NPS follow-up questions ask why someone answered “How likely is it that you would recommend [company] to a friend or colleague?” the way they did, on a scale from 0 to 10. Detractors (0–6), passives (7–8) and promoters (9–10) each need a different follow-up, and when the answer is “it's fine,” Spoke asks one more.
Detractors are reacting to the price rise; passives are held back by support handoffs.
An example of the line the Agent Results Brief leads with.The standard Net Promoter Score question, from Fred Reichheld of Bain & Company, is “How likely is it that you would recommend [company] to a friend or colleague?” People answer from 0 (not at all likely) to 10 (extremely likely), and the score puts them in one of three groups. Every NPS follow-up question starts from which group they're in.
Unhappy customers. Ask what's behind the score and what would change it.
Satisfied, not enthusiastic. They count in the total but add nothing to the score. Ask what kept it from a 9.
Enthusiastic fans who recommend you. Ask why, so you know what to protect.
NPS = % promoters − % detractors (a number from −100 to 100)
Under every question is the follow-up Spoke asks when the first answer is vague. That second question is where the real answer usually is.
What's the main reason for your score?
When did that last get in your way?
Which part of working with us affected your score most?
What happened the last time you ran into it?
What were you hoping we'd do for you when you signed up?
How close have we come to that so far?
What would have to be different for you to score us higher?
Of those changes, which one matters most to you?
What, if anything, do you do outside our product to get the job done?
How much time does that take you in a normal week?
What would you tell a friend who asked you about us?
What's the one thing you'd make sure they knew?
What almost made you stop using us?
What kept you from leaving?
What's the one thing that kept this from a 9 or 10?
When did you last notice it?
What do you value most about working with us?
What would you miss first if it went away?
What's working well enough, but not great?
What would great look like there?
What's missing for you right now?
What do you do today to get around it?
What would make us an easy recommendation?
What's the closest we've come to that so far?
What would make you look at other options?
When has that come closest to happening?
What's the main reason you'd recommend us?
What happened the last time you noticed it?
Who would you recommend us to?
What problem would they be trying to solve?
How would you describe us to that person?
What's the first thing you'd tell them?
What was it like before you started with us?
What changed first once you did?
What would you add, if you could add one thing?
Which part of your work would that help with?
What's one thing we should never change?
What would it mean for you if it did change?
Every person gets interviewed, not surveyed. Spoke asks, listens, and follows up on what that person actually said. Here's an example.
Read the answer, not the transcripts: themes ranked by how many people raised them, the quotes behind each, and the places where two groups disagree.
The reasons split by score. Detractors point to the price rise and to slow support, like a new location that took four days of emails to set up. Passives are held back by support handoffs: they explain the problem twice. Promoters praise support when it reaches them fast, so promoters and detractors disagree on it. What to do: show what the new price buys, fix the ticket handoff, then make setup self-serve.
An example Agent Results Brief. The counts, themes and quotes are illustrative, not from a real survey.
A fixed follow-up gets one answer and stops. When the reply is “it's fine” or “great service,” Spoke asks one more question about what that person said. One example per score group:
The 5 isn't about the product as a whole. It's one setup step that needs a support email and makes the customer look slow to their boss.
A 7 that looks satisfied hides a handoff problem: the ticket history isn't reaching whoever picks it up.
What this promoter values is specific: a problem fixed before they had to ask. That's the thing to protect as you grow.
One open follow-up after the score is enough if it fits the group. Here is the first follow-up to ask, what you're trying to learn, and the probe for a vague reply.
| Score | What you want to learn | Ask first | If the answer is vague |
|---|---|---|---|
| Detractors (0–6) | The specific thing behind the low score, and whether it can be fixed | “What's the main reason for your score?” | “When did that last get in your way?” |
| Passives (7–8) | The one gap between okay and a recommendation | “What's the one thing that kept this from a 9 or 10?” | “When did you last notice it?” |
| Promoters (9–10) | What they value, in their words, so you can protect it | “What's the main reason you'd recommend us?” | “What happened the last time you noticed it?” |
The question wording is Fred Reichheld's original, from “The One Number You Need to Grow” (Harvard Business Review, December 2003); Bain & Company's own page uses the shorter “How likely are you to recommend us to a friend or colleague?” The three groups and the formula (% promoters − % detractors, from −100 to 100) follow Bain's Measuring your Net Promoter Score. Timing, by common practice: ask transactional NPS soon after one interaction, relational NPS every quarter or twice a year. The same question has variants, such as product NPS (“recommend [product]”) and eNPS for employees (“recommend [company] as a place to work”). More on writing probes in follow-up questions.
The number is the easy part. The reason is what gets lost. More on neutral wording in open-ended survey questions.
A detractor and a promoter are telling you different things. Ask each what fits their score.
A follow-up that assumes the answer gets agreement, not the reason.
“Fine” and “great service” are where the reason starts. Ask about the last time it happened.
Every extra question costs answers to the ones that matter. Keep to the score and one or two open follow-ups.
What do you want to learn, and who's answering?
Drafted the spec. Should I ask promoters who they'd recommend us to, or keep it to why they scored high?
One open question that fits the score. Ask detractors “What's the main reason for your score?”, passives “What's the one thing that kept this from a 9 or 10?”, and promoters “What's the main reason you'd recommend us?” If the reply is vague, ask about the last time it happened.
Anything above 0 means more promoters than detractors. Compare against your own last score and your industry, not a universal number.
Detractors score 0 to 6, passives 7 or 8, and promoters 9 or 10. Passives count in the total number of answers but neither add to nor subtract from the score.
Subtract the share of detractors from the share of promoters: NPS = % promoters − % detractors. The result runs from −100 (all detractors) to 100 (all promoters).
Reply to each one within a few days, ask about the specific thing they named, and tell them what you changed.
Relational NPS asks about your company overall, on a regular schedule such as every quarter. Transactional NPS asks right after one interaction, such as a purchase or a support ticket. The follow-ups on this page work for both; for transactional NPS, ask about that one interaction.
Free to start. No card. Your first ten agent surveys are on us.
