Knowledge base / Excel
LOG10 in Excel: base-ten logarithms, LOG and LN
I explain logarithms through powers of ten, compare LOG with LN and show why adding one changes the analysis.

LOG10 asks which power of 10 produces a given positive number. For 100, the answer is 2 because ten squared is 100. I start with that relationship because it makes the result easier to interpret. A logarithm is different from dividing the input by ten.
Syntax and availability
The function has the same name in English and Polish Excel:
=LOG10(A2)
Its argument must be a positive real number. Microsoft’s LOG10 documentation defines the function and lists versions including Microsoft 365, Excel 2024, 2021, 2019 and 2016. It does not require dynamic arrays.
The English examples use decimal points and commas between arguments where needed. Regional settings can differ. LOG10 itself takes one argument, so a whole-number example has no argument separator to confuse with a decimal separator.
Derive simple results from powers
These synthetic inputs make the definition visible:
| Input | Formula | Result |
|---|---|---|
| 1 | =LOG10(1) |
0 |
| 10 | =LOG10(10) |
1 |
| 100 | =LOG10(100) |
2 |
| 1000 | =LOG10(1000) |
3 |
| 0.1 | =LOG10(0.1) |
-1 |
| 0.01 | =LOG10(0.01) |
-2 |
Ten to the power zero equals one. Ten to the power minus one equals one tenth. A negative result is therefore valid for a positive input below one. Do not confuse a negative logarithm with attempting to calculate the real logarithm of a negative input.
An input between 10 and 100 has a result between 1 and 2. Results need not be integers. Values other than simple powers generally require an approximation, and cell formatting controls how many decimal places are displayed.
LOG10, LOG and LN use specific bases
LOG without an explicit base calculates a base-ten logarithm. Microsoft defines its default base as 10. For a positive input, these expressions therefore describe the same base-ten calculation:
=LOG10(100)
=LOG(100)
=LOG(100,10)
Each gives 2 here. To request another base, specify it explicitly. For example, =LOG(8,2) gives 3 because two cubed equals eight. The syntax and default are documented in LOG.
Use LN for the natural logarithm, whose base is e. There is no need to type an approximation of e when the task specifically requires a natural logarithm. See Microsoft’s LN reference. Name the base in the column description because “log” alone can mean different things across fields.

Understand what the transformed scale represents
Compare synthetic values of 10, 100 and 1000. Successive differences are 90 and 900, while the ratio is ten in both steps. LOG10 produces 1, 2 and 3. Equal distances on this scale therefore correspond to equal multiplicative changes, making it useful for quantities spanning orders of magnitude.
The transformation does not guarantee a normal distribution or a valid statistical model. I avoid using “normalization” as a promise that any dataset will be repaired. Distribution, units and the analytical question still require attention. A logarithm changes how differences are represented, rather than improving source quality or resolving the meaning of an observation.
Also distinguish transforming a column from setting a chart axis to logarithmic spacing. The former creates transformed measure values; the latter changes their visual placement. Describe which operation you used so another analyst does not transform an already transformed column while assuming it contains original amounts.
Zeros, missing values and negative inputs
For real-valued calculations, LOG10 requires an input greater than zero. Zero and negative inputs fall outside that domain. Before applying the function to a column, inspect those records as well as missing values and numbers stored as text. Decide how to handle them separately from writing the expression.
Do not automatically add one merely to avoid a zero problem. LOG10(x+1) is a different transformation: when x is 9, it gives 1, whereas LOG10(9) is below 1. The shift changes interpretation and depends on the data’s unit. It can be a deliberate methodological choice, but it needs a reason.
Similarly, taking absolute values to conceal negative inputs removes the distinction between a loss and an equally sized gain. If the sign matters, choose an analysis that preserves it. The ABS entry explains why changing a sign changes the information represented by the value.
Exercise and next step
Create a column of powers of ten and their reciprocals. Write expected logarithms first by identifying the exponent. Add zero and explain why it needs a separate treatment. The absence of a real logarithm does not justify silently changing the input until the formula returns a number.
Finally, compare LOG10(100), LOG(100) and LN(100). The first two use the same base; the third uses another. Continue with the LOG entry for a wider discussion of bases. In a report, retain the transformation name, its base and the handling rule for inputs outside its domain.
Key takeaways
Does LOG default to the natural logarithm?
No. Excel LOG defaults to base 10; LN calculates the natural logarithm.
Can LOG10 return a negative value?
Yes. A positive input below 1 has a negative logarithm, such as LOG10(0.1) = -1.
Should I automatically add one to every input?
That changes the transformation and its interpretation. Use a shift only with a reason appropriate to the data and its unit.
AI was used in an automated way while working on this article. If you want to learn how to do it too, check out this project.
