# NPS survey question and calculation Ask a recommendation question, calculate NPS from a labelled example, and investigate the reasons behind the score. If nothing was unclear or difficult, say so. Skip questions about steps you did not experience; use not applicable where appropriate. ## 1. How likely are you to recommend us to a friend or colleague? Answer from 0 to 10. Answer: ____________________ ## 2. What is the main reason for your score? Answer: ____________________ ## 3. Which part of your recent experience influenced your answer? Answer: ____________________ ## 4. What would make recommending us easier? Answer: ____________________ ## 5. What should we understand about the context of your response? Answer: ____________________ ## Author notes / illustrative keys where applicable 1. The recommendation rating supplies the defined zero-to-ten input for NPS grouping. Keep the scale consistent; the resulting score describes respondents and cannot explain the reason for sentiment on its own. 2. The reason adds explanatory context to a recommendation score. Preserve the respondent's stated driver without assuming that the same rating has the same meaning across customers. 3. A recent touchpoint links the rating to an experience the team can investigate. Distinguish that event from longer-term perceptions before attributing a change in NPS to one interaction. 4. Conditions for recommending reveal obstacles that may be practical, relational or situational. Explore a concrete change without treating a stated intention as an actual referral or guaranteed higher score. 5. Response context helps interpret differences in tenure, recent use or situations represented. Avoid collecting unnecessary identifiers and do not assume contextual explanation makes a voluntary sample representative. Interpretation caution: NPS is a score from minus 100 to plus 100, not a percentage or a diagnosis of why sentiment changed. ## The same score can have different reasons Synthetic responses show why a score alone does not choose the next service action. ### Score 6: “I waited for the second support reply.” What it supports: Investigate the handoff between replies and the stated update expectation. What it cannot establish: This single explanation does not establish how often the delay occurs. ### Score 6: “The delivery date changed after checkout.” What it supports: Inspect which delivery information the customer saw before committing. What it cannot establish: The score does not establish a support-response problem. ### Score 6 with no explanation What it supports: Keep the score and optional missing reason explicit. What it cannot establish: Do not infer a cause or fill in a reason from other respondents. ## Illustrative finding and follow-up Observation: Synthetic sample: 55 promoters, 25 passives, and 20 detractors among 100 responses. NPS = 55% minus 20% = +35. Action to test: Read the reasons behind each response group before deciding which customer experience to improve. Follow-up: Repeat the same invitation rule and scale after people can encounter the chosen service change. Report response count and relevant sample changes alongside NPS; inspect reasons before attributing any score movement. ## Collection protocol When and whom to ask: Choose and state one consistent relationship-review interval or named post-experience touchpoint; invite customers who fit that definition and do not mix those two sampling plans in one comparison. Decision owner: The customer experience owner documents the invitation rule and asks the relevant service owner to investigate the most specific reasons behind responses. Follow-up: Repeat the same invitation rule and scale after people can encounter the chosen service change. Report response count and relevant sample changes alongside NPS; inspect reasons before attributing any score movement. Response handling: Keep the score and optional explanation separate. Avoid unnecessary identifiers, explain who reads comments and do not label a voluntary sample representative.