinced.
- Add-to-cart to purchase: High cart abandonment usually means friction at checkout — surprise shipping costs, forced account creation, or limited payment options.
- AOV vs. break-even: If every stage looks healthy but ROAS still lags, your order value is too low for your margins — add bundles, upsells, or a free-shipping threshold.
What ROAS mistakes should you avoid?
The most common way to hurt a business is to optimize ROAS in isolation. Chasing a higher ROAS almost always means shrinking to your most efficient audiences — which caps volume and total profit.
Total profit, not ROAS, is the goal. A campaign at 3:1 spending $100,000 can generate far more profit than one at 8:1 spending $5,000. There's a real trade-off curve between efficiency and scale, and the profit-maximizing point is often a deliberately lower ROAS at much higher spend.
Watch these traps as you optimize. Each one makes the reported number look better while the business does worse.
- Judging campaigns on platform-reported ROAS alone — Meta and Google both count the same sale, so summed platform ROAS overstates reality. Reconcile against blended ROAS (total revenue / total spend).
- Cutting spend the moment ROAS dips — short attribution windows underreport revenue on longer buying cycles; give data time before killing a campaign.
- Ignoring margin — a rising ROAS built on discounts can lower actual profit. Track POAS.
- Over-restricting audiences — narrow targeting can spike ROAS but starve growth. Broad + clean signal usually scales better.
- Testing one variable at a time too slowly — with limited daily conversions, glacial testing wastes weeks. Prioritize high-impact tests (creative, offer) first.