Service Feedback Form Template
An NPS-anchored pulse for service businesses — the recommend score, the reason behind it, and different follow-ups for critics and champions.
One score and one honest sentence — that is all we need to know where we stand with you.
The recommend question earns its reputation not because the score itself is magic, but because it sorts your customers into three groups that deserve three different conversations. This template has that sorting built in: detractors (0–6) are asked what one fix would raise their score, promoters (9–10) are asked what they say when they recommend you, and passives get neither — just the open reason field everyone shares.
Why these fields. The 0–10 recommend scale is required and comes first, before any priming from other questions can nudge it. The always-on "what's behind the number" field is where the actual insight lives; an NPS score without its reason is a mood ring. The detractor branch is phrased as "the one thing we'd need to fix" because asking for a single fix produces an actionable answer, while "any other comments?" produces sighs. The promoter branch does double duty: "what do you tell people" surfaces your real-world pitch in customer language — which is frequently better than your marketing copy — and tells you which strength to double down on. The matters-most question calibrates everything else: a 6 from someone who values speed reads differently than a 6 from someone who values price. Use-again intent is the behavioral crosscheck, because people inflate recommend scores more readily than they misstate their own plans.
What we left out. Demographic questions (your CRM knows), star-rating batteries for every service dimension (the matters-most single-select achieves the same calibration at a fifth of the cost), and mandatory contact fields — candor first.
Who uses this. Cleaning companies, IT service providers, accounting firms, and trades businesses send it after each engagement; subscription services run it quarterly. Anyone whose growth depends on word of mouth should be running some version of this loop.
Make it yours. The two Logic thresholds are editable — some teams widen the detractor branch to 0–7. Add a webhook to push scores into your dashboard, or export the CSV and compute your score by segment: by month, by service line, by whatever your first question becomes. If you serve clients repeatedly, keep one form per year rather than per job, so the trend stays in one file.
Catching the polite detractor. The most dangerous respondent in a service business scores you an 8 and quietly picks "I'll shop around" — too courteous to criticize, already gone in spirit. The recommend number only becomes trustworthy next to use-again intent: whenever the score and the stated plan disagree, the reason field on that response deserves a slower read. Watch the matters-most mix over time as well. When the share choosing price creeps up against quality, your clientele is drifting toward bargain hunters, and that shift shows up here quarters before it reaches your margins.
Frequently asked questions
How is the NPS score calculated from responses?
Export the CSV and compute promoters (9–10) minus detractors (0–6) as percentages of total responses. Every response carries the raw 0–10 answer with a timestamp.
Why do some respondents see different questions?
Two Logic rules branch on the score: 6 and below reveals the fix-one-thing question, 9 and above reveals the what-do-you-tell-people question. Both thresholds are editable in the Logic panel.
When should a service business send this?
Within a day of completing the work, while the experience is concrete. For ongoing services, a quarterly rhythm keeps scores comparable without fatiguing clients.
Can I stop the same client from answering twice in one cycle?
Enable duplicate prevention in Settings — by device or IP — and repeat submissions within the cycle are declined politely.