Why Do We Have a 5x ROAS but No Profit?

What does a 5x ROAS actually mean?

ROAS — return on ad spend — is the revenue an ad platform attributes to your ads, divided by what you spent on them.

Google Ads describes its target ROAS as the average conversion value you’d like to get for each dollar you spend on ads, and reports the result as “conversion value / cost.” It’s a clean, useful measure of how an ad account turns spend into revenue by its own rules. What it isn’t is a profit figure.

That gap — between what a platform credits to its ads and what the business keeps — is the whole subject of this article. Google itself notes that a ROAS target should be set based on your business goals, which is a polite way of saying the platform can’t know whether 5x is enough for you. There’s no universal “good” ROAS. The same 5x can be comfortable for one company and a slow leak for another.

Where does the money from a 5x ROAS actually go?

Take one month at an established direct-to-consumer product company. The numbers are illustrative, not benchmarks — but the structure is the one most product businesses run on. The ad platforms, added together, report $250,000 of attributed revenue on $50,000 of spend. Here’s what the company’s own books show for the same month.

Every order in that month was contribution-positive. Acquisition, on its own terms, cleared its cost. And the platforms’ 5x was honest by their rules. The business still ended the month about $6,000 short once its fixed costs were paid.

Notice one more thing. The company’s own revenue divided by media spend — often called MER — is 5.6x here, even better than the platform ROAS. It still didn’t produce a profit. A healthier-looking ratio doesn’t change what the order economics and fixed costs underneath it say.

ROAS is measured at the top of the ledger. Profit is decided at the bottom.

Why doesn’t a strong ROAS turn into profit?

When the dashboard and the business disagree, the cause usually sits in one of three places. Revenue has costs attached to it that ROAS doesn’t see. Some of the revenue in the report may not be new. And averages can hide what the next dollar of spend is doing.

• What revenue has to pay forProduct cost, order costs and the full cost of acquisition
• Whether the revenue is newAttribution, returning customers and double-counted credit
• What the average hidesThe next dollar of spend, and the fixed costs still waiting

Gross margin sets what an order can afford

Two $100 orders at the same 5x ROAS — $20 of media each — are not the same business. At a 70% gross margin, the order has $70 left after product cost: $50 after the ad, $36 after $14 of order costs. At a 30% margin, it has $30, then $10, then minus $4. Same dashboard, opposite economics.

Illustrative. Both orders show a 5x ROAS.

A handy rule of thumb: the break-even ROAS on a business’s own revenue is roughly 1 divided by its contribution margin before acquisition. In the example above that margin is 37.5%, so the business needs about 2.7x just to cover media. The complete guide to e-commerce marketing walks through the same math with two products side by side.

Discounts, returns and fulfillment come out before anything is kept

Attributed revenue is usually the order value the platform recorded. It doesn’t automatically reflect the promotion that discounted it, the return that reversed it, the shipping the company absorbed, or the packaging, pick-and-pack and payment processing behind it. Shopify defines net sales as gross sales minus returns, allowances and discounts — and is careful to add that net sales aren’t the same as profitability. Order costs still come after that.

Count each cost once. If a cost already sits inside your product cost, don’t subtract it again as a fulfillment cost.

Acquisition costs more than media

Platform ROAS divides by media spend only. If the company also pays for media management and acquisition-focused creative every month, those are part of what it costs to win customers. In the example, media-only ROAS is 5.0x; attributed revenue over the full $59,000 of acquisition cost is about 4.2x.

That doesn’t mean every marketing salary or brand campaign belongs in acquisition cost. The useful line is simple: include costs that exist because you’re buying customers, and leave out costs you’d carry anyway. Whatever you decide, decide it once and keep the definition stable, so the number means the same thing next quarter.

Is the revenue in the ad report really new revenue?

Not necessarily. Attributed revenue is revenue the platform can connect to an ad. It isn’t proof the sale happened because of the ad.

Attributed isn’t the same as incremental

Meta, for example, credits a website purchase to an ad when it happens within a set number of days after someone viewed or clicked it. That’s a legitimate, documented rule. But “this customer saw our ad before buying” and “this customer bought because of our ad” are different claims. The second one — incrementality — asks what would have happened without the ad at all.

That’s genuinely hard to measure. In a study of 15 large advertising experiments at Facebook, researchers found that the observational methods commonly used in the industry often failed to reproduce the results of randomized experiments on the same campaigns. The point isn’t that ads don’t work. It’s that attribution and causation can diverge, and a report alone can’t tell you by how much.

It matters for the economics. Suppose only half of the $250,000 in the example would not have happened without the ads. The incremental return is then 2.5x — and at a 37.5% contribution margin, that $125,000 leaves about $46,900 to pay for $50,000 of media. Illustrative, but it shows how a strong-looking 5x and a money-losing incremental result can describe the same month.

Returning customers and existing demand show up in ad reports

Ads often reach people who already know you: past customers, people searching your brand name, visitors retargeted after browsing. Those ads can have real value — keeping you visible, closing a hesitant buyer. But they’re different from finding a new customer.

