OperatorRadar
DiscoverAI ToolsAI AgentsInsights
Submit a ToolFind My Solution
OperatorRadar

Find the tools, systems, and ideas that move your business forward. Timeless business thinking, rebuilt for the AI era.

Discover

  • Discover
  • AI Tools
  • AI Agents
  • Software
  • Agencies
  • Categories

Decide

  • Decision Guides
  • Compare
  • Find My Solution
  • Insights

Execute

  • Prompts
  • Workflows
  • Submit a Tool
  • Contact

Company

  • About
  • Privacy
  • Terms

© 2026 OperatorRadar. All rights reserved.

Built by Ekofi

  1. Home
  2. Prompts
  3. Generate FAQ Candidates from Support Tickets
Featured
beginner
Claude or ChatGPT (verify current model choice)
v1.0.0

Generate FAQ Candidates from Support Tickets

Mine your support queue for the questions you answer 50 times a month. Turn ticket patterns into an FAQ that actually stops tickets.

Full prompt

You are a support operations analyst. Your job is to find the questions that are costing us the most time and repeat effort.

Analyze the following {{TICKET_COUNT}} anonymized support tickets from the past {{TIME_PERIOD}}. For each ticket, note:
- The core question or problem
- How long it took to resolve
- What type of customer asked it ({{CUSTOMER_SEGMENT_FIELD}})
- Whether it was a repeat of a previous ticket

Then output:
1. A list of FAQ candidates ranked by impact (frequency × average resolution time)
2. For each candidate: the question, ticket count, average resolution time in minutes, affected segments, and a one-sentence answer
3. Flag any questions that appear in {{THRESHOLD_PCT}}% or more of tickets
4. Suggest which FAQ items would save the most support time if published

Tickets:
{{TICKET_DATA}}

Output as a structured list. Be specific about numbers. Do not generalize.

How to use

Export your last 30–90 days of support tickets (Zendesk, Intercom, Freshdesk, or CSV). Include ticket subject, resolution time, customer segment, and resolution notes. Paste into the prompt. Run weekly to catch emerging patterns. Publish top 3–5 FAQ candidates immediately; test impact by tracking ticket volume before/after.

Example input

Ticket 1: Subject: 'How do I reset my API key?' | Resolution time: 4 min | Segment: Enterprise | Notes: User forgot password, sent reset link
Ticket 2: Subject: 'API key reset' | Resolution time: 3 min | Segment: Startup | Notes: Same issue, sent reset link
Ticket 3: Subject: 'Can I use the same API key across environments?' | Resolution time: 12 min | Segment: Enterprise | Notes: Explained key scoping, sent docs
Ticket 4: Subject: 'API key reset' | Resolution time: 5 min | Segment: SMB | Notes: Repeat of Ticket 1
Ticket 5: Subject: 'Billing invoice not showing' | Resolution time: 8 min | Segment: Startup | Notes: Invoice delay, checked backend

Example output

FAQ Candidates (ranked by impact):

1. How do I reset my API key?
   - Frequency: 3 tickets (10% of sample)
   - Avg resolution time: 4 min
   - Affected segments: Enterprise, Startup, SMB
   - Answer: Visit Settings > API Keys > Reset. You'll receive a confirmation email. Old keys expire in 24 hours.
   - Impact: 12 min saved per week (estimated)

2. Can I use the same API key across environments?
   - Frequency: 1 ticket (3% of sample)
   - Avg resolution time: 12 min
   - Affected segments: Enterprise
   - Answer: No. Generate separate keys for dev, staging, and production. This prevents accidental data exposure.
   - Impact: 12 min saved per occurrence

3. Why isn't my invoice showing in billing?
   - Frequency: 1 ticket (3% of sample)
   - Avg resolution time: 8 min
   - Affected segments: Startup
   - Answer: Invoices appear 24–48 hours after payment processing. Check your email spam folder or contact billing@company.com.
   - Impact: 8 min saved per occurrence

Customization tips

  • Weight by segment: If Enterprise asks a question once but SMB asks it 10 times, don't assume Enterprise is more important. Calculate impact as (frequency × resolution time × cost-per-minute-of-support).
  • Track resolution time, not just volume: A question asked 5 times in 2 minutes each is less urgent than one asked 2 times in 15 minutes each. Prioritize by total time saved.
  • Flag 'explain' vs. 'bug' questions: Some tickets reveal product confusion (FAQ-able). Others reveal bugs (engineering backlog). Separate them in your output.
  • Run this monthly, not once: Support patterns shift with new features, pricing changes, and seasonal customer cohorts. Build FAQ refresh into your ops calendar.
  • Test FAQ impact: After publishing, track whether ticket volume for that topic drops. If it doesn't, the answer wasn't clear enough or customers aren't finding it.