PERCENTRANK.INC in Collabora Office: Inclusive Positions and Stable Benchmark Comparisons
Calculate Collabora inclusive percentage ranks with explicit significance, observed-value controls and a benchmark expansion example.
Verification status: See the evidence and test scope described in this guide. How formula examples are checked
Quick answer: PERCENTRANK.INC returns a relative position on a scale including 0 and 1. The smallest and largest observed values occupy those endpoints in the independent distinct-value model used here.
Availability and version history
- Introduced: original Collabora release and launch date are not established by the reviewed sources.
- Availability: currently documented in Collabora Office 26.04 desktop help.
- Platform scope: online and mobile deployments are not inferred from this desktop-help release.
- Checked: October 14, 2026. Function-specific documentation and independent calculations; no native Collabora execution.
Syntax and an important default limit
=PERCENTRANK.INC(data;value;[significance])
The source makes significance optional and describes it as significant digits used for the rounded percentage result. It does not state its omitted numerical default in the dedicated section reviewed. The worked formulas therefore explicitly use 6.
The independent numeric position model is (position-1)/(n-1) for distinct ordered observations. The examples choose exact terminating ranks, so no result depends on an unverified rounding-versus-truncation boundary.
The source documents an error when the requested value lies outside the total data range, but it does not name the exact display. Missing interior values, mixed types and tie policies are not borrowed from another application.
Example 1: compute all five observed positions
Enter 3,12,24,48,96 in A2:A6. Put the same five targets in C2:C6, enter this in D2 and fill down:
=PERCENTRANK.INC($A$2:$A$6;C2;6)
Expected results are 0,0.25,0.5,0.75,1. The position of 48 is fourth; (4-1)/(5-1)=0.75.
The absolute data range prevents accidental changes to the benchmark while the target cell moves. Ranking does not sort or move either the source data or the targets.
Example 2: record order is not rank order
Use the same benchmark values in another range, ordered 96,3,48,12,24. Their modeled ranks in that row order are 1,0,0.75,0.25,0.5.
The target 48 retains rank 0.75. Reordering complete records is a presentation change for this model; adding, removing or altering benchmark observations is a data change.
Keep identifiers with values when sorting a real table so that a correctly computed score remains attached to the right record.
Advanced example: add a new largest benchmark
Append 192 to the original sorted benchmark and expand the range to A2:A7:
=PERCENTRANK.INC(A2:A7;48;6)
Independent expected result: 0.6. The target remains fourth, but the denominator is now five. The old maximum 96 moves from rank 1 to 0.8, while the new 192 takes rank 1.
This illustrates why scores from different benchmark periods are not automatically comparable. Record the observation count and range as well as the target value.
Advanced example: rank cutoffs versus magnitude cutoffs
Return to the original five observations. A rule accepting modeled ranks at least 0.7 selects 48 and 96. A rule accepting values at least 70% of the maximum instead uses threshold 67.2 and selects 96 alone.
Both rules contain a percentage, but they answer different questions. A percentage rank measures position among observations, while a percentage of a maximum measures magnitude relative to one value.
Neither rule here promises a fixed fraction of rows. Only two of five values meet the rank cutoff, or 40% of this finite dataset.
Advanced check: positive affine conversion
Apply 2×value+1 to the original benchmark, producing 7,25,49,97,193. The transformed target 97 still has modeled rank 0.75.
Its location on the measurement scale changed; its ordering among the five observations did not. Negative scaling reverses order and needs a different directional interpretation.
Errors, aliases and format scope
Against the original benchmark, 200 is outside the range 3 through 96 and reaches the documented error condition. The exact native code and interior interpolation behavior remain unverified.
The dedicated section does not settle range-text/Boolean handling, duplicate ranks, singletons, empty inputs or nonpositive significance. PERCENTRANK supplies some different explicit details, but those are not automatically shared.
The namespace COM.MICROSOFT.PERCENTRANK.INC is documented outside the cited OpenFormula 1.3 standard. PERCENTRANK.EXC uses an exclusive scale; neither that distinction nor a format note proves a retirement or universal import/export behavior.
References and verification
Complete function-specific documentation reviewed. Worked outputs are independent mathematical models or explicitly documented rules, not application screenshots or native test results. Precision, type and error boundaries retain their stated limits.