A widely cited set of field experiments at eBay found that paid search on its own brand name had no measurable short-term benefit, and that new and infrequent customers responded to ads while frequent buyers — whose purchases ads didn’t change — accounted for most of the spend. eBay is an unusually well-known brand, so those results won’t transfer directly. The question they raise does: how much of your ad revenue is new-customer acquisition, and how much is monetizing demand you already had?

More than one platform can claim the same order

Meta, Google, Google Analytics and your store each measure from a different vantage point, with different rules about what gets credit and when. Google’s own help documentation notes that differences between Google Analytics and Google Ads are common and can occur even with a correct setup — Google Ads, for instance, records a conversion on the date of the click rather than the purchase.

None of that means anyone is wrong. It means the platforms’ attributed revenue shouldn’t be added together and treated as company revenue. For why channel-to-channel ROAS comparisons are especially unreliable, see Meta Ads vs. Google Ads for e-commerce.

Why can ROAS look fine while the business gets worse?

Because ROAS is an average, and an average can stay attractive while the last dollar you added is losing money.

The average and the next dollar are different questions

In the example, suppose the first $40,000 of spend was credited with $225,000 of revenue, and the last $10,000 with $25,000. Blended, that’s still 5.0x. But the last $10,000 ran at 2.5x — below the roughly 2.7x the business needs just to cover media. Even taking that credited revenue at face value, at a 37.5% contribution margin the final increment left about $9,400 to pay for $10,000 of ads.

Whether and when this happens varies by business, product and channel; there’s no universal point where it starts. But it’s the reason “our ROAS is still 5x” doesn’t, by itself, answer “should we add another $20,000?” The useful question is what the next increment is producing — in new customers, acquisition cost and contribution.

Fixed costs are still waiting at the end

Contribution-positive orders don’t guarantee a profitable company. What’s left after acquisition still has to cover salaries that aren’t in product cost, rent, insurance, software, administrative expense, equipment and, where there’s borrowing, financing costs. In the example, $46,000 of contribution met $52,000 of fixed costs.

For manufacturers there’s a wrinkle worth knowing. As Harvard Business Review has pointed out, cost of goods sold can include fixed costs — factory labor and production overhead among them — which is why gross margin and contribution margin aren’t the same number. Knowing which of your costs actually move with orders is what makes the arithmetic trustworthy. That’s territory to work through with whoever keeps your books; the point here is simply that ROAS sees none of it.

What should you look at besides ROAS?

Keep ROAS. Just pair each question you’re really asking with the number that actually answers it.

On MER: dividing total revenue by total ad spend is a useful company-level check, because it doesn’t depend on any platform’s attribution. But it’s still a revenue-to-spend ratio. It doesn’t subtract costs and doesn’t prove that the ads caused the revenue — the example above had a 5.6x MER and still lost money.

Should you increase ad spend if ROAS is good but profit is weak?

Not on ROAS alone. More spend at the same economics usually makes the gap bigger, not smaller — unless the thing limiting profit is somewhere that more volume actually helps.

The useful management question isn’t “is our ROAS good?” It’s closer to this: is marketing generating profitable demand that the business can responsibly absorb? Before approving more budget, it’s worth working through a short sequence:

1. Rebuild the month in your own numbersNet sales, product cost, order costs and all-in acquisition cost — not platform revenue.
2. Separate new customers from returning onesKnow what a genuinely new customer costs, and how much of the ad revenue is repeat business.
3. Ask what the next increment is doingCompare the most recent budget increase with the average, not just the blended ROAS.
4. Check the operation and the cashMore orders need inventory, fulfillment and working capital before the money comes back.
5. Decide how much to trust the measurementIf platform reports and company results rarely reconcile, fix that before spending more.

None of this means becoming a media buyer. It means asking the people running your advertising — in-house or outside — to connect their reports to your business’s economics. A good team will welcome the question. If you’re sizing the budget itself, setting an e-commerce marketing budget from contribution works through that decision step by step.

Coast333’s perspective here comes from its founder’s roughly eight years inside a direct-to-consumer manufacturer handling about 10,000 orders a month — across operations and, eventually, as Head of Marketing — where marketing decisions sat alongside inventory, fulfillment and margin.

Frequently asked questions

Keep going

Sources & further reading

1. About Target ROAS biddingGoogle Ads Help. How Google defines ROAS and target ROAS.
2. Conversion discrepancies between Google Analytics and Google AdsGoogle Ads Help. Why the two report different numbers.
3. About actions attributed to your adMeta Business Help Center. How purchases are credited to ads.
4. A Comparison of Approaches to Advertising Measurement: Evidence from Big Field Experiments at FacebookGordon, Zettelmeyer, Bhargava and Chapsky, Marketing Science, 2019.
5. Consumer Heterogeneity and Paid Search Effectiveness: A Large-Scale Field ExperimentBlake, Nosko and Tadelis, Econometrica, 2015.
6. What Are Net Sales?Shopify. Net sales defined, and why they aren’t profitability.
7. Contribution Margin: What It Is, How to Calculate It, and Why You Need ItHarvard Business Review, 2017.

The worked example and every dollar figure in this article are illustrative, created to show the arithmetic. They aren’t industry benchmarks or client data.