Return on Investment Calculator

Return on investment = (Total returned − Initial investment) ÷ Initial investment × 100

Initial investment

What went in at the start, as a single figure; it is both the thing being measured and the denominator

Amounts returned

One number per period, in order, separated by spaces or commas — the length of this list is what defines how many periods the investment ran for

Total returned

The sum of that list, added at face value with no discounting; the page prints it so the percentage can be checked against the raw arithmetic

Net profit

Total returned minus the initial investment — the same quantity as the numerator of the ratio, in money rather than in percent

Return on investment

Net profit divided by the initial investment, as a percentage; it can be negative down to −100%, which is the case where nothing at all came back

Number of periods

How many figures were in the list; printed because a return means little without it, and because it is the only thing on the page that says how long the money was tied up

Use it when the returns arrive on a schedule you can write down and you want the plain answer to how much came back relative to what went in — a piece of equipment, a marketing campaign, a property with annual cash flow, anything where you can list the yearly figures. Use it as the first screen on a deal, before deciding whether the timing is worth modelling properly. What it is not is a rate: because nothing is discounted, it cannot be compared with an interest rate, a bond yield or a hurdle rate, and two deals with the same total returned but different timings look identical here even though they are not. It also ignores how long the money was tied up — that is what the period count beside it is for, and what the annualized return page does properly. And it says nothing about risk or about what else you could have done with the money.

1. 10,000 in, five equal returns of 2,500

   1. Add the list at face value: 2,500 × 5 = 12,500 returned
   2. Net profit: 12,500 − 10,000 = 2,500
   3. Return on investment: 2,500 ÷ 10,000 = 0.25, so 25.00%
   4. Count the periods: five figures in the list, so the money was tied up for five periods

   The default, and the shape most people mean by a return on investment. The 25% is the whole answer to how much came back, and it is deliberately not annualized: spread over five periods it is nothing like 25% a year, and the period count printed beside it is the only hint on the page of that. Note that the same 12,500 returned as 10,000 in the first period and 2,500 in the second gives exactly the same 25% here, because nothing is discounted.

2. Breaking even: 50,000 in, 20,000 + 15,000 + 10,000 + 5,000 back

   1. Add the list: 20,000 + 15,000 + 10,000 + 5,000 = 50,000
   2. Net profit: 50,000 − 50,000 = 0
   3. Return on investment: 0 ÷ 50,000 = 0.00%
   4. Four figures, so four periods — and the returns are shrinking as they go

   A zero return is a real answer and this is where the page stops looking like a sales pitch. Half the money came back in the first period and only a tenth in the last, which is the pattern that looks healthy month to month and ends up exactly even. Read against the previous example, this is also the clearest demonstration that timing is invisible here: 50,000 back is 50,000 back whether it arrives immediately or over four years, and the only thing that distinguishes the two schedules on this page is the order of the numbers in a list whose total is the same.

3. A loss: 20,000 in, 4,000 + 3,000 back

   1. Add the list: 4,000 + 3,000 = 7,000
   2. Net profit: 7,000 − 20,000 = −13,000
   3. Return on investment: −13,000 ÷ 20,000 = −0.65, so −65.00%
   4. Two figures, two periods — and both of them are printed as negative money on the panel

   The page does not warn, colour the result red or refuse to answer; a negative return on investment is a normal output, and the floor is −100%, which is the day nothing at all comes back. Two things are worth noticing about this one. The money did come back — 7,000 of it, in two periods, with no period showing zero — so a reader checking period by period would see two payments and no obvious failure; the loss lives entirely in the comparison with what went in. And the −65% is not a rate of anything: it is a total, and spread over two periods it is worse than −65% a year, again the reason the period count is on the panel.

Three limits, and the first is the one that matters most. Nothing here is discounted: every period's amount is added at face value, so money returned in period one and money returned in period twenty count the same, and the result is a ratio rather than a rate. That is why it cannot be compared with an interest rate, a bond yield or a hurdle rate — for that, the cash flows have to be discounted, which is what the net present value page does. Second, it does not annualize. A 25% return over two periods and a 25% return over ten periods are the same number on this page, and the period count beside it is the only thing that separates them; the annualized return page is the tool for the rate. Third, it is a single-path answer: it takes the amounts you type as given, so it cannot express a range of outcomes, a probability of not being repaid, or the fact that the later a payment is, the less likely it is to arrive at all. Treat it as the first screen on a deal, not the last.

How is this different from the plain return on investment calculator?

Only in the shape of the input. Both divide net profit by the initial investment, and both take a single initial investment. Here the returns are a list — one figure per period, as many as you type — where the simpler page takes one amount returned. So this page answers what came back over a schedule and prints the number of periods, while the other answers a single exchange. If your money came back once, use the other one; if you can write down a figure per year, use this one.

Why is nothing discounted?

Because that is what a return on investment is, and doing it any other way would make the number something else. Each period's amount is added at face value, so the total returned is literally what you got. The cost is that timing disappears: 12,500 returned as five equal 2,500 payments and 12,500 returned as 10,000 in the first period and 2,500 in the second both give 25% here. If the timing matters to the decision, the cash flows have to be discounted, which is the net present value page's job.

How do I enter the amounts returned?

One number per period, in order, separated by spaces or commas — 2500 2500 2500 2500 2500 for five years of 2,500. The list's length is what sets the period count, so there is no separate field for it, and adding a figure lengthens the investment. The amounts are the cash you received, so they should not be negative: a period in which you paid more in is not a return, and entering a negative figure is rejected rather than quietly netted off.

What does the number of periods tell me that the percentage does not?

How long the money was tied up, which the percentage is silent about. A 25% return over two periods is roughly 12% a period; the same 25% over five periods is roughly 4.6% a period, and the two are very different deals. Because this page does not annualize, the period count is the only place that information appears, so read the two together. It is also the reason the same percentage means different things in different industries: a return on investment of 25% is ordinary for a project measured in months and remarkable for one measured in decades.

Can the return on investment be negative, and how negative?

Yes, and the floor is −100%, which is the case where nothing at all came back. A loss is a normal output here and the page does not flag it specially. The percentage is measured against the initial investment, so −65% means 65% of what you put in is gone, not that you received 65% less than you hoped. Note that a deal can look fine period by period and still be negative overall — every payment can arrive on time and still total less than the initial investment, which is exactly what happens in a deal whose later periods were overestimated.

Should the initial investment include costs beyond the purchase price?

It should be everything you had to commit at the start, because it is the denominator. That normally means the purchase price plus the costs of getting the thing running: fees, installation, closing costs, the initial outlay on anything that had to be bought before any return could arrive. Do not put ongoing costs in here, though. Money you spend in a later period is not part of the initial investment, and the honest way to handle it in this model is to net it against that period's return — enter the period's cash after those costs, rather than inflating the denominator with them.

• Circular A-94, Guidelines and Discount Rates for Benefit-Cost Analysis of Federal Programs — why cash flows at different dates are not added together at face value when the timing is meant to be part of the answer — Office of Management and Budget, Executive Office of the President (United States)
• Annual Return — Investor.gov glossary, for the annualizing step this page deliberately does not take — U.S. Securities and Exchange Commission, Investor.gov (United States)
• Compound Interest — Investor.gov glossary, the effect that makes a dollar returned later worth less than a dollar returned now — U.S. Securities and Exchange Commission, Investor.gov (United States)