Knowledge base / Excel
ROUNDUP in Excel: digits, negative values and package counts
Distinguish decimal places from multiples. Follow negative values, whole package counts and the difference between rounding rows and rounding a total.

ROUNDUP rounds away from zero to a specified number of digits. A positive value may increase, while a negative value may become more negative. I state that direction explicitly because the everyday meaning of up can suggest moving to a larger number on a number line.
Its second argument specifies digits, not a multiple. An argument of 2 means two decimal places; it does not mean the next multiple of two. All quantities below are synthetic and are small enough to follow directly.
Syntax and the meaning of digits
Microsoft’s documentation gives the syntax:
=ROUNDUP(number,num_digits)
Positive num_digits selects places after the decimal point, zero selects a whole number, and negative values move rounding to the left of the decimal point. A value of -1 means tens, while -2 means hundreds. Polish Excel calls the function ZAOKR.GÓRA.
These examples use English function names, commas between arguments and decimal points. Check regional settings if your installation expects different separators. Do not append explanatory comments to a formula being pasted into a cell. Each expected result is shown separately below.
Examples you can follow by hand
| Formula | Result | Meaning of the second argument |
|---|---|---|
=ROUNDUP(3.14159,2) |
3.15 | Two decimal places |
=ROUNDUP(5.3,2) |
5.3 | Already within the requested precision |
=ROUNDUP(5.3,0) |
6 | A whole number away from zero |
=ROUNDUP(4567,-2) |
4600 | Hundreds away from zero |
=ROUNDUP(-2.345,2) |
-2.35 | Two places, further from zero |
For 3.14159 at two places, consider 3.14 and 3.15. The away-from-zero direction selects 3.15. For -2.345 it selects -2.35, although that number is smaller than the input. This comparison prevents the word up from being mistaken for a guarantee that the output is numerically larger.
To display 5.30, use an appropriate number format. The numeric value 5.3 and its appearance with two decimal places are separate concerns. An additional calculation is unnecessary merely to show a trailing zero in the finished report.
Count enough packages for all items
Suppose there are 17 items and each package can hold 8. Enter 17 in A2, 8 in B2 and the formula in C2:
=ROUNDUP(A2/B2,0)
The result is 3 packages. Two packages hold 16 items, so the remaining item needs a third. This gives the package count; it does not promise equally filled packages. For 16 items the answer is 2, and for zero items it is 0.
The example assumes a nonnegative item count and a positive capacity. A capacity of zero is an input problem. Before suppressing an error, establish whether it means missing information, an unavailable package or a typing mistake. Rounding cannot determine that business meaning on behalf of the person preparing the table.

Decimal places and multiples are different requirements
CEILING handles multiples. For positive 14 with a step of 5, it returns 15. By contrast, =ROUNDUP(14,5) leaves 14 unchanged: the second argument requests five decimal places rather than a five-unit increment.
Similarly, a half-unit increment differs from one decimal place. Before writing a formula, I describe the requirement in words: two decimal places, enough whole packages or multiples of five. That description helps select the function and makes it easier for another person to verify what the second argument means.
ROUND selects the nearest value according to its rounding rules, while ROUNDDOWN moves towards zero. For -2.345 at two places, ROUNDDOWN gives -2.34. Direction relative to zero is a more reliable guide than the everyday word down.
Decide when rounding happens
Take two synthetic components of 1.21 each. Rounding each away from zero to one decimal place gives 1.3 and 1.3, totalling 2.6. Their original sum is 2.42, which becomes 2.5 if you round only the total. The two sequences implement different requirements.
Label whether a report rounds individual rows, groups or its final result. Do not choose the stage merely because the formula is convenient. Likewise, do not carry the package example into a billing process without agreeing its rules. Knowing a spreadsheet function does not establish the business policy governing its use.
An exercise and a boundary check
Consider 4.01, 4 and -4.01 at zero decimal places. Expected results are 5, 4 and -5. Return to the packaging example and compare 15, 16 and 17 items. Results of 2, 2 and 3 show the boundary at which an extra package becomes necessary.
For rounding to multiples in the other direction, see FLOOR. Beside your own formula, leave the unit and a short reason for the direction chosen. A later change in precision should be a deliberate change to the report’s rule.
Key takeaways
Does the second ROUNDUP argument specify a multiple?
It specifies digits. A value of 2 means two decimal places, 0 means whole numbers and -2 means hundreds.
How does ROUNDUP handle negative values?
It rounds away from zero. For example, -2.345 at two places becomes -2.35, which is numerically smaller.
What does rounding the item count divided by capacity produce?
For nonnegative items and positive capacity, it gives enough whole packages to hold all items. It does not make every package equally full.
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.
