Why a 3x ROAS Can Still Lose Money

The math that fools most dashboards

ROAS = revenue ÷ ad spend. Spend $1,000, sell $3,000, and the dashboard celebrates a 3x. But if your product margin is 30%, that $3,000 of revenue is only $900 of gross profit — you spent $1,000 to make $900, losing $100 while the platform reports success.

This is the most common way small e-commerce and dropshipping stores bleed money: optimizing toward a number that was never connected to profit.

Break-even ROAS: one division

Break-even ROAS = 1 ÷ gross margin. At 40% margin you need 2.5x just to stay even; at 25% margin, 4x; at 60%, only 1.67x. 'Good ROAS' isn't an industry benchmark — it's a property of your own margin, and every product in the same account can have a different break-even.

Run your numbers through the ROAS calculator: enter spend, revenue, and margin, and it shows both the vanity ROAS and the profit-weighted one, plus the break-even threshold for your margin.

Platform ROAS is inflated on top of that

Meta and Google attribute conversions generously — view-through windows, brand searches, and customers who would have bought anyway all get credited to ads. Platform-reported ROAS commonly runs 1.5–2x higher than blended reality (total revenue ÷ total spend, sometimes called MER). Trustworthy advertisers watch both: platform ROAS for optimization signals, blended ROAS for whether the business is actually working.

When running below break-even is correct

One exception: if you have measured lifetime value, a first-order loss can be rational. A subscription with a $120 average lifetime profit can profitably buy customers at a $60 first-order loss — but only if retention is real, not hoped for. The dangerous version is 'we'll make it back on repeat purchases' without the LTV data to prove it.