ROI vs ROAS: The Difference and Which Metric Decides What

The ROI vs ROAS question comes down to scope: ROAS (return on ad spend) divides ad revenue by ad spend and measures nothing but the efficiency of the ads themselves. ROI (return on investment) divides net profit by total cost, including product, shipping, fees and people, and answers whether the operation actually makes money. The working rule is simple: use ROAS for campaign and budget decisions inside the platforms, and ROI for channel and investment decisions above them. This guide walks through both formulas, break-even ROAS and the exact decision each metric should own, with numbers you can copy. In 2026, the spread of automated bidding (tROAS, ROAS goals) makes the distinction even more critical: if the target you hand the platform ignores your margin, the optimization sprints in the wrong direction at full speed.

What is the difference between ROI and ROAS?

ROAS is the ratio of ad revenue to ad spend, and it deliberately ignores every other cost. ROI is the ratio of net profit to total investment, so product costs, shipping, payment fees and tooling all enter the denominator. The difference is scope: ROAS answers 'are the ads working', ROI answers 'is this making money'. The same campaign can look brilliant on ROAS and be underwater on ROI.

• ROAS: ad revenue ÷ ad spend. Example: 40,000 in revenue on 10,000 of spend = 4.0× (400%).
• ROI: (revenue - total costs) ÷ total costs × 100. Every cost of delivering the sale is included.
• Scope: ROAS is a tactical, campaign-level metric; ROI is a strategic, business-level metric.
• Cadence: ROAS supports daily and weekly optimization; ROI belongs in monthly and quarterly reviews.

How do you calculate ROAS and ROI?

Run both formulas on the same example: 10,000 units of ad spend produced 40,000 of tracked ad revenue, so ROAS = 40,000 ÷ 10,000 = 4.0×. For ROI, add the remaining costs: if product, shipping and fees total 26,000, total cost is 36,000 and net profit is 4,000. ROI = 4,000 ÷ 36,000 ≈ 11%. The same campaign is 'excellent' on ROAS and only just above water on ROI. Expressed as a percentage, a 4.0× ROAS equals 400%; ROI, however, is always a profit-based percentage, so never mix the two values in one column.

1. Pull ad revenue and spend from the platform: ROAS = 40,000 ÷ 10,000 = 4.0×.
2. Add the cost of goods, shipping and fees for those orders: 26,000.
3. Compute net profit: 40,000 - 36,000 = 4,000.
4. Compute ROI: 4,000 ÷ 36,000 × 100 ≈ 11%.

What is break-even ROAS and how do you find it?

Break-even ROAS is the minimum ROAS at which a campaign stops losing money, and the formula is 1 ÷ gross margin. At a 40% margin, break-even is 1 ÷ 0.40 = 2.5×; any ROAS below that loses money no matter how good it looks in the dashboard. With industry benchmarks putting typical blended averages around 2.5-3×, a low-margin product can be unprofitable at a perfectly 'average' ROAS. Use the pre-advertising contribution margin in the formula: revenue minus product cost, shipping, fees and a returns allowance. Subtracting only the product cost understates the threshold.

Why can a high ROAS still lose money?

Because the only cost inside the ROAS formula is ad spend; product cost, shipping, returns and fees never enter it. On a 15% margin product, even a 5.0× ROAS sits below the 6.7× break-even and produces a loss on every order. Meanwhile a 40% margin product at 3.2× ROAS clears its 2.5× threshold and banks profit. The campaign that looks worse in the dashboard can be the one actually earning. The illusion also works in reverse: killing a 'low' ROAS campaign in a high-margin category means voluntarily abandoning profitable volume.

The structural fix for this illusion is optimizing toward profit instead of revenue. We cover profit-based bidding and measurement in our POAS guide; this article stays focused on the decision split between ROI and ROAS.

Which decision should each metric drive?

Match the metric to the blast radius of the decision. In-platform moves such as pausing ad sets, rotating creatives or shifting budget between campaigns run on ROAS. Investment questions such as 'does this channel deserve more money next quarter' run on ROI. And when you monitor all channels together, a top-level ratio that ignores attribution fights, MER, becomes the reference. The same split applies to reporting: presenting ROAS alone to a manager or client is half the picture; put margin and total profit next to it. Together, the ROAS-MER-ROI trio forms an unbroken decision chain from daily to quarterly.

• Daily and weekly optimization: ROAS (plus CPA). It is fast and available in-platform; creative and audience calls live here.
• Budget between campaigns: margin-adjusted ROAS. Never compare campaigns selling different-margin products on raw ROAS.
• Channel totals: platform ROAS inflates when attribution overlaps; track MER for blended efficiency.
• Quarterly investment: ROI. If you doubt the ads create sales at all, verify with incrementality testing.

What are the most common ROI and ROAS mistakes?

Four mistakes dominate in practice: treating the two metrics as synonyms, setting one target ROAS across products with different margins, feeding inflated platform-attributed revenue into ROI math, and panicking at a ROAS dip instead of diagnosing its cause. All four share one root: not knowing what each metric deliberately leaves out.

• One target ROAS for everything: a 20% margin product and a 60% margin product cannot share a 4.0× goal; one loses money, the other leaves volume on the table.
• Trusting platform revenue blindly: attributed revenue can overlap across platforms; base ROI on backend or accounting revenue.
• Reading a ROAS drop with one metric: diagnose the cause first; our 5-step diagnosis guide exists for exactly that.
• Using ROI for campaign optimization: ROI is slow and aggregate; it signals too late for daily calls, where ROAS does the work.

In short: ROAS is speed, ROI is truth, and they belong on the same screen, read together with margin context. Ads Sensor unifies your Meta, Google, TikTok, Criteo and GA4 data in one panel, compares campaign ROAS side by side, catches risks and proposes reasoned actions. Doing this read manually stops scaling as accounts and campaigns multiply; seeing the context next to every campaign turns 'which one do I pause' into a minutes-long question. Try it on your own account by joining the beta.