ROAS vs ROI for Ecommerce: Why a 5x ROAS Can Still Lose Money

An ad platform showing 5x ROAS does not mean the business earned $4 of profit for every $1 spent. ROAS divides attributed revenue by advertising cost. It says nothing, by itself, about product cost, shipping, refunds, payment fees or the overhead needed to run the company. That is why a campaign can look excellent in the ads dashboard and still leave the store with less money than before.

This article uses one made-up order throughout so you can check every number by hand. It is a teaching example, not a client result. Further down there is a calculator where you can replace each figure with your own.

What ROAS measures, and what it leaves out

ROAS is attributed revenue divided by advertising spend. If a platform credits $5,000 of purchases to $1,000 of spend, ROAS is 5.0x. Three things sit behind that number and none of them is profit.

First, attributed means the platform decided to credit itself under its own rules. Meta, Google and your store can credit the same order, so the total across channels can exceed what the store actually sold. Article Meta Says You Made $80,000. Google Claims $45,000. Shopify Shows $93,000 walks through that overlap.

Second, revenue is not what is left over. The product still has to be bought or made, packed, shipped, paid for through a card processor and, sometimes, refunded.

Third, ROAS ignores which customers are new. A campaign that mostly reaches people who were about to buy anyway can report a strong ROAS and add very little to the business.

Work through one order, not a slogan

Take a hypothetical brand with $100 average order revenue and an ad account reporting 5x ROAS. At that ratio each order costs $20 in advertising, which looks healthy. Now count what the order itself costs: $55 for the product, $15 for fulfillment and shipping, and $12 for payment fees, returns and other variable costs.

The dashboard says five-to-one. The order loses two dollars before rent, salaries or software. Nothing in the ads account is "wrong". The ratio simply answers a smaller question than the one the owner cares about.

Three measures with three different jobs

ROAS monitors how efficiently ad spend is credited with sales. It depends on attribution settings, so compare it only within one platform and one setup.

Contribution after advertising equals revenue, minus the variable costs of that order, minus the advertising cost. This is the number that tells you whether orders help pay for fixed costs and profit.

ROI needs a defined investment and a defined return. Businesses use the word loosely. If you publish or report ROI, say whether the numerator is incremental net profit or something else, and which costs sit in the denominator. Do not treat contribution margin as audited net profit.

How to calculate break-even ROAS

Break-even ROAS comes from your contribution margin before ads, which is contribution divided by revenue.

In the example the margin before ads is $18 on $100, or 18%. Break-even ROAS is 1 ÷ 0.18 = 5.56x. At 5x the first order loses money. If you also need to keep 5% of revenue after acquisition for overhead and profit, only 13% of revenue is available for advertising. The target becomes 1 ÷ 0.13 = 7.69x.

The table shows how fast the threshold climbs as margin falls. These are arithmetic results, not benchmarks.

A store with a 50% margin breaks even at 2x. A store with a 15% margin needs 6.67x just to stand still. They can run on the same ad platform with the same reported 4x and have opposite outcomes.

Why discounts move the target

A discount lowers revenue but not costs, so the margin falls faster than the price. Take 20% off the same order: the customer pays $80, costs stay at $82, and the order would already be below zero before ads. This is why the break-even number belongs in every promotion decision. The related article on what a 25% Black Friday discount does to profit shows the same effect with a different order.

The same 4x ROAS in three different businesses

To see why a benchmark cannot replace your own break-even figure, hold the reported ROAS fixed and change only the margin. In each case the platform credits $400 of revenue to $100 of spend, so every $100 of revenue costs $25 in advertising. These are hypothetical margins.

The first business has room to spend more and still earn. The second is exactly at break-even and has nothing left for overhead. The third loses $10 on every $100 it sells, and a dashboard that reports 4x will show no sign of it. The advice "aim for 4x" is right for one of these businesses, neutral for the second and wrong for the third.

What to do when ROAS is below break-even

A gap between your reported ROAS and your break-even ROAS has four possible causes. Fix the one that is actually responsible.

