How to Calculate the Geometric Mean in Excel

The geometric mean is the nth root of the product of n numbers. If that sounds abstract, here's the practical version: it's the correct way to average values that multiply together, like growth...

Key Insights

  • The geometric mean is essential for calculating average rates of return, growth percentages, and any multiplicative data—using arithmetic mean in these scenarios produces mathematically incorrect results
  • Excel’s built-in GEOMEAN() function handles most cases, but understanding the manual calculation methods (PRODUCT/POWER or EXP/LN) gives you flexibility when dealing with edge cases
  • Zero and negative values break geometric mean calculations entirely, requiring data transformation or alternative approaches before you can proceed

Introduction to Geometric Mean

The geometric mean is the nth root of the product of n numbers. If that sounds abstract, here’s the practical version: it’s the correct way to average values that multiply together, like growth rates, percentages, and ratios.

Most people default to the arithmetic mean (add everything, divide by count) because it’s familiar. But arithmetic mean fails spectacularly with multiplicative data. Consider an investment that gains 100% one year and loses 50% the next. The arithmetic mean suggests a 25% average return. The reality? You’re back where you started—0% actual return. The geometric mean correctly calculates this.

The formula for geometric mean is:

GM = (x₁ × x₂ × x₃ × ... × xₙ)^(1/n)

Or equivalently: take the product of all values, then raise it to the power of 1 divided by the count.

When to Use Geometric Mean

Use geometric mean when your data represents:

Rates of return and growth rates. Investment performance, population growth, compound interest—anywhere values compound over time.

Ratios and proportions. Price-to-earnings ratios, aspect ratios, any comparison expressed as one value divided by another.

Percentage changes. Revenue growth percentages, inflation rates, year-over-year metrics.

Log-normally distributed data. Data that’s skewed right with a long tail, common in finance and biology.

The key indicator: if your values multiply together to produce a meaningful result, use geometric mean. If they add together, use arithmetic mean.

Here’s a concrete example of arithmetic mean failure. A stock goes up 50% year one, down 33.33% year two:

  • Year 1: $100 → $150 (×1.50)
  • Year 2: $150 → $100 (×0.6667)

Arithmetic mean of returns: (50% + -33.33%) / 2 = 8.33%

But you ended with exactly what you started with. The true average return is 0%.

Geometric mean: √(1.50 × 0.6667) = √1.0 = 1.0 → 0% return

The geometric mean gets it right.

Using the GEOMEAN Function

Excel provides the GEOMEAN() function for straightforward geometric mean calculations. The syntax is simple:

=GEOMEAN(number1, [number2], ...)

You can pass individual values, cell references, or ranges. For a dataset of growth multipliers in cells A1 through A10:

=GEOMEAN(A1:A10)

Important: GEOMEAN() expects positive numbers representing multipliers, not percentage changes. If your data shows percentage returns (like 10%, -5%, 15%), you must convert them to multipliers first by adding 1.

For percentage returns in column A, where A1 contains 10% (0.10):

=GEOMEAN(A1:A10+1)-1

This array operation adds 1 to each value, calculates the geometric mean, then subtracts 1 to convert back to a percentage return.

A practical example with explicit values:

=GEOMEAN(1.10, 1.15, 0.95, 1.08)

This calculates the geometric mean of four growth multipliers representing 10%, 15%, -5%, and 8% returns. The result is approximately 1.0679, indicating an average return of about 6.79% per period.

Manual Calculation Method

Understanding the manual calculation helps when GEOMEAN() isn’t suitable or when you need to customize the calculation.

Method 1: PRODUCT and POWER

The direct translation of the geometric mean formula:

=POWER(PRODUCT(A1:A10), 1/COUNT(A1:A10))

Breaking this down:

  • PRODUCT(A1:A10) multiplies all values together
  • COUNT(A1:A10) gives you n (the number of values)
  • 1/COUNT(A1:A10) gives you the exponent (1/n)
  • POWER() raises the product to that power

This method is transparent and easy to audit. You can see exactly what’s happening at each step.

Method 2: EXP and LN (Logarithmic approach)

This method uses logarithms to convert multiplication into addition:

=EXP(AVERAGE(LN(A1:A10)))

The mathematical principle: the log of a product equals the sum of logs. So:

  • LN(A1:A10) takes the natural log of each value
  • AVERAGE() computes the arithmetic mean of those logs
  • EXP() converts back from log space

This approach is numerically more stable for very large or very small numbers. It’s also useful when you’re already working with log-transformed data.

