Skip to content
ToolShelf

ROI calculator

Work out the return on an investment or a campaign as a percentage and a multiple, with the annualised figure when you say how long the money was tied up.

The investment

What went in

The figure you have is

Everything you got back, including the original amount

How long the money was in

Optional, for the annualised rate

Without this you get the plain return, which ignores time entirely, 30% in one year and 30% over ten look identical until it is filled in.

Return on investment

—

Enter what went in and what came back.

The formula, and the word that ruins it

ROI is one subtraction and one division: what came back, minus what went in, divided by what went in. Put in $2,000, get back $2,600, and the $600 of gain is a 30% return.

The word that causes trouble is “return”, because it is used for both the total and the profit. “$2,600 back” is 30%. “$2,600 of return”, meaning profit on top of the original, is 130%. Same numbers, and more than four times the answer. That is why this tool asks which figure you have rather than picking one and hoping.

Plain ROI has no sense of time

It is the measure’s biggest weakness and the easiest to forget. A 30% return is excellent over a year and poor over a decade, and the percentage is identical either way.

The annualised figure fixes it by asking what constant yearly rate would compound to the same result. That 30% is:

  • over 1 year, 30% a year
  • over 2 years, 14.02% a year
  • over 5 years, 5.39% a year
  • over 10 years, 2.66% a year

Notice that it is not 30 divided by the years. Growth compounds, so the yearly rate falls more slowly than a straight division would suggest, which is also why a rough “divide by the years” understates a long investment.

Annualising a few months tells you less than it appears to

The arithmetic runs happily in the other direction, and it is worth knowing what it is saying when it does. A 10% gain in a single month annualises to 213.8% a year, not because that is what the year will do, but because that is what twelve identical months would compound to.

It is a fair way to compare two short investments with each other. It is not a forecast, and the tool labels it whenever the period entered is under a year.

ROI on a marketing campaign

The spend is the investment. The return should be the profit the campaign generated, not the revenue, using revenue treats the cost of delivering what you sold as though it were profit, and can make a campaign that lost money look like a success.

The harder problem is one no calculator can solve: attribution. ROI credits the campaign with everything that happened in the window, including the sales that would have arrived anyway. Treat the figure as a comparison between campaigns rather than as a measurement of what one campaign caused.

Where this is the wrong measure

Money added or taken out along the way. ROI compares one amount in against one amount out. A position you kept topping up needs a money-weighted return instead, a spreadsheet’s XIRR function is the usual way to get one.

Anything where risk differs. A 10% return on a guaranteed one-year deposit and a 10% return on a speculative venture are not the same result, and no percentage can say so.

And this is a nominal figure throughout. Tax on the gain, transaction and management fees, and inflation over the period all reduce what you actually kept, and none of them are included. Over a long holding period inflation in particular matters, subtracting it from the annualised rate gives a rough real return.

Questions

How do I calculate ROI?
Subtract what you put in from what came back, divide by what you put in, and multiply by 100. Putting in $2,000 and getting back $2,600 gives (2,600 − 2,000) ÷ 2,000 = 0.3, which is a 30% return. If you only know the profit rather than the total, that is the numerator already, $600 ÷ $2,000 is the same 30%.
Is the return the total that came back or just the profit?
It depends which you were told, and the difference is enormous: $2,600 back on $2,000 is a 30% return, while $2,600 of profit on $2,000 is a 130% return. This is the most common way an ROI figure gets quoted wrongly, which is why the tool asks rather than guessing. Check which one your figure is before comparing it with anyone else's.
What is annualised ROI, and why does it matter?
It is the constant yearly rate that would compound to the same result, (end ÷ start) to the power of 1/years, minus 1. Plain ROI ignores time entirely, so a 30% return looks identical whether it took one year or ten. Over two years that 30% is 14.02% a year; over ten it is 2.66%. Annualising is the only way to compare investments of different lengths.
Can I annualise a return from a few months?
You can, and the tool will, but read it carefully. Annualising assumes the same result repeats for the rest of the year: a 10% gain in one month annualises to 214%, which describes that month rather than predicting the next eleven. It is useful for comparing two short investments with each other and misleading as a forecast, so the figure is labelled wherever a period under a year is entered.
How do I calculate ROI on a marketing campaign?
Put the campaign spend in as the amount invested and the profit it generated, not the revenue, as the return. Using revenue counts the cost of delivering what you sold as if it were profit and can make a loss-making campaign look successful. The harder problem is attribution: ROI credits the campaign with every sale in the window, including the ones that would have happened anyway.
What is a good ROI?
There is no cross-industry answer, and any figure quoted as one is describing a particular market. A return has to be judged against how long it took, how much risk it carried, what else the money could have done, and inflation over the period. A 10% return on a guaranteed one-year investment and a 10% return on a speculative five-year one are not comparable results.
Does this account for inflation, tax or fees?
No. It measures the nominal cash return: what went in against what came back. Tax on the gain, transaction fees, management charges and the fall in what money buys over the period all reduce the real return and none of them are modelled. Over a long holding period inflation in particular makes a substantial difference, subtract it from the annualised figure for a rough real return.
What if I added or withdrew money along the way?
Then this is the wrong measure. ROI compares one amount in against one amount out, and money moving in or out during the period breaks that. What you need is a money-weighted return, which accounts for when each amount arrived, a spreadsheet's XIRR function is the usual way to get one. Using ROI on a position you kept topping up will overstate or understate the result depending on the timing.

More tools

Business & freelancing

Hourly rate calculator

What to charge, once the unbillable hours are counted

$60,000 take-home, $6,000 costs, 20% tax$67.50 over 1,200 billable hours
Business & freelancing

Freelance project quote calculator

A fixed price from an hours estimate, with room to be wrong

40 hours at $75, 10% contingency, 20% margin$4,125.00
Business & freelancing

Commission calculator

Flat or tiered, before and after the split

$60,000 sale, 3% to $50,000 then 6%$2,100.00