ROI Calculator

How to use this calculator

Unlike the projection tools on this site, this one looks backward: you already know what went in and what came out, and the job is to measure it properly.

• Amount invested: the purchase price or capital committed.
• Amount received / current value: sale proceeds for a closed position, or today's value for an open one.
• Additional costs: fees, commissions, transaction taxes. These join the invested amount in the denominator, which is where most DIY ROI figures go wrong.
• Holding period: how long the money was committed; this converts the raw percentage into an annualized rate.

The table re-annualizes the same total return over 1, 2, 3, 5 and 10 years. It's worth a look once, because it shows how completely the holding period changes what a "40% gain" means.

ROI vs annualized return — the trap in the headline number

Return on investment is the simplest metric in finance: net profit divided by what you put in. The default example ($10,000 in, $14,000 back) is a 40% ROI. Clean, intuitive, and dangerously incomplete, because it says nothing about time. A 40% gain in one year is exceptional; the same 40% spread over three years works out to 11.87% per year, and over ten years to just 3.42% — savings-account territory. Same ROI, radically different investments.

The per-year figure is the annualized return, or CAGR: the steady compound rate that would turn the start amount into the end amount over the holding period. The naive shortcut (40% ÷ 3 years ≈ 13.33%/yr) overstates the truth, because compounding means each year's growth builds on the last; the true geometric answer is (14,000 ⁄ 10,000)^1⁄3 − 1 = 11.87%. The gap between 13.33% and 11.87% is small here and grows with the numbers: averaging yearly gains always flatters volatile results. A portfolio that gains 50% then loses 50% has an "average return" of zero yet has lost 13.4% a year in reality; geometric math is the only version that matches your account balance.

Annualized return matters because it's the only number that transfers across investments of different sizes and durations. You can't rank a 25% ROI over two years against a 60% ROI over five until both become annual rates (11.80% and 9.86%, the smaller headline wins). It also lets you weigh a result against reference points: long-run historical averages are commonly quoted around 4% for savings, 5% for bonds, 8% for broad equities — history, not prophecy, but useful for judging whether a result compensated its risk.

The other discipline this calculator enforces is counting every cost. Fees, commissions, stamp duty and taxes paid belong in the denominator: they were capital you committed. Add $500 of costs to the default example and ROI drops from 40% to 33.33%, and the annualized rate from 11.87% to 10.06%. On thin-margin investments, ignored costs routinely turn paper profits into real losses.

Formula and methodology

Total return counts everything committed, including costs:

Annualizing converts the total into a compound yearly rate:

The two agree only at exactly one year. Shorter than a year, annualizing extrapolates (a 20% gain in 6 months annualizes to 44%); longer than a year, it compresses. Both formulas assume a single outlay and a single ending value; for drip-fed investing, an XIRR-style calculation is the right tool instead.

Worked example

What changes the result

• The holding period transforms the verdict. The same 40% total is 40%/yr over one year, 18.32% over two, 11.87% over three, 6.96% over five, 3.42% over ten. Never quote an ROI without its duration.
• Costs compound in reverse. Every dollar of fees both shrinks the profit and grows the base it's measured against: the $500 in the example cut the annualized rate by 1.8 points.
• Open positions flatter themselves. Using today's value assumes you could exit at it; illiquid assets (property, private stakes) often can't be sold at the marked price, net of selling costs.
• Risk is invisible in ROI. An 11.87%/yr result from a diversified fund and the same from a single speculative bet are not equivalent outcomes — one of them was luck-weighted.

Assumptions and limitations

• The model assumes one payment in, one value out. If you added or withdrew money along the way, CAGR misattributes the result; money-weighted measures (XIRR) handle dated cash flows.
• Don't annualize sub-year holdings for decision-making: a lucky 6-month trade annualized to 44% implies a repeatability nothing guarantees. The engine computes it; treat it as extrapolation, clearly labeled.
• The context bars (savings ~4%, bonds ~5%, equities ~8%) are rough historical long-run averages, shown only for scale: they are not predictions, offers, or targets, and any given decade lands well away from them.
• Inflation isn't deducted; a 3.42%/yr result during 3% inflation barely broke even in real terms.