๐ Geometric Mean Calculator: Find the Nth Root of a Product
By Shihab Mia ยท Updated 2026-07-17
The geometric mean is the nth root of the product of your values: multiply all n numbers together, then take the nth root. For 2 and 8 it is the square root of 16, which is 4. Use this geometric mean calculator for anything that grows by multiplying, such as investment returns, population growth, price indexes or bacteria counts, where the ordinary average gives the wrong answer. Paste your positive numbers below and you get the geometric mean, the arithmetic mean beside it for comparison, the product, and the count. Values must be positive: one zero collapses the product, and a negative value would need the root of a negative number.
What is the Geometric Mean Calculator?
The geometric mean of n positive numbers is the nth root of their product. For two numbers it is the square root of their product, for three the cube root, and so on. Written out, GM = (x1 x x2 x ... x xn)^(1/n). It answers a different question from the familiar arithmetic mean: instead of asking what single value, added n times, gives the same total, it asks what single value, multiplied by itself n times, gives the same product. That makes it the natural average whenever quantities combine by multiplication rather than addition. The same idea can be written with logarithms: the log of the geometric mean equals the arithmetic mean of the logs, so GM = exp((ln x1 + ln x2 + ... + ln xn) / n). That identity is not just a curiosity, it is how this geometric mean calculator computes the result internally.
The classic use is averaging growth rates and returns. If an investment grows 10 percent one year and falls 10 percent the next, the arithmetic mean of +10 and -10 is zero, yet you actually end up with 99 percent of your money. Working with the growth factors 1.10 and 0.90, the geometric mean is the square root of 0.99, about 0.99499, which correctly says the portfolio shrank about 0.5 percent per year on average. This is why fund performance is reported as a compound (geometric) return rather than a simple arithmetic one, and it is exactly the compound annual growth rate: CAGR = (ending value / beginning value)^(1/years) - 1, which is the geometric mean of the yearly growth factors minus 1.
Three facts are worth committing to memory. First, for positive data the harmonic mean is less than or equal to the geometric mean, which is less than or equal to the arithmetic mean (HM <= GM <= AM), with equality only when every value is identical. The wider the spread in your values, the bigger the gaps. Second, the geometric mean is only defined for positive numbers, so convert percentage changes into growth factors (a 5 percent fall becomes 0.95, never -5) before averaging. Third, the geometric mean is not distorted by a single huge outlier the way the arithmetic mean is, because it averages multiplicative distance rather than additive distance. Average 1 and 1000 arithmetically and you get 500.5, which resembles neither number; geometrically you get 31.6, which sits sensibly between them on a log scale.
Geometric means run quietly through official statistics. The US Bureau of Labor Statistics has used a geometric mean formula to average prices inside most CPI item categories since January 1999, because it lets the index reflect shoppers substituting between similar goods when relative prices move. The UN Development Programme builds the Human Development Index as the geometric mean of its three normalized dimension indices (health, education and income), deliberately so that a country cannot fully offset a terrible score in one dimension with a brilliant score in another. The US Environmental Protection Agency writes recreational water quality criteria as geometric means of bacteria counts, for example an enterococci geometric mean of 35 CFU per 100 mL for marine waters alongside a statistical threshold value of 130, because bacteria counts are roughly log-normal and one storm-driven spike would otherwise wreck a simple average.
Geometry gives the mean its name. In a right triangle, the altitude drawn from the right angle to the hypotenuse is the geometric mean of the two segments it cuts the hypotenuse into: h = sqrt(p x q). This is the geometric mean theorem, also called the altitude rule, and it is why the geometric mean of two numbers is sometimes called their mean proportional: a / g = g / b means g = sqrt(a x b). The same construction squares a rectangle, producing a square of side sqrt(a x b) with exactly the same area as an a by b rectangle. Any time you need one number to stand in for two multiplied dimensions, the geometric mean is the number you want.
One practical note on accuracy. Multiplying a long list of values directly can overflow past what a computer can store, or underflow to zero, long before you get to take the root. Averaging the natural logs and exponentiating at the end sidesteps that completely and returns the identical answer, so this tool stays exact on lists that would break a naive product. If you prefer a spreadsheet, Excel and Google Sheets both provide GEOMEAN, which accepts a range and ignores empty cells and text but errors on zero or negative values, exactly as the mathematics requires.
When to use it
- Turning several years of investment returns into one compound annualized figure (CAGR) that matches your actual ending balance.
- Averaging growth rates for revenue, users, population or traffic, where each period multiplies the last rather than adding to it.
- Combining ratios and index numbers, such as price relatives in a consumer price index or performance benchmark ratios.
- Building composite scores from indicators measured on different scales, the way the Human Development Index averages its three dimension indices.
- Averaging environmental or lab data that is log-normal and spiky, such as bacteria counts in water quality monitoring.
- Solving geometric mean theorem problems: finding the altitude of a right triangle or the mean proportional between two lengths.
How to use the Geometric Mean Calculator
- Type or paste your positive numbers, separated by commas, spaces, or new lines.
