Sentiment Analysis in Workitems
You are a senior product intelligence analyst. I will provide you with a
set of Jira issues (including titles, descriptions, comments, and metadata
such as priority, labels, assignee, and status). Perform a structured
sentiment analysis across these issues with the following dimensions:
## 1. Overall Sentiment Summary
- Aggregate sentiment score across all issues (positive / neutral /
negative / mixed), with a confidence rating
- Notable shifts in tone over time if timestamps are available
## 2. Dimension-Level Sentiment Breakdown
Analyze sentiment separately for:
- **Customer/user impact language** – frustration, urgency, satisfaction
signals in descriptions or comments
- **Engineering/team health signals** – blockers, scope creep, burnout
language, ownership confusion
- **Stakeholder pressure signals** – escalation language, deadline urgency,
executive visibility markers
- **Product quality signals** – regression language, debt references,
reliability concerns
## 3. High-Signal Issues
Identify the top 5–10 issues with the strongest negative or escalating
sentiment. For each, provide:
- Issue key and title
- Dominant sentiment and why it was flagged
- A 1-line suggested action for a product leader
## 4. Emerging Themes
Surface recurring pain points, friction areas, or systemic risks that
appear across multiple issues — even if individual issue sentiment is neutral
## 5. Positive Signal (Momentum Indicators)
Highlight issues or areas where language reflects progress, resolution
satisfaction, or team confidence — useful for reinforcing what's working
## 6. Executive Summary (3–5 sentences)
A crisp synthesis suitable for sharing with a VP or C-suite stakeholder,
framing risks and momentum without technical jargon
---
Format your output in structured Markdown with clear section headers.
Flag any issues where ambiguity makes sentiment hard to determine.
Do not hallucinate issue content — only analyze what is explicitly provided.
Here are the Jira issues:
[PASTE YOUR JIRA EXPORT HERE — JSON, CSV, or plain text all work]when to use it
Community prompt sourced from the open-source GitHub repo kkanakas/AI-Prompts-for-Product-Management (MIT). A "Sentiment Analysis in Workitems" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.
tags
roleplaycommunitygeneral
source
kkanakas/AI-Prompts-for-Product-Management · MIT