CEILING.PRECISE in Google Sheets: Choose the Mathematical Upper Grid Point

• Introduced: first launch date or introduction version not established by the official history reviewed.
• Availability: documented in the rolling Google Sheets service; this guide uses ordinary cells in the web editor, without a claim about identical mobile authoring workflows.
• Checked: October 14, 2026 against the official CEILING.PRECISE reference. Examples were independently calculated; no native Sheets execution was performed.

=CEILING.PRECISE(number, [significance])

Significance is optional and defaults to 1. Google documents use of its magnitude regardless of sign. There is no negative-direction mode argument. The examples here use nonzero steps and ordinary numeric values. Official CEILING.PRECISE reference

For a positive step, the mathematical result is the smallest grid multiple at or above the input. This explains why positive values move away from zero but negative values move toward zero.

PRECISE does not mean unlimited arithmetic precision. The name identifies this direction convention; it does not eliminate finite representation or decide how many decimals to show.

=CEILING.PRECISE(17.2)
=CEILING.PRECISE(-17.2)
=CEILING.PRECISE(17)

Expected results: 18, −17, 17. Each answer is the first integer no smaller than the input.

A whole-number cell display is not enough to predict this result if the cell stores a hidden fraction. Keep the source visible with sufficient displayed precision while checking a rounding rule.

=CEILING.PRECISE(17.2,5)
=CEILING.PRECISE(17.2,-5)
=CEILING.PRECISE(-17.2,5)
=CEILING.PRECISE(-17.2,-5)

Expected results: 20, 20, −15, −15. The sign of the second argument does not reverse the result. The negative input lies between −20 and −15, and −15 is the upper neighbor.

If a negative quantity must instead move away from zero, use a separately chosen function or documented mode. Do not attempt to produce that policy by merely negating this function’s significance.

=CEILING.PRECISE(17,5)
=CEILING.PRECISE(20,5)

Expected results: 20 and 20. The first input is an integer but not a multiple of five. The second is already on the permitted grid.

This distinction matters for package sizes and timetable increments. A check that a value has no fraction is not a check that it is divisible by the operational step.

For fictional demands 15, 21, 22 in A2:A4, use =CEILING.PRECISE(A2,7) in B2 filled down. Expected capacities: 21, 21, 28.

The block counts =B2/7 are 3, 3, 4. Spare capacity =B2-A2 is 6, 0, 6. Demand totals 58, while separately allocated capacity totals 70.

If the demands may share capacity, =CEILING.PRECISE(SUM(A2:A4),7) gives 63, or nine blocks. Separate rounding requires ten blocks. The difference is caused by the grouping rule, not a calculation error.

Use the method that matches the actual allocation constraint. Rounding every row and rounding one aggregate are not interchangeable formulas.

For grid points starting at 1 and repeating every 4, use:

=1+CEILING.PRECISE(F2-1,4)

For F2 values 2, 5, 6, −2, expected results are 5, 5, 9, 1. The allowed sequence is …, −3, 1, 5, 9, … . Subtracting the origin makes ordinary multiples applicable; adding it back restores the required grid.

For −2, the transformed value −3 rounds to zero before adding one. This is a useful negative-offset control that a positive-only test would miss.

The checked Google page prints a positive result in one example with a negative input, conflicting with its own toward-zero explanation. This guide does not copy that sign inconsistency. Its stated rule independently gives =CEILING.PRECISE(-9.1,0.25) as −9, and the corresponding positive input gives 9.25.

These are expected calculations, not native observations. Use a tolerance when testing approximate measurements close to boundaries. An output formatted to two decimals does not prove the underlying input was an exact decimal value.

ROUND chooses nearest decimal-place rounding, and ROUNDUP moves away from zero by decimal places. Neither is automatically equivalent to a mathematical upper multiple for negative inputs.

The dedicated Google page does not establish a full zero-step or invalid-input error matrix. Keep steps nonzero in this documented workflow, validate numeric parameters and test any additional edge cases in the target file before making them part of an operational rule.

Verification: Expected results use fictional data and independent calculations rather than native Google Sheets observations. See the editorial testing standards.