- For growth rates, enter growth factors rather than percentages: a 10 percent rise is 1.10, a 10 percent fall is 0.90.
- Read off the geometric mean, shown next to the arithmetic mean so you can see the gap between them.
- Check the count and product fields to confirm every value was read correctly.
- To convert a geometric mean of growth factors back into a rate, subtract 1 and multiply by 100. A result of 1.0677 means about +6.77 percent per period.
Formula & method
Worked examples
Find the geometric mean of 2 and 8.
- Multiply the values: product = 2 x 8 = 16
- There are n = 2 values, so take the square root
- GM = 16^(1/2) = 4
- For comparison, arithmetic mean = (2 + 8) / 2 = 5
Result: Geometric mean = 4 (arithmetic mean = 5)
Find the geometric mean of 1, 3 and 9.
- Multiply the values: product = 1 x 3 x 9 = 27
- There are n = 3 values, so take the cube root
- GM = 27^(1/3) = 3
- For comparison, arithmetic mean = (1 + 3 + 9) / 3 = 4.333
Result: Geometric mean = 3 (arithmetic mean = 4.333)
Average four yearly growth factors: 1.10, 0.90, 1.25, 1.05, and read the CAGR.
- Convert each percentage change to a factor first (+10% to 1.10, -10% to 0.90, +25% to 1.25, +5% to 1.05)
- Multiply the factors: product = 1.10 x 0.90 x 1.25 x 1.05 = 1.299375
- There are n = 4 values, so take the fourth root
- GM = 1.299375^(1/4) = about 1.06774
- Subtract 1 to get the rate: 1.06774 - 1 = 0.06774, so about +6.77 percent per year
- Check it: 1.06774^4 = 1.2994, matching the real 29.94 percent total growth
- The arithmetic mean of the factors is 4.30 / 4 = 1.075, which would overstate growth to +7.5 percent
Result: Geometric mean about 1.0677, a CAGR of about +6.77% per year (arithmetic mean = 1.075)
Geometric mean versus arithmetic mean for sample data sets
| Values | Geometric mean | Arithmetic mean | Gap |
|---|---|---|---|
| 4, 4, 4 | 4 | 4 | None, all values identical |
| 2, 8 | 4 | 5 | 1.00 |
| 1, 3, 9 | 3 | 4.333 | 1.333 |
| 1.10, 0.90 | 0.99499 | 1.000 | 0.005 |
| 1, 100 | 10 | 50.5 | 40.5 |
| 1, 1000 | 31.623 | 500.5 | 468.9 |
Which average to use for which kind of data
| Situation | Best average | Why |
|---|---|---|
| Adding up quantities (heights, test scores) | Arithmetic mean | Values combine by addition |
| Multiplying rates or returns over time | Geometric mean | Values compound period to period |
| Averaging ratios or index numbers | Geometric mean | Ratios are multiplicative |
| Data spanning several orders of magnitude | Geometric mean | Resists single large outliers |
| Averaging speeds over a fixed distance | Harmonic mean | Rates share a fixed numerator |
| Data containing zero or negative values | Arithmetic mean | Geometric mean is undefined there |
Converting percentage changes into growth factors before averaging
| Percentage change | Growth factor to enter |
|---|---|
| +25% | 1.25 |
| +10% | 1.10 |
| +5% | 1.05 |
| 0% (no change) | 1.00 |
| -5% | 0.95 |
| -10% | 0.90 |
| -50% | 0.50 |
Common mistakes to avoid
- Including zero or negative numbers. The geometric mean is only defined for positive values. A single zero makes the whole product zero, and a negative value would require the root of a negative number. Convert percentages to positive growth factors (1.05, not -0.05) before averaging. Excel and Google Sheets GEOMEAN return an error for the same reason.
- Averaging percentages instead of growth factors. Do not take the geometric mean of figures like 10 and -10. Turn each percentage change into a factor first: +10 percent becomes 1.10 and -10 percent becomes 0.90, average those to get 0.99499, then subtract 1 to read it back as about -0.5 percent per period.
- Using the arithmetic mean for compound returns. The arithmetic mean overstates average growth whenever the values vary, because a percentage gain and an equal percentage loss do not cancel. Use the geometric mean to report the true compound (annualized) rate, which is what actually reproduces your ending balance.
- Forgetting to subtract 1 at the end. The geometric mean of growth factors is itself a factor, not a rate. A result of 1.0677 is not 1.07 percent growth, it is about 6.77 percent growth. Subtract 1 and multiply by 100 to convert.
- Confusing the geometric and harmonic means. They are different averages. The geometric mean suits rates that multiply over time, while the harmonic mean suits averaging rates over a fixed quantity, such as speeds over the same distance. For positive data the ordering is always HM <= GM <= AM.
- Multiplying a long list directly and hitting overflow. Products of many values grow astronomically fast and can overflow a calculator or underflow to zero before you take the root. Average the natural logs and exponentiate instead. This tool does that for you, so the answer stays exact on long lists.
