A CSAT survey template is a short, reusable question set: one satisfaction rating plus an optional comment. Intercom teams should start with native conversation ratings rather than a second form. Use a custom survey only when you need a score on a moment Intercom does not rate (onboarding call, implementation milestone, or a non-inbox channel).
Formula and Intercom emoji classes: what CSAT is. Building the weekly view: Intercom CSAT report. Bias and coverage: CSAT survey response bias.
A CSAT template only pays off once the scores reach someone who can act. Supportman posts every Intercom rating to Slack the moment a customer submits it, with the remark and conversation link attached.
Use Intercom ratings first
Intercom already asks customers to rate the conversation with five emoji choices. Amazing and Great are positive; Okay and Bad are neutral; Terrible is negative. CSAT is positive ÷ all ratings. You do not need a SurveyMonkey form on top of that inbox unless you are measuring a different event.
Stacking a custom CSAT email on every closed Intercom conversation trains people to ignore both surveys. One rating per event.
When to send a custom CSAT survey
| Moment | Use native Intercom rating? | Use a custom template? |
|---|---|---|
| Closed inbox conversation | Yes | No |
| Fin-only resolution | Yes, then segment Fin vs human | Only if you need extra questions Intercom cannot ask |
| Onboarding or implementation call | No | Yes — different event |
| Quarterly relationship check | No | Use NPS or a relationship survey, not CSAT |
| Help-center visit | No | Article feedback, not a CSAT email |
Copy-paste templates
Inbox (if you must replicate Intercom outside Intercom)
- How satisfied were you with this support conversation? (Amazing / Great / Okay / Bad / Terrible, or 1–5 with 4–5 = positive)
- What is the main reason for your rating? (optional, one box)
Onboarding or implementation
- How satisfied were you with onboarding so far? (1–5)
- What still blocked you from completing [named job]?
After a recovered miss
- Did we resolve the issue we missed the first time? (yes / no)
- How satisfied are you with how we handled the follow-up? (1–5)
Do not add NPS, CES, and CSAT to the same three-question email. Each extra question drops completion and mixes loyalty with a single interaction.
Keep the scale labeled. “1–5” without “5 = very satisfied” produces noise. If you operate in Intercom, prefer the same five labels the inbox already uses so CSAT from two tools can be compared.
Keep the response rate honest
- Send immediately after the event, not in a weekly digest.
- One reminder, not three.
- Report CSAT with response count and response rate. 90% from 8 replies is not a program.
- Do not survey customers who just left a Terrible on the same conversation.
- Route negatives the same day. Template without a queue is a spreadsheet. Send Intercom ratings to Slack.
Question design that leads (“How satisfied were you with our excellent support?”) is response bias. Details: CSAT survey response bias.
Frequently asked questions
What should a CSAT survey include?
One rating question and one optional comment. That is enough for ops. Add a second rating only when you are measuring a different event, such as onboarding vs the ticket.
How many questions should a CSAT survey have?
One or two. Past three, you are running a research study and will watch completion fall.
Should I use a 1–5 or 1–10 CSAT scale?
Use 1–5, or Intercom's five emoji labels, so “positive” is unambiguous (4–5, or Amazing/Great). A 1–10 scale needs an extra rule for what counts as positive and does not match Intercom Reports.
Do I need a CSAT survey if I use Intercom?
Not for inbox conversations. Use conversation ratings, then a custom template only for events Intercom does not score.
When should I send a CSAT survey?
Immediately after the interaction you want to measure. Delayed surveys mix that conversation with everything that happened since.