1. The margin is too thin for paid acquisition. Look at price, product cost, shipping charged versus shipping paid, and payment fees. A small change in margin moves the threshold a lot. At 18% the break-even is 5.56x. Lifting the margin to 25% brings it down to 4.00x.
2. The conversion path leaks. More sales from the same spend raise ROAS without touching margin. Diagnose the funnel first with the Shopify conversion rate and clicks without sales articles.
3. The offer is too discounted. A promotion lowers the margin and raises the threshold in one move. Recalculate break-even for the promotional price, not the regular one.
4. The credit is overstated. If the platform is claiming orders that would have happened anyway, its ROAS is higher than the real return. Compare it with store-level numbers before cutting or scaling.

If none of these can be fixed in the time you have, the honest conclusion is to spend less on that campaign until one can.

Common mistakes with ROAS

• Using one ROAS target for every product. Margins differ by product. A single target over-spends on the low-margin items and under-spends on the high-margin ones.
• Setting the target from last month's platform average. A past average describes what happened, not what is affordable.
• Leaving shipping out of the margin. If you offer free shipping, the carrier cost belongs in the variable costs of the order.
• Ignoring returns. Revenue that comes back is not contribution. Use net revenue after expected returns.
• Mixing new and returning customers. Returning customers often cost less to win. A blended ROAS can hide a new-customer ROAS that is below break-even.

Reconcile before you rely on the number

A campaign reporting 5x does not automatically mean 5x in unique, incremental revenue. Platforms credit overlapping purchases, attribution windows differ and returns arrive later. Before treating any ROAS as an investment decision, compare it with Shopify orders for the same period, with refunds and with your contribution per order. If the dashboard and the store disagree, diagnose the drop in ROAS before changing the campaign.

A useful companion metric is store-level efficiency: Shopify net sales divided by total advertising spend across all channels. It does not prove incrementality, but it cannot be inflated by two platforms claiming one sale.

When losing money on the first order can be acceptable

Paying more than the first order earns is a legitimate strategy if repeat purchases are real and arrive soon enough. That should be modeled from actual cohorts, not from an optimistic "lifetime value" multiple. The CAC payback model shows how to count only realized contribution and how long the cash is tied up. Seasonal stock clearance and deliberate brand campaigns can also justify a lower return, provided the trade-off is written down and approved.

A weekly review you can actually run

1. Pull Shopify net sales, orders and refunds for the closed week.
2. Calculate contribution per order before ads from real product, shipping and fee costs.
3. Divide 1 by that margin to get break-even ROAS, then add your required margin to get target ROAS.
4. Compare the target with what each platform reports, and note which channels overlap.
5. Separate new customers from returning customers wherever the data allows.
6. Decide for each campaign: scale, hold, fix the offer, or cut.

Run it with your own numbers

The calculator below opens on the example order: $100 of revenue, $55 product cost, $15 fulfillment and shipping, and $12 of other variable costs, with no contribution required after ads. It returns a maximum CPA of $18 and a break-even ROAS of 5.56x. Replace the values with your own to get your threshold, not an industry average. Payment fees and returns have their own fields here, so enter them there instead of inside the other-costs figure.

From the field

Common questions

Is 5x ROAS good?

It depends entirely on your margin. At a 50% margin before ads, 5x leaves a large profit. At an 18% margin it loses money. Compare ROAS with your own break-even figure.

Is ROAS the same as ROI?

No. ROAS is revenue per advertising dollar. ROI relates profit to an investment. A campaign can have a high ROAS and a negative ROI.

Should I optimize for ROAS or for contribution?

Report both, but make decisions on contribution after ads. ROAS is a useful signal for day-to-day management, as long as it is checked against a break-even number you calculated.

Does platform ROAS include shipping and tax?

That depends on how purchase value is sent to the platform and on your store settings. Check what your pixel or server events pass as the value, and compare it with Shopify's net sales definition.

Sources

1. Shopify, analytics revenue and conversion definitions help.shopify.com
2. Google Analytics, purchase and refund measurement support.google.com

Nenad Vasic

Founder and CEO of Fenix Network Media. His work focuses on the economics behind customer acquisition, conversion, measurement and profitable growth. Calculations in this article are original; third-party figures belong to their publishers and are credited in the sources.