Support ticket trend analysis and fixes

Analyze a batch of tickets to find top issue themes, root causes, deflection opportunities and product or documentation fixes worth making.

Change note: Initial version

Prompt

Analyze the batch of support tickets below for {{product}} and tell me what to fix.

Ticket data (one per line or pasted export):
{{tickets}}

Period and volume: {{period}}
Team goal: {{goal}}

Do this:
1. Group tickets into 5-8 themes. For each theme: name, approximate count and percentage, a representative customer quote (names removed), and the likely root cause.
2. Classify each theme's best fix: product change, documentation or help article, in-app copy, macro or training, or no action. Explain the choice briefly.
3. Rank the top 5 actions by expected ticket reduction against effort. Show your reasoning as High/Medium/Low for each.
4. Highlight anything unusual: sudden spikes, high-emotion tickets, churn risk, security or legal issues.
5. State the limits of this analysis: sample size, ambiguous tickets, anything you couldn't classify.

Count honestly; if the data is too small to draw conclusions, say so. Do not invent numbers, and show how you got each count.

Output as a short report with a table for theme counts and a numbered action list.

Variables

• {{product}}Producttextrequired

  Default: Brightloop

• {{tickets}}Ticketslongrequired

  Default: 1. Can't find where to invite teammates 2. Gantt blank 3. Invoice shows wrong company name 4. How do I invite my team? 5. Export to CSV timing out 6. Where do I change company name on invoices? 7. Invite link says expired 8. Gantt view empty again 9. Need receipt for tax 10. Can't invite user, error

  Paste ticket subjects or summaries, ideally 50+

• {{period}}Period and volumetextrequired

  Default: Last 7 days, 10 tickets shown (sample)

• {{goal}}Goaltextrequired

  Default: Reduce ticket volume by 15% this quarter

Example output

## Theme counts
| Theme | Count | % |
|---|---|---|
| Inviting teammates | 4 | 40% |
| Invoice/billing details | 3 | 30% |
| Gantt view blank | 2 | 20% |
| CSV export timeout | 1 | 10% |

(Tickets 1, 4, 7, 10 = invites; 3, 6, 9 = invoices; 2, 8 = Gantt; 5 = export.)

## Ranked actions
1. **Redesign the invite flow and add an inline help link** (product + in-app copy). Impact High, effort Medium. 40% of the sample; "expired link" and error cases suggest a real flow problem.
2. **Let admins edit company name and tax details on invoices** (product). Impact Medium, effort Low.
3. **Empty-state message on Gantt explaining required dates** (in-app copy). Impact Medium, effort Low.
4. Publish a help article on invites (documentation). Impact Medium, effort Low.
5. Investigate large CSV exports (product). Impact Low in this sample, effort Medium.

## Caveats
Only 10 tickets, so percentages are directional. Re-run with at least 50 tickets before committing roadmap time.

Notes

Showing which ticket numbers fall into each theme keeps the model honest about its counts, and classifying the type of fix turns the report into a backlog. The more tickets you paste, the more reliable it gets.