Zoho Sheet WRAPROWS: Row Width, Padding and Mean Examples

• Introduced: Included in the fifteen-function launch linked from Zoho’s March 2023 release history. No numbered cloud release is given.
• Availability: Documented in the current Zoho Sheet web-service help; no final supported-version endpoint is established.
• Scope: Mobile, offline and imported-workbook parity remain separately unverified.
• Checked: October 14, 2026, against the official WRAPROWS entry. Layouts and numerical examples were independently calculated; no native Zoho Sheet execution was performed.

WRAPROWS(values; wrap_count; [fill_with])

The official entry requires the source range and wrap count. The optional fill value defaults to an unavailable-value error when an incomplete final row needs padding. Its examples use both an omitted fill and the text "-".

WRAPCOLS instead fills downward by columns. The two have distinct output contracts; they are not alternate spellings. TOROW flattens data into one row rather than creating several rows.

Create Readings, reserving A1:T12. Put Reading in A1 and enter 11,13,17,19,23,29,31,37 down A2:A9. Leave E2:G4, I2:L3, N2:P4 and R2:R10 empty.

Enter =WRAPROWS(A2:A9;3;"-") in E2. The expected E2:G4 rectangle is:

There are eight readings and nine positions, so only G4 is padding. It is text, not a ninth observation.

Zoho’s dynamic-array instructions say to enter the formula in one anchor cell and press Enter. Results occupy adjacent cells automatically. Leave each complete output rectangle empty and do not fill duplicate formulas across it. An obstruction produces the documented #SPILL! error.

In I2 enter =WRAPROWS(A2:A9;4). The two output rows in I2:L3 are 11,13,17,19 and 23,29,31,37. There is no incomplete row and no padding. In N2 enter =WRAPROWS(A2:A9;3) to repeat the three-column layout with the default error in its ninth position P4.

On Readings, put =SUM(A2:A9) in T2 and =SUM(I2:L3) in T3. Both expected totals are 180. T4 =COUNT(E2:G4) is 8, and T5 =AVERAGE(E2:G4) is 22.5. The documented COUNT and AVERAGE rules exclude the nonnumeric text padding in these explicit examples. Avoid aggregating the separate N:P layout as if its error were a valid reading.

Put =TOCOL(E2:G4;0;FALSE) in R2. The expected R2:R10 sequence is 11,13,17,19,23,29,31,37,-. TOCOL with FALSE scans rows, matching the way WRAPROWS placed the input. The final text padding is retained because ignore=0.

This is not an exact eight-cell round trip: the rectangular result has gained one padding cell. Preserve the original count instead of treating every result position as original data.

Copy Readings as ChangedReading and change only A6 from 23 to 230. The corresponding positions F3, I3, O3 and R6 become 230. T2/T3 become 387, T4 remains 8 and T5 becomes 48.375. The layout dimensions and final padding do not change.

Create BatchMeans, reserving A1:M10. Put 10,20,30,40,50 in A2:A6. In D2 enter =WRAPROWS(A2:A6;3;"-") and in H2 enter =WRAPROWS(A2:A6;3;0).

D2:F3 is 10,20,30 / 40,50,-. H2:J3 is 10,20,30 / 40,50,0. The last zero is an explicit numeric fill, not an original measurement.

Enter the following formulas in L2:L9:

L4 gives the two row means equal weight even though they summarize three and two readings. The overall mean uses five observations: (3×20+2×45)/5=30. L7 instead includes the synthetic zero, increasing its denominator to six. A tidy rectangular layout does not make the statistical interpretation automatic.

Copy BatchMeans as GenuineZero and change only A6 from 50 to 0. The text-padded row becomes 40,0,- and the zero-padded row becomes 40,0,0. L2:L9 now gives 20,20,20,20,20,16.6666666667,5,6, with the repeating value rounded for display. The real zero remains part of the five-value dataset; it must not be removed merely because a different position uses zero as padding.

Use a single row or column and positive whole wrap counts for these exercises. The checked entry does not settle zero or fractional counts, a count larger than the source length, two-dimensional input, real blanks, formula-empty strings or imported errors. Do not infer those outcomes from another spreadsheet program.

The fill argument supplies positions beyond the end of the source sequence. It is not a documented instruction to replace every empty-looking source cell. The default is described as #N/A, while the shared error table uses #N/A!; the exact localized display was not tested.

Check source order, anchor position, wrap count and the full empty output area before changing data. A layout can be mathematically correct while grouping the wrong records if the source sequence is incomplete. Keep the original vector alongside the reshaped result.