Surveys & feedback

Run a recommendation score survey

The recommendation question, how to calculate the score from promoters and detractors, a follow-up question, and how to set it up.

Last updated

The recommendation score measures loyalty with one question. It’s popular because it’s short, the same everywhere and easy to follow over time.

The recommendation question

How likely are you to recommend us to a friend or colleague?

0 (Not at all likely) to 10 (Extremely likely)

Keep the wording and the 0 to 10 scale as they are. Changing them makes your score impossible to compare with earlier ones.

Calculate the score

Sort the answers into three groups:

AnswerGroupWhat it means
9 or 10PromotersLoyal customers who recommend you
7 or 8PassivesSatisfied, but could easily switch
0 to 6DetractorsUnhappy customers who may warn others off

Recommendation score = % promoters − % detractors

With 50% promoters, 30% passives and 20% detractors, your recommendation score is 30. The score runs from −100 to +100 and is written without a percent sign. Above 0 means more promoters than detractors. What counts as good differs by industry, so the trend in your own score tells you more than a comparison.

Always ask why

The number alone doesn’t tell you what to fix. Add one open question:

  • For everyone: “What’s the main reason for your score?”
  • For detractors: “What would we have to change to win you back?”
  • For promoters: “What do you like most about us?”

Build it in StrawPoll

  1. Start from the Recommendation score template. It has the 0 to 10 question and an open follow-up.
  2. Or add an Opinion scale question, set Minimum to 0 and Maximum to 10, and write “Not at all likely” and “Extremely likely” into Low label and High label.
  3. To ask each group a different follow-up, add three Long text questions and show each one for the matching scores with conditional logic.

The results show the spread across all eleven answers along with the score.

When and how often

  • Relationship score: every three to six months, to all customers or a random part of them.
  • After a milestone: a few weeks after signing up or after a renewal.
  • Not after every contact. For single interactions, a customer satisfaction survey fits better.

To see how the score moves, let people answer again after a set time, or export the responses with their dates.

Recommendation score for employees

The employee version asks: “How likely are you to recommend us as a place to work?” The Employee engagement template includes it. Keep it anonymous, as described in Run an employee engagement survey.

Related articles