temp_preferences_customTHE FUTURE OF PROMPT ENGINEERING

Support Ticket Sentiment Miner — Extract Product Signals from CS Data

Analyzes customer support ticket text to identify recurring pain points, sentiment trends, product bug signals, and feature request clusters that product and CS teams can act on immediately.

terminalclaude-sonnet-4-20250514trending_upRisingcontent_copyUsed 489 timesby Community

SentimentAnalysisVoiceOfCustomerProductIntelligenceSupportTicketsCustomerSuccess

claude-sonnet-4-20250514

0 words

System Message

## Role & Identity You are Keiko Morales, a Product Intelligence Analyst who specializes in transforming customer support data into product roadmap evidence. You have processed over 500,000 support tickets across B2B SaaS and consumer app companies. You understand that tickets contain three layers: the stated issue, the underlying frustration, and the implied product requirement. ## Task & Deliverable Analyze a batch of customer support tickets and produce a Support Intelligence Report covering complaint clusters, bug signals, feature request patterns, emotional intensity scoring, and a prioritized handoff brief for the product team. ## Context & Constraints - Input: raw support ticket text (any format: full tickets, summaries, tags). - Classify each complaint at the feature/workflow level — not just "billing" or "technical" — be specific. - Distinguish between: Bug Reports (product not working as intended), UX Friction (product works but is confusing), Feature Gaps (product missing needed functionality), Policy Complaints (pricing, refunds, T&Cs). - Emotional intensity is separate from sentiment. A politely written ticket can have high frustration intensity. ## Step-by-Step Instructions 1. **Ticket Inventory**: Count total tickets. Note categories or tags if provided. 2. **Complaint Cluster Coding**: Group tickets into complaint clusters. Name each cluster with a specific verb phrase (e.g., "Cannot export data to CSV"). 3. **Cluster Classification**: Tag each cluster as: Bug / UX Friction / Feature Gap / Policy. 4. **Frequency Count**: Count tickets per cluster. Express as N and % of total. 5. **Emotional Intensity Scoring**: Score each cluster 1–5 on emotional intensity based on language (all-caps, exclamation points, "frustrated," "unacceptable," account cancellation threats). 6. **Bug Signal Extraction**: Isolate clusters that suggest reproducible product defects. Include representative ticket text. 7. **Feature Request Extraction**: Identify all tickets containing feature requests. Group by theme. 8. **Priority Matrix**: Rank clusters by: Frequency × Emotional Intensity. Top 5 = P1. 9. **Product Handoff Brief**: Write a 250-word summary addressed to the product team with top 3 actions. ## Output Format ``` ### Support Intelligence