NPS, CSAT, and CES
NPS, CSAT, and CES are three standard satisfaction metrics: NPS measures how likely someone is to recommend you, CSAT measures how satisfied they were with one interaction, and CES measures how much effort that interaction cost them.
The three questions, plainly. NPS (Net Promoter Score) asks how likely you are to recommend this to a colleague on a 0 to 10 scale and reports the share of promoters minus the share of detractors. CSAT (Customer Satisfaction) asks how satisfied you were on a short scale and reports the percentage of positive answers. CES (Customer Effort Score) asks how easy it was to get what you needed and reports the average.
For a single documentation page, CES is usually the honest fit and NPS is usually the wrong instrument. Nobody recommends a company because of one reference page, so NPS collected there measures brand mood rather than the page. A "was this page helpful" rating is CSAT-shaped, and "how easy was it to find what you needed" is CES-shaped and closest to something a documentation team can actually fix.
All three produce a number, and the number on its own is not actionable. The follow-up comment is where the fix lives. A widget that asks for a rating and then one open sentence returns the metric and the reason in the same submission, which is why per-page ratings are worth very little without a text field next to them.
Frequently asked questions
Which metric should a documentation site use?
A per-page helpful or not-helpful rating, which is CSAT-shaped, with an optional comment field. Keep NPS for product-wide surveys where the recommend question makes sense.
What is a good NPS or CSAT score for documentation?
Cross-company benchmarks mislead, because scales, wording, placement, and audiences all differ. Track your own trend per page and compare a page to itself before and after a rewrite.
Can you measure all three at once?
You can, but response rate falls with every extra question. One rating plus one open field outperforms a three-metric survey on almost every site we see.