Wiki

What is the Net Promoter Score in customer feedback programs?

Net Promoter Score asks how likely customers are to recommend a service and turns the answers into a single loyalty figure.

People answer on a scale from zero to ten, promoters are the nines and tens, detractors are zero through six, and the score subtracts the detractor share from the promoter share.

How it works

Passives, the sevens and eights, are left out of both shares, and the result can range from minus one hundred to plus one hundred.
The figure is a loyalty metric for product and service teams, not a detector for intrusions and not a proof that a system is safe.

Scripted responses, purchased reviews, and duplicate accounts can push the number up without any change in real customer experience.
Security and trust teams should treat a sudden perfect score with the same skepticism they would give a sudden perfect review page.

Concrete examples

  • A retailer sees the score jump in a week when hundreds of tens arrive from accounts created the same hour.
  • A support team celebrates an improved score, then learns a vendor suppressed unhappy replies before the calculation.
  • A bank keeps the score next to complaint data and notices the score rose while fraud complaints also rose.

Protective measures

  • Collect the score from verified customers and resist incentives that pay people to tick a ten.
  • Compare the score with complaint volume and fraud reports instead of managing the number alone.
  • Investigate clusters of identical answers, new accounts, and impossible completion times.
  • Tell executives the score does not measure whether accounts, payments, or data are being abused.

Related articles

Deception through internet services, which a loyalty score will not reveal, is explained in What is online fraud carried out through internet services?.

Designing the categories behind a metric or a model is described in What is label engineering for security data and detection models?.

Learning from public information, including public reviews, is covered in What is open source intelligence gathered from public information?.

‹ Back to the wiki