Net promoter score (NPS)

Net Promoter Score, or NPS, converts willingness to recommend into a compact customer KPI. It is easy to communicate, but customer satisfaction, loyalty and future revenue cannot be inferred from the score alone.

When to use it

• Answer: “To what extent are our customers willing to recommend us?”
• Assess the KPI within the Customer perspective.
• Plan representative collection, formula, frequency, privacy and follow-up.
• Compare like with like and retain the underlying response distribution.

Origins

NPS was developed by Fred Reichheld with Bain & Company and Satmetrix and promoted as a way to hold organisations accountable for customer relationships. Related names and marks have trademark or service-mark status; check current usage guidance when publishing them commercially.

The method became prominent through the claim that recommendation intent could act as a concise loyalty and growth signal. Independent evidence does not establish that it is universally superior to satisfaction, repurchase or behavioural measures.

What it is

Perspective: Customer perspective.

Key performance question: To what extent are our customers satisfied and loyal?

Ask how likely a customer is to recommend the specified company, product or service to a friend or colleague. The score groups respondents into Promoters, Passives and Detractors. Those groups simplify reporting but discard information within the original scale.

Do not state that each promoter will remain loyal or refer, or that every detractor will damage the brand. These are labels derived from a response, not observed behaviours.

How to use it

Measurement

Keep wording, entity, timing, channel, population and language consistent. Analyse recommendation alongside actual retention, repurchase, complaints, effort and satisfaction.

Data collection method

Use a survey by mail, online or phone and protect personal data. Two common designs are:

1. Relationship NPS: sample customers independently and ask about the overall relationship, optionally using a comparable competitor design.
2. Transactional NPS: ask after a defined touchpoint to diagnose that experience.

Avoid surveying so often that customers are burdened, and prevent employees from selecting only likely promoters.

Formula

On a 0 to 10 scale:

• Promoters score 9–10.
• Passives score 7–8.
• Detractors score 0–6.

Subtract the percentage of Detractors from the percentage of Promoters.

Frequency

Collect often enough to support the decision, not simply because the question is short. The inherited example suggests monthly sampling from 10% of customers; validate that cadence against volume, seasonality, contact burden and statistical power.

Source of the data

Use a documented sample of eligible customers, with consent and privacy controls appropriate to the jurisdiction and contact channel.

Cost/effort in collecting the data

The question is inexpensive, but representative sampling, multilingual design, nonresponse analysis, qualitative coding and closed-loop improvement require real effort.

Target setting/benchmarks

The 2011 US and UK industry report below is historical. Do not use it as a current benchmark without re-sourcing the methodology, date, population and competitive context.

Example

A company surveys 1,000 customers on a 0–10 scale, where 0 means not at all likely, 5 is neutral and 10 is extremely likely.

The score is:

Report the number of responses, promoter/passive/detractor proportions, uncertainty and full scale distribution.

Top practical tip

Ask what the customer particularly values and what one thing should improve. Use the answer to prioritise service changes, and test whether those changes affect behaviour as well as the score. Preserve the 9–10 promoter and 0–6 detractor thresholds when comparing waves.

Top pitfall

Grouping responses can hide distribution shifts and increase uncertainty. In the inherited Zappos illustration the survey was used in two contexts: 1) after an order on a 0–10 scale; 2) after a service interaction on a 0–10 scale. Preserve context rather than comparing unlike scores, and inspect the original 10-point response pattern instead of accepting claims such as a nine-fold error increase without statistical review.

Further reading

Net promoter is a customer loyalty metric associated with Fred Reichheld, Bain & Company, and Satmetrix; consult current trademark guidance.

Frederick F. Reichheld, The one number you need to grow, Harvard Business Review, December 2003.

Fred Reichheld, The Ultimate Question: Driving Good Profits and True Growth, Boston, MA: Harvard Business School Press, 2006.

www.netpromoter.com/

www.satmetrix.com

http://customermetrics411.com/other-metrics-net-promoter.html