Both methods produce identical results for valid inputs. Choose based on your preference and context.

Handling Edge Cases

GEOMEAN() and manual methods share the same mathematical limitations. Here’s how to handle them.

Zero values

Any zero in your dataset makes the product zero, and the geometric mean becomes zero. This is mathematically correct but rarely useful.

Workaround 1: Filter out zeros before calculation:

=GEOMEAN(IF(A1:A10<>0, A1:A10))

Note: In older Excel versions (pre-365), enter this as an array formula with Ctrl+Shift+Enter.

Workaround 2: Replace zeros with a small positive number (use with caution—this changes your data):

=GEOMEAN(IF(A1:A10=0, 0.0001, A1:A10))

Negative values

Geometric mean is undefined for negative numbers. You can’t take the nth root of a negative product (in real numbers).

For data that can go negative, like profit/loss or temperature changes, geometric mean isn’t appropriate. Consider:

  • Using arithmetic mean instead
  • Working with absolute values if sign doesn’t matter
  • Transforming data (shift all values positive)

For percentage returns that go negative (like -20%), remember to convert to multipliers (0.80) which are always positive.

Empty cells and text

GEOMEAN() ignores empty cells and text values. This is usually what you want, but verify your COUNT() matches expectations in manual calculations.

=GEOMEAN(A1:A10)  ' Ignores blanks automatically

Very large or small numbers

With extreme values, PRODUCT() can overflow or underflow. The logarithmic method handles this better:

=EXP(AVERAGE(LN(A1:A100)))

This keeps intermediate calculations in a manageable range.

Practical Example: Investment Returns

Let’s build a complete investment return analysis. You have annual returns for a stock over 5 years and want to calculate the true average annual return.

Setup your data:

Row Column A (Year) Column B (Return %) Column C (Multiplier)
1 2019 15% =1+B1
2 2020 -8% =1+B2
3 2021 22% =1+B3
4 2022 -12% =1+B4
5 2023 18% =1+B5

Column B contains the raw percentage returns (0.15, -0.08, 0.22, -0.12, 0.18). Column C converts these to multipliers.

Calculate geometric mean return:

=GEOMEAN(C1:C5)-1

Result: approximately 6.14%

Compare with arithmetic mean:

=AVERAGE(B1:B5)

Result: 7%

The arithmetic mean overstates the actual return by nearly a full percentage point.

Verify with ending value:

Starting with $10,000:

=10000*PRODUCT(C1:C5)

Result: $13,471.94

Now calculate what 5 years at the geometric mean rate would produce:

=10000*(1+GEOMEAN(C1:C5)-1)^5

Result: $13,471.94

The geometric mean correctly predicts the ending value. Try the same with arithmetic mean:

=10000*(1+AVERAGE(B1:B5))^5

Result: $14,025.52

The arithmetic mean overshoots by over $500.

Complete formula reference for your worksheet:

' Cell D1: Geometric Mean Return
=GEOMEAN(C1:C5)-1

' Cell D2: Arithmetic Mean Return (for comparison)
=AVERAGE(B1:B5)

' Cell D3: Total Return over period
=PRODUCT(C1:C5)-1

' Cell D4: Verify ending value
=10000*PRODUCT(C1:C5)

Summary

The geometric mean is the mathematically correct average for multiplicative data. Excel makes it easy with GEOMEAN(), but understanding the manual methods gives you flexibility.

Quick reference:

Scenario Formula
Basic geometric mean =GEOMEAN(A1:A10)
Percentage returns to average return =GEOMEAN(A1:A10+1)-1
Manual (PRODUCT/POWER) =POWER(PRODUCT(A1:A10),1/COUNT(A1:A10))
Manual (logarithmic) =EXP(AVERAGE(LN(A1:A10)))
Excluding zeros =GEOMEAN(IF(A1:A10<>0,A1:A10))

Decision rule: If your values compound (multiply together over time or across factors), use geometric mean. If they accumulate (add together), use arithmetic mean. When in doubt, ask yourself: “Does the product of these values have meaning?” If yes, geometric mean is your tool.

Liked this? There's more.

Every week: one practical technique, explained simply, with code you can use immediately.