Net Promoter Score (NPS): formula and margin of error
NPS is the share of promoters (9–10) minus the share of detractors (0–6), on a scale from −100 to +100. Its margin of error is wider than most dashboards suggest.
The formulas
NPS = (%promoters − %detractors) × 100 Var = p⁺ + p⁻ − (p⁺ − p⁻)² MoE (95%) = 1.96 × √(Var ÷ n) × 100 points
NPS is a difference of two proportions taken from the same sample. Its variance is therefore not the single-proportion p(1 − p). This is the Wald interval described by MeasuringU and Rocks (2016).
Worked example
400 responses: 210 promoters, 120 passives, 70 detractors.
- NPS 35.0 ± 7.4 points (27.6 to 42.4).
- Responses needed for ±5 points at this split: 888.
A month-on-month move from 35 to 39 on 400 responses is noise, not a trend.
Try it
Usage
import { npsStats, parseScores } from 'contact-center-math/nps'
npsStats(parseScores('10 9 8 6 10 7 9 3 10'))