Customer Satisfaction (CSAT)

Know how your customers feel about your support. Scitor automatically sends a simple satisfaction survey after tickets are resolved β€” customers rate their experience, and you get measurable data to track support quality over time.

Pro feature β€” CSAT surveys require a Pro plan.

How it works

  1. Ticket is closed β€” when an issue is closed, Scitor checks the CSAT configuration
  2. Delay period β€” a configurable delay (default: 1 hour) gives the team time to reopen if needed
  3. Survey sent β€” an email with rating buttons is sent to the original customer
  4. Customer rates β€” clicking a rating records the score and posts a comment on the issue
  5. Report included β€” CSAT metrics appear in /generate-report output

Configuration

Add a csat section to your .github/scitor.yaml:

csat:
  enabled: true
  delay: 1h              # Wait before sending survey (default: 1h)
  exclude_labels:        # Don't survey these ticket types
    - "spam"
    - "duplicate"
    - "internal"

Configuration options

Option Type Default Description
enabled boolean false Enable CSAT surveys
type string csat Survey variant: csat, nps, thumbs, or ces
delay duration 1h Delay before sending survey after ticket close
scale number 5 Rating scale: 5 for 1-5 stars, 2 for thumbs up/down (legacy; prefer type)
include_comment boolean true Allow customers to leave an optional free-text comment
exclude_labels string[] ["spam", "spam:high", "spam:medium", "duplicate", "internal"] Labels that prevent surveys

Survey types

Scitor supports four survey variants. Pick the one that best matches the metric you care about by setting type: in your csat config.

csat β€” Customer Satisfaction (1-5 stars, default)

Classic 1-5 star rating. Customers click one of: ⭐ Bad, ⭐⭐ Poor, ⭐⭐⭐ OK, ⭐⭐⭐⭐ Good, ⭐⭐⭐⭐⭐ Excellent.

csat:
  enabled: true
  type: csat

How was your support experience?

⭐ Bad · ⭐⭐ Poor · ⭐⭐⭐ OK · ⭐⭐⭐⭐ Good · ⭐⭐⭐⭐⭐ Excellent

The report shows the average rating (e.g. 4.3 / 5) and the distribution across the five buckets.

nps β€” Net Promoter Score (0-10)

The industry-standard β€œwould you recommend us?” question, scored 0 (not at all) to 10 (extremely likely).

csat:
  enabled: true
  type: nps

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

[0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]

0 β€” Not at all likely Β· 10 β€” Extremely likely

Responses are bucketed:

  • Promoters (9-10) β€” loyal enthusiasts
  • Passives (7-8) β€” satisfied but not enthusiastic
  • Detractors (0-6) β€” unhappy customers

The NPS Score is % Promoters βˆ’ % Detractors, expressed as a number from βˆ’100 to +100. Anything above 0 means more promoters than detractors; above +50 is excellent.

thumbs β€” Thumbs up / Thumbs down

Single-question yes/no satisfaction check β€” fastest to answer.

csat:
  enabled: true
  type: thumbs

Was your issue resolved satisfactorily?

πŸ‘ Yes β€” πŸ‘Ž No

The report shows the percentage of positive responses.

ces β€” Customer Effort Score (1-7)

Measures how easy it was for a customer to get help, on a 1-7 scale anchored at β€œVery difficult” β€” β€œNeutral” β€” β€œVery easy”. Lower effort generally correlates with higher loyalty.

csat:
  enabled: true
  type: ces

How easy was it to get help?

[1 Very difficult] [2] [3] [4 Neutral] [5] [6] [7 Very easy]

The report shows the average CES (e.g. 5.8 / 7) and the distribution.

Survey email

Customers receive an email like:

How was your support experience?

Your support request β€œHelp with login” has been resolved.

⭐ Bad · ⭐⭐ Poor · ⭐⭐⭐ OK · ⭐⭐⭐⭐ Good · ⭐⭐⭐⭐⭐ Excellent

Clicking a rating opens a page where they can optionally leave a comment.

Commands

/csat β€” View CSAT summary

Shows all-time and last-30-day CSAT metrics including:

  • Total surveys sent and responses received
  • Response rate percentage
  • Average rating
  • Rating distribution

/csat skip β€” Skip survey for this ticket

Use this on any ticket to prevent a survey from being sent when it’s closed. Useful for internal tickets, test issues, or situations where a survey would be inappropriate.

Reliable by design

  • One survey per ticket β€” duplicate surveys are prevented automatically
  • Secure links β€” each survey has a unique cryptographic token
  • 30-day expiry β€” survey links expire after 30 days
  • One rating per survey β€” ratings can only be submitted once

Report integration

CSAT metrics are automatically included in /generate-report output:

πŸ“Š Customer Satisfaction (CSAT)

| Metric | Value |
|--------|-------|
| Surveys sent | 25 |
| Responses received | 18 |
| Response rate | 72% |
| Average rating | 4.2 / 5 |

⭐⭐⭐⭐⭐ Excellent: 10
⭐⭐⭐⭐ Good: 5
⭐⭐⭐ OK: 2
⭐⭐ Poor: 1

Lifecycle

Event Action
Issue closed Schedule survey (after delay)
Issue reopened Cancel pending survey
/csat skip used Mark survey as skipped
Exclude label present No survey scheduled
Customer unsubscribed Survey not sent

GitHub comment

When a customer submits a rating, a comment is posted on the issue:

πŸ“Š Customer Satisfaction Survey

Rating: ⭐⭐⭐⭐⭐ Excellent Comment: Great support, resolved quickly!

Was this article helpful?

Scitor β€” Turn GitHub into your support platform