ROI vs. Cost-Effectiveness: Formulas, Break-Even, and Evaluation Limits
Shusaku Yosa
ROI expresses profit relative to an investment; cost-effectiveness evaluates whether spending delivers an intended outcome. When the numerator is profit after deducting the investment and other included costs, break-even ROI is 0%. A positive ROI below 100% does not, by itself, mean the investment remains unrecovered.
Choose a measure for the decision
A basic ROI formula is included profit ÷ included investment × 100. Revenue is not profit, and the investment denominator must be specified. ROAS, which can describe attributed sales relative to ad spend, answers a different question.
Question | Possible measure | Required condition or limitation |
|---|---|---|
Did the investment leave a profit? | ROI | Consistent cost scope and period |
How much attributed revenue accompanies ad spend? | Revenue-based ROAS | Attribution and spend definition; not profit |
How efficiently were nonmonetary outcomes achieved? | Cost per valid outcome | Comparable quality, population, and horizon |
When will cash return? | Cash-flow plan and recovery period | Accounting profit does not establish payment timing |
Google Ads' official ROI example includes both advertising and product costs in the denominator. A marketing ROI using ad spend alone therefore has a different scope. Publish the formula alongside the name.
A fictional example: 50% ROI is not “half recovered”
Suppose a fictional initiative has attributed revenue of ¥2,500,000, product costs of ¥1,000,000, and ad spend of ¥1,000,000. Use the same period, with no additional costs, tax, or returns.
Profit after included costs is ¥2,500,000 − ¥1,000,000 − ¥1,000,000 = ¥500,000. The advertising cost has already been deducted. Dividing this profit by ad spend gives 50% ROI. It describes ¥500,000 remaining after the specified costs, not a failure to recover half the ad spend.
Item | Amount or calculation | Interpretation |
|---|---|---|
Attributed revenue | ¥2,500,000 | Not proof of incremental revenue |
Included costs | ¥1,000,000 product costs + ¥1,000,000 advertising | Deduct ¥2,000,000 once |
Ad-spend-denominator ROI | ¥500,000 / ¥1,000,000 × 100 = 50% | Investment defined as ad spend |
Total-cost-denominator ROI | ¥500,000 / ¥2,000,000 × 100 = 25% | Investment includes product costs |
Both percentages use the same profit. They differ because they express return relative to different investments. Do not compare them across campaigns as though their denominators were identical.
Separate break-even from cash recovery
With a positive investment denominator, zero profit after all included costs means 0% ROI. A 100% ROI means profit equals the investment amount. Treating 100% as the break-even threshold demands an additional return equal to an investment that has already been deducted.
A profitable initiative can still create a cash shortage when receivables arrive after expenses are paid. Check prepayments, customer payment dates, inventory, and taxes separately. An ROI excluding overhead, production, or management fees also does not represent the company's final net profit.
Do not invent monetary returns for nonmonetary outcomes
Training, awareness, and customer experience may lack a defensible immediate monetary value. HM Treasury's Green Book 2026 addresses cost-effectiveness and effects that cannot be monetized in public-policy appraisal. Its context is public policy, but the distinction helps explain why a valid outcome measure is preferable to an unsupported monetary ROI.
In a separate fictional training example, program A costs ¥100,000 and yields 20 verified completions; B costs ¥150,000 and yields 50. Cost per completion is ¥5,000 and ¥3,000 respectively. Compare these only if participant populations, completion criteria, learning quality, and periods align. Completion counts alone do not establish profit.
Include scale, timing, and incrementality
- Identify whether the decision is to continue, expand, or stop.
- Fix the revenue, cost, population, and observation period; separate actuals from forecasts.
- Review absolute profit, additional investment, and payment timing as well as ratios.
- Distinguish attributed outcomes from a demonstrated incremental effect.
- Test a manageable change and track both outcomes and additional costs.
A high historical ROI does not guarantee the same return on added spending. Read about reported ROAS and Target ROAS bidding for advertising decisions, and using LTV in marketing for longer-horizon customer value. Keep short-term actuals distinct from lifetime forecasts.