Glossary
- Geometric mean
- The nth root of the product of n positive values, the average for quantities that combine by multiplication.
- Arithmetic mean
- The familiar average: the sum of the values divided by how many there are.
- Harmonic mean
- The reciprocal of the arithmetic mean of the reciprocals, used for averaging rates over a fixed quantity. It is always the smallest of the three means for positive data.
- Growth factor
- A multiplier representing a percentage change, where +5 percent is 1.05 and -5 percent is 0.95.
- Product
- The result of multiplying all the values together, the quantity whose nth root is taken.
- nth root
- The value that, raised to the power n, gives the original number. The square root is the 2nd root, the cube root the 3rd.
- CAGR
- Compound annual growth rate: the single constant yearly rate that reproduces actual cumulative growth, found as the geometric mean of the yearly growth factors minus 1.
- Geometric mean theorem
- In a right triangle, the altitude to the hypotenuse equals the geometric mean of the two hypotenuse segments it creates: h = sqrt(p x q). Also called the altitude rule.
- Mean proportional
- Another name for the geometric mean of two numbers, from the proportion a / g = g / b, which gives g = sqrt(a x b).
- Log-normal data
- Data whose logarithms are normally distributed, common in bacteria counts, incomes and particle sizes. Its typical value is best summarised by the geometric mean.
Frequently asked questions
What is the geometric mean?
The geometric mean of n positive numbers is the nth root of their product. For two numbers it is the square root of their product, for three the cube root, and so on. It is the right average when values combine by multiplication, such as growth rates and ratios.
What is the geometric mean formula?
GM = (x1 x x2 x ... x xn)^(1/n). Multiply all n values, then raise the product to the power 1/n. An equivalent form using logarithms is GM = exp((ln x1 + ln x2 + ... + ln xn) / n), which gives the same answer without overflowing on long lists.
How do I find the geometric mean of two numbers?
Multiply the two numbers and take the square root: GM = sqrt(a x b). The geometric mean of 2 and 8 is sqrt(16) = 4, and the geometric mean of 4 and 9 is sqrt(36) = 6. This value is also called the mean proportional between the two numbers.
How is the geometric mean different from the arithmetic mean?
The arithmetic mean adds the values and divides by the count, suiting quantities that add up. The geometric mean multiplies the values and takes the nth root, suiting quantities that multiply, like returns over time. For positive data the geometric mean is always less than or equal to the arithmetic mean.
Why must all the numbers be positive?
The geometric mean relies on a product and a root. A zero makes the product zero and collapses the result to zero, while a negative value would require taking a root of a negative number, which is not a real number. So only positive values are allowed.
Can you calculate a geometric mean with negative numbers?
Not directly. The standard fix is to work with positive growth factors instead: a change of -10 percent is entered as 0.90, not as -10. Some tools shift every value up by a constant before averaging, but that changes the answer and is not a true geometric mean, so report it carefully if you do it.
How do I use the geometric mean for investment returns?
Convert each periodic change into a growth factor (a 7 percent gain is 1.07, a 4 percent loss is 0.96), take the geometric mean of those factors, then subtract 1. The result is the compound (annualized) return that matches your actual cumulative growth.
Is the geometric mean the same as CAGR?
Effectively yes. CAGR is the geometric mean of the periodic growth factors minus 1. Computing (ending value / beginning value)^(1/years) - 1 gives the identical number, because the ratio of ending to beginning value is exactly the product of all the yearly growth factors.
Is the geometric mean always smaller than the arithmetic mean?
For positive numbers it is always less than or equal to the arithmetic mean, and they are equal only when every value is identical. The more spread out the values, the larger the gap. The full ordering for positive data is harmonic mean <= geometric mean <= arithmetic mean.
How do I calculate a geometric mean in Excel or Google Sheets?
Use =GEOMEAN(A1:A10). Both Excel and Google Sheets ignore empty cells and text in the range but return an error if any value is zero or negative. For growth rates, put the growth factors in the range, then subtract 1 from the result to read it as a percentage.
Why does this tool use logarithms instead of multiplying directly?
Multiplying many values can overflow into a number too large to store, or underflow to zero, before the root is taken. Computing exp of the average of the natural logs gives the identical geometric mean without that risk, so the result stays accurate even on long lists.
What is the geometric mean theorem in geometry?
In a right triangle, the altitude drawn from the right angle to the hypotenuse is the geometric mean of the two segments it splits the hypotenuse into: h = sqrt(p x q). If the segments are 4 and 9, the altitude is sqrt(36) = 6. It is also known as the altitude rule.
Sources
- Consumer Price Index Frequently Asked Questions (geometric mean formula used for most basic indexes since January 1999) , US Bureau of Labor Statistics
- Human Development Report Technical Notes: the HDI is the geometric mean of the three dimension indices , United Nations Development Programme (2025)
- Recreational Water Quality Criteria (enterococci geometric mean of 35 CFU/100 mL with an STV of 130 for marine waters) , US Environmental Protection Agency, Office of Water 820-F-12-058 (2012)
- GEOMEAN function documentation , Microsoft Support