ROI Calculator

Return on investment is the most quoted performance figure in business and investing because it is trivially simple. It is also the easiest to make misleading, because it ignores two things that usually matter most: time and risk.

• A 25% ROI over one month and a 25% ROI over ten years are not comparable. The figure says nothing about duration. This is the single biggest source of misuse.
• Cost must be complete. Purchase price alone is not the cost if there were fees, commissions, holding costs, tax or the value of your time.
• Gross and net ROI differ. Net ROI subtracts all costs associated with generating the return; gross ROI does not. Only the net figure is useful for decisions.

The same trade looks excellent or mediocre entirely depending on how long the money was tied up. Annualising is what makes a six-month flip comparable to a ten-year hold - and it is the figure you should quote when comparing opportunities.

Note. Annualisation assumes the return could have been repeated each year. That assumption is often wrong for one-off deals, so treat the annualised figure as a comparison aid rather than a claim about achievable yearly returns.

• Time. Covered only by annualising. A 100% return over twenty years is 3.5% a year - unimpressive, but the raw figure looks superb.
• Risk. A 30% return from a speculative venture and a 30% return from a government bond show the same ROI. They are not the same achievement.
• Cash flow timing. ROI treats all gains as arriving at the end. Money received early can be reinvested; money locked up cannot.
• Opportunity cost. The return you did not earn elsewhere is not in the calculation.
• Tax. Unless you input after-tax figures, ROI overstates what you keep.
• Leverage. Borrowing to invest inflates both ROI and the risk. A 20% return on a 50% borrowed position is a much larger return on your own cash - and a much larger loss if it goes wrong.

None of these make ROI wrong. They make it a starting point rather than a conclusion.

Marketing campaign. Spending $4,000 produced $11,000 in attributable revenue on a product costing 40% of price to deliver. Gross ROI on revenue would be 175%. Net of delivery cost the gain is $2,600, so net ROI is 65%.

Property. Bought for $250,000, sold for $310,000 after 3 years, with $18,000 in fees, improvements and taxes. Net gain $42,000, total ROI 16.8%, annualised 5.3%. Compare that against the annualised figure of a low-cost index fund over the same three years before deciding the property outperformed.

ROI is most useful when you have several options and need to rank them quickly. The discipline is to compute each one the same way.

1. Use identical inputs. Same cost definition, same period, same treatment of tax. A comparison built from inconsistent inputs is worse than no comparison.
2. Annualise everything before ranking, so no option wins on duration alone.
3. Apply a risk haircut. If one option is materially riskier, its ROI needs to exceed the alternatives by a margin you consider sufficient. There is no formula for that margin - it is a judgement, but it should be made explicitly rather than skipped.
4. Include the do-nothing case. The baseline return on cash or a default investment is a legitimate alternative with its own risk profile.
5. Look at total return, not just ROI percentage. A 50% return on $500 earns $250. A 20% return on $50,000 earns $10,000. Percentage alone can point you at the wrong-sized decision.