Break-Even CPA vs ROAS for Dropshipping: Formulas and Examples

CPA and ROAS answer different questions

Cost per acquisition (CPA) is ad spend divided by the number of attributed purchases. Return on ad spend (ROAS) is attributed revenue divided by ad spend. Neither metric alone measures actual profit, because revenue is not contribution margin.

For a simple first-order product model, break-even CPA is the amount left from a sale after non-ad variable costs. Break-even ROAS is selling price divided by break-even CPA. Both depend on the assumptions and attribution window used.

Step 1: calculate contribution before advertising

Start with the actual selling price after discounts. Subtract item cost, shipping and handling, payment processing, and an evidence-based allowance for refunds, replacements, or other variable order costs. The result is contribution before advertising, not net profit after fixed overhead or taxes.

Example assumptions: selling price $40; item cost $12; shipping $5; payment fees $1.50; expected refunds/replacements allowance $1.50. Contribution before advertising is $40 − $12 − $5 − $1.50 − $1.50 = $20. This example is hypothetical, not a real supplier quotation.

Step 2: calculate break-even CPA and ROAS

In the example, the break-even first-order CPA is $20. If ads cost $15 per attributed purchase, contribution after ads is $5 per order before fixed costs and taxes. At $22 CPA, it is a $2 loss per order on the same assumptions.

Break-even ROAS equals $40 revenue divided by $20 allowable ad spend = 2.0×. At a measured ROAS of 2.5×, the implied CPA is $16 if each attributed order averages $40. At 1.5×, implied CPA is about $26.67. These conversions assume the same attribution basis and order value.

• Contribution before ads = selling price − non-ad variable costs.
• Break-even CPA = contribution before ads (for one attributed first-order purchase).
• Break-even ROAS = selling price ÷ break-even CPA, when break-even CPA is positive.
• Contribution after ads = contribution before ads − CPA.

What if your margin or average order value changes?

A bundle, discount, shipping increase, or return-rate change can alter the threshold. Recalculate for each meaningful offer and destination market rather than copying a single ROAS target across products.

If contribution before ads is zero or negative, there is no positive first-order acquisition budget that breaks even. Improve the economics or reconsider the offer before buying traffic.

Avoid false precision in ad-platform dashboards

Ad platforms can report attributed purchases differently. Compare the same time window and attribution settings, reconcile orders with store records, and account for cancellations, refunds, repeat purchases, and taxes when appropriate.

A profitable first order is not guaranteed by an attractive ROAS figure. Fixed expenses, creative costs, cash flow, and fulfillment issues can still change the business outcome.

A pre-launch decision checklist

Record the date, destination country, currency, price, discount, supplier quote, shipping method, payment fees, and refund assumptions. Define a maximum CPA and a spending cap before launching a test. Recalculate after real orders provide better evidence.

• Confirm your product and shipping costs for the intended market.
• Calculate contribution before ads using conservative assumptions.
• Set first-order break-even CPA and ROAS thresholds.
• Track actual attributed purchases and reconcile with order data.
• Review refunds and fulfillment costs before scaling spend.

Frequently asked questions

Is a 2.0× ROAS always profitable? No. The break-even ROAS depends on your variable cost structure and can be much higher or lower.

Should you use CPA or ROAS? Use both alongside contribution margin and verified order data. CPA is intuitive for a consistent order value; ROAS can help when order values vary.

Does break-even CPA include fixed costs? Not in the simple model above. To target overall business profitability, allocate relevant fixed expenses and set a stricter acquisition-cost limit.