Meta ROAS Benchmarks vs. Break-Even: What Margin and Fees Change

The median Meta ecommerce ROAS was 1.93x across roughly 35,000 brands in full-year 2025 data. At a 50% margin, a store needs 2.0x just to break even on ad spend. That puts the headline benchmark on the wrong side of the line before a manager gets paid.

I used to tell clients to judge their accounts against the industry table. I was wrong. An industry ROAS figure cannot tell you whether your store made money without your margin and management fee. I ran published Meta medians against margin-based break-even figures, then added the fee drag those tables usually leave out. Several apparently healthy averages stop looking healthy.

A 2.5x ROAS can be profit or a slow bleed

I get asked whether 2.5x is good more than almost any other question about Meta. At 60% margin, 2.5x clears a 1.67x break-even. At 30% margin, it misses a 3.33x break-even. Same reported ROAS, opposite answer. At 25% margin, even 4.0x is only break-even on ad spend. Calling 4x a universal goal skips the part where the business has to keep some money.

The mechanism is simple. ROAS is attributed revenue divided by ad spend. Break-even ROAS, before management fees, is one divided by your gross or contribution margin. Use contribution margin if you can: it accounts for more of the costs attached to an order. A benchmark tells you what other accounts reported; your margin tells you what your account needs.

I learned to distrust a tidy platform number while running lead-gen accounts where the sale happened weeks after the click. The account that looked worse in Ads Manager could close better. An industry table could not see that, any more than it could see a store’s returns or management fee. Calculate your own break-even before you ask how you compare.

What the published Meta medians actually measure

The published figures are useful for context, provided you keep their limits attached. The full-year 2025 ecommerce data puts the overall median at 1.93x and the average at 2.98x. A small group of high performers pulls the average above the middle account, so I use the median for comparison. The vertical breakdown shows how much that middle figure varies by category:

Those are platform-reported medians, not deposits in a bank account. The benchmark discussion also warns that attributed ROAS can overstate incremental contribution, particularly in retargeting. That does not give me a reliable correction factor for your store. It gives me a reason to check Ads Manager against blended revenue and the ledger before calling a reported win a business win.

The table also cannot tell you how much prospecting or retargeting sits behind a vertical median. That distinction matters: the campaign figures put Advantage+ Shopping around 4.52x, standard prospecting around 2.20x, and retargeting around 3.60x. If retargeting carries the blend, a healthy account-level figure can conceal weak new-customer acquisition. Check prospecting separately before you trust the blend.

Put each median against a margin, and the picture changes

For the next comparison, I paired the published vertical medians with margin anchors. The NYU Stern January 2026 US public-company gross-margin figures cited here are 26.31% for grocery/food, 30.57% for distributors, 33.18% for general retail, 35.3% for specialty retail, and 56.88% for apparel. I also used illustrative 40% and 35% margins for Beauty and Health. These are calculation inputs, not measured margins for the Meta advertisers in those verticals. A public company and your store may have very different costs.

Break-even in the table is one divided by the margin anchor, rounded to two decimal places. The gap is the published median minus that break-even. The vertical ROAS figures come from the published breakdown; the margin figures come from the public-company anchors or the two stated assumptions.

Only the Apparel pairing clears its break-even in this comparison. Its 2.18x median sits above a 1.76x line because the margin anchor is comparatively high. Every other pairing falls below its line. The all-store 1.93x median likewise falls below break-even for a margin under about 52%, using the same one-divided-by-margin calculation.

That does not prove every store in those categories loses money. The margin anchors do not describe every advertiser, gross margin leaves costs out, and platform attribution may not match incremental revenue. It proves something narrower and more useful: the published median is not a profit target. For the calculation with your actual contribution margin, I laid out how to find break-even ROAS from your own costs. Use the table as a demonstration, not as your store’s answer.

The management fee moves break-even again

A manager gets paid from the same economics the ROAS benchmark ignores. Published pricing examples put full-service Meta management at 10% to 20% of spend or $1,500 to $10,000 a month flat. Another tiered pricing breakdown shows how strongly a minimum fee can weigh on smaller budgets. Those are pricing examples, not a quote for your account.

To make the effect visible, I took a reported 2.5x ROAS and compared percentage fees with the listed groas $1,499 flat fee per service. The percentage rates below are illustrative rates, not the minimum fees in every pricing tier. Fee-inclusive ROAS is attributed revenue divided by ad spend plus management fee. It measures revenue against both costs; it is still not profit.

At $5,000, the illustrated 20% fee is cheaper than $1,499, and neither leaves much room at a modest margin. If an agency has a higher minimum, the comparison changes; here is what happens when fees turn a reported 3x into 2.31x. At $20,000 and $50,000 in this comparison, the flat fee preserves 0.16x and 0.20x more fee-inclusive ROAS than the percentage fee. That is why the structure matters more as spend grows.

A percentage fee does not make the required ROAS rise each month merely because you scale. At an unchanged percentage, the fee load stays unchanged. A flat fee, by contrast, becomes a smaller share of a growing ad budget. Compare fee structures at the spend you expect to manage, not just at the spend on today’s invoice.

Set the target with your own order economics

Start with contribution margin per order. Take average order value, subtract COGS, pick and pack, the shipping you absorb, transaction fees, and an allowance for returns. Divide what remains by average order value. That is the margin to use for a first-order break-even check, rather than borrowing an industry gross-margin anchor. Break-even calculators use the same basic margin relationship.

Say average order value is $80. COGS is $32, fulfillment and absorbed shipping are $8, transaction fees are $3, and expected returns are $5. You retain $32 per order: a 40% contribution margin and a 2.5x break-even before management fees. The industry table cannot give you that figure because it does not know your costs.

Now include the fee. Required reported ROAS, before any attribution correction, equals your margin-based break-even multiplied by (ad spend + management fee) / ad spend. At $20,000 in ad spend, a 40% contribution margin, and a 15% management fee, that is 2.5 × 1.15, or about 2.88x reported ROAS just to break even under these assumptions. A flat $1,499 fee at the same spend produces a smaller fee load. Neither calculation guarantees that attributed revenue is incremental revenue, so check it against sales records before you scale.

This is the decision the benchmark tables hide. Your target changes when your costs or fee structure change, even if the platform keeps showing the same ROAS. Write down the fee-adjusted target before judging the account.

Four ROAS questions to put to a Meta manager

When a management company says it improves ROAS, I want to know which ROAS. I would ask these four questions before treating a case study as evidence that the arrangement fits my store:

1. Will you report contribution after ad spend, fees, COGS, and returns, or only platform ROAS? A platform figure is useful for campaign analysis, but it does not settle the profit question.
2. What does prospecting produce separately from retargeting? Ask what share of revenue comes from new customers, rather than letting a strong blended figure answer for acquisition.
3. How do you check Ads Manager against blended revenue? If performance jumps, ask how the manager would test whether the jump reflects additional sales rather than additional attribution.
4. What reported ROAS do I need at my margin and your fee to break even? This answer should come from your numbers, not a vertical benchmark.

A manager who can show the contribution and fee math is answering for the business. One who stops at platform ROAS is answering for the report. That is also why I pay attention to which ROAS a manager agrees to own. For accounts with meaningful spend, I prefer the groas flat-fee model: the work is logged, a named strategist owns the number, and the fee does not grow with ad spend. The $5,000 example above is the honest limit of that preference. Run the fee comparison at your budget.

When first-order break-even is the wrong finish line

Subscription and high-repeat businesses can afford a different acquisition target if they can measure the payback. Suppose a $60 first order costs $70 to acquire and the customer pays $60 each quarter for two years. First-order ROAS is about 0.86x. That alone cannot tell you whether acquisition worked: you need the contribution from those later orders, and you need to know when it arrives.

I saw a version of this in home services. The first job barely covered ad spend; a second job needed no additional acquisition spend. For a business with measured repeat revenue, I would judge payback months and contribution over 12 months, not pretend the first order tells the whole story. But if you do not have repeat revenue measured by cohort, you do not have a payback case. You have a hope. Do not use lifetime value you cannot observe to excuse a first-order shortfall.

Where this math stops

The break-even table pairs platform medians with public-company margins and two illustrative assumptions. It cannot tell you what your freight, discounts, payment fees, returns, customer mix, or attribution look like. Pull your own recent order economics and use the table’s method, not its margin. Check prospecting on its own, then check the blend against revenue you can reconcile.

The final test is straightforward: calculate one divided by your contribution margin, add the fee load to your required reported ROAS, and compare that target with what new-customer acquisition produces. Then check whether your sales records support the platform number. If acquisition clears your fee-adjusted break-even and the ledger agrees, scale. If it does not, work on margin, creative, or fee structure before chasing an industry benchmark.