Free Template · from the makers of NominalQC, QC software for small manufacturers
Free Cp/Cpk calculator (Excel template)
A process capability study without the statistics homework: paste your measurements, type the spec limits, and read the verdict. Within-process Cp/Cpk (the I-MR moving-range method), overall Pp/Ppk, expected and observed % out of spec, a histogram that turns out-of-spec bars red, and a run chart with the spec limits drawn on.
This log, without the spreadsheet
NominalQC computes X̄-R, I-MR, and Cp/Cpk live from the measurements your team already records — no exports, no pasting, no formula babysitting.
Try the live demoOpens on the real app with seeded data · no signup, no email · read-only · from $29/mo when you’re ready.
Free template — no email, no signup.
Prefer it in your inbox, with two short notes on using it?
What’s in the download
- Paste up to 500 measurements — n, mean, min/max, and both standard deviations compute live
- Cp and Cpk from within-process variation (average moving range ÷ 1.128 — the same estimate an I-MR control chart uses)
- Pp and Ppk from overall variation, so the Cpk-vs-Ppk gap exposes instability
- Observed % out of spec (counted) and expected % out of spec (from the normal model)
- Color-coded verdict: CAPABLE (≥1.33) / MARGINAL / NOT CAPABLE — thresholds explained on the sheet
- Histogram with out-of-spec bars in red, and a run chart with dashed LSL/USL lines that resize to your data
- 100% macro-free — works in Excel 2010+ on Windows and Mac, no security prompts
The zip contains the workbook (.xlsx) and a README quick-start. Sample data is included — paste your measurements over it.
Where this spreadsheet breaks
A calculator answers "were we capable last month?" It can’t answer "are we drifting right now?":
1.Capability is a rear-view mirror
By the time someone exports data and pastes it into the calculator, the process has been drifting for weeks. The number is honest — and late.
NominalQC computes SPC charts and capability live from the inspection measurements your team already records — no export, no paste, no lag.
2.The data entry is the weak link
Measurements live on paper checksheets or scattered files. The paste-into-Excel step is where transcription errors and cherry-picking sneak in.
In NominalQC measurements are captured once, on the inspection form, against the plan — and the same numbers feed the SPC charts untouched.
3.One characteristic at a time
The calculator holds one dimension. A real part has twenty — and nobody re-pastes twenty datasets a week.
NominalQC tracks every characteristic on every part continuously, with X̄-R and I-MR charts and capability per characteristic, selectable in two clicks.
Who built this
I spent 16 years in manufacturing quality — machining, fabrication, assembly — and in every shop I ended up building the same Excel tools with macros because the software budget went elsewhere. These workbooks are the ones I kept rebuilding. Eventually I stopped rebuilding spreadsheets and built NominalQC instead — but the spreadsheet is still the right starting point for a lot of shops, so here they are, cleaned up and free.
Questions
Is it really free?
Yes. You get the full workbook and the setup guide in exchange for an email address. Two short follow-up emails about running the process, then silence — unsubscribe anytime.
Does it work on Mac or older Excel?
Yes — this workbook is 100% macro-free, so there’s no "Enable Content" prompt and it works in Excel 2010 or later on Windows and Mac.
What’s the difference between Cpk and Ppk?
Cpk uses within-process variation (estimated from consecutive-point moving ranges) — what the process is capable of when stable. Ppk uses overall variation across all the data — how it actually performed, drift included. A big gap between them means the process wanders: the machine could do it, but something (setups, shifts, material lots) keeps moving the aim.
How many measurements do I need?
More is better: 30 is a common minimum for a rough read, 50–100+ for numbers you’d show a customer. Below ~25 the estimates get noisy — treat them as directional. And check the histogram first: if the shape isn’t roughly normal, the indices can mislead regardless of sample size.
More free templates
Ready to skip the spreadsheet stage?
Inspections, NCRs, CAPAs, LPAs, and gauge calibration — from $29/month flat, unlimited users.
Start your free trial