Knowledge base / Excel
TRUNC in Excel: removing digits and choosing between TRUNC and INT
I explain how to discard digits without conventional rounding. Small examples show why the sign matters and why changing a number's appearance differs from changing its value.

Sometimes a calculation needs a number with some digits discarded. Excel’s TRUNC function does that. I first decide whether the task needs truncation, conventional rounding or just a different display format. Those choices can produce different results in the next step of a report.
This revision also corrects the Polish partner of this page. Its old text described removing spaces under the incorrect function name ODETNIJ. The correct Polish name for TRUNC is LICZBA.CAŁK. Both pages now explain the same numeric operation, while the historical addresses remain available to existing links.
TRUNC syntax
Microsoft’s documentation describes two arguments:
=TRUNC(number,[num_digits])
The first argument is the number to truncate. The optional second argument specifies precision. Omitting it uses 0, leaving the integer part. Square brackets mark an optional argument in the syntax description; you do not type them into an actual formula.
My English examples use a decimal point and commas between arguments. The Polish version uses localized function names, a decimal comma and semicolons. Separators depend on regional settings, so I check the argument tooltip in the installed Excel version when a pasted formula is rejected.
This operation does not require a new dynamic array function. Microsoft lists support including Excel 2016, 2019, 2021, 2024 and Microsoft 365. I use individual values throughout this entry, avoiding a separate discussion of array behavior across versions.
Truncation moves toward zero
All values below are synthetic examples. You can trace each result by crossing out fractional digits, without building a large workbook.
| Formula | Result | What remains |
|---|---|---|
=TRUNC(6.8) |
6 | The integer part of a positive value |
=TRUNC(-6.8) |
-6 | The integer part with its sign |
=TRUNC(0.8) |
0 | No whole unit |
=TRUNC(-0.8) |
0 | No whole unit on the negative side either |
I do not inspect whether the first discarded digit is greater than five. For 6.8, the result remains 6. For -6.8, the result is -6, which is greater than the input and closer to zero. That is why saying only “round down” leaves an important ambiguity.
If A2 contains 6.8, =TRUNC(A2) returns 6. A2 keeps its original value. I generally leave the source in one column and the transformed result in another. This makes it easier to identify where information is being discarded.
The second argument: decimals and tens
Positive precision keeps digits after the decimal point. For the synthetic value 26.87, truncating to one decimal place produces 26.8. Keeping two places leaves 26.87. TRUNC can therefore return a fractional result when the second argument requests it.
| Formula | Result |
|---|---|
=TRUNC(26.87,1) |
26.8 |
=TRUNC(-26.87,1) |
-26.8 |
=TRUNC(26.87,0) |
26 |
=TRUNC(26.87,-1) |
20 |
=TRUNC(-26.87,-1) |
-20 |
Negative precision works to the left of the decimal point: -1 truncates to tens. I discard the units and fractional part while retaining the sign. With -26.87, that produces -20, again showing the direction toward zero. Microsoft’s ROUNDDOWN precision guide provides a useful comparison for decimal positions and negative precision.

TRUNC versus INT
For positive 6.8, both functions return 6. That example alone hides the difference. I add a negative value to make the choice visible.
| Input | TRUNC with the second argument omitted | INT |
|---|---|---|
| 6.8 | 6 | 6 |
| -6.8 | -6 | -7 |
INT returns the nearest integer less than or equal to its input. For -6.8, that integer is -7. Microsoft explicitly explains this behavior in the INT documentation.
The task’s rule decides the function. Use TRUNC to discard the fractional part toward zero. Use INT when the calculation needs the lower integer boundary. Agreement on positive inputs does not make the two functions interchangeable for every dataset.
A calculated value and a displayed value
Reducing decimal places on the ribbon ordinarily changes the display. A value of 6.8 may appear as 7 while later calculations still use 6.8. Microsoft’s rounding guide distinguishes formatting from changing a value through a function.
TRUNC produces a result of 6. Multiplying that result by 2 gives 12. Multiplying the original 6.8 by 2 gives 13.6. This small comparison shows why fewer visible digits need not mean fewer digits in the calculation.
I also decide where truncation belongs in the process. Truncating synthetic values of 1.8 and 1.8 separately gives 1 and 1, totaling 2. Their original sum is 3.6; truncating only after summing gives 3. Those are different processing rules. A report should make its chosen rule clear.
What this function does not resolve
TRUNC is not a text cleaning function. For spaces, use the separate TRIM entry. The old Polish description of text processing does not belong in these numeric examples.
I also avoid using truncation as a universal fix for discrepancies. If a difference comes from a wrong unit, missing record or incorrect formula, discarding digits makes the cause harder to find. First establish why values differ, then choose the precision rule.
When filling down, inspect references too. =TRUNC(A2) should read A3 on the next row. For deliberate control over copied formulas, see my explanation of dollar signs in Excel references.
Exercise: predict each result first
Enter 6.8, -6.8, 26.87 and -26.87 in A2:A5. In column B, use TRUNC without its second argument. In C, set precision to 1; in D, set it to -1. Before filling the columns, write down which digits should remain.
Finally, compare TRUNC and INT on the first two rows. Explain the difference for the negative input and choose the rule that fits your task. Being able to justify that choice is more useful than picking whichever function name feels familiar.
Key takeaways
What is the Polish name for TRUNC?
It is LICZBA.CAŁK. The historical Polish URL in this pair contains the incorrect name ODETNIJ, but the revised page explains the numeric TRUNC function.
How does TRUNC differ from INT?
With zero precision, TRUNC removes the fractional part toward zero. INT returns the nearest integer less than or equal to the input. For -6.8, the results are -6 and -7 respectively.
Does reducing displayed decimal places truncate the value?
Ordinary cell formatting changes how a number appears rather than its stored value. TRUNC returns a new result for use in subsequent calculations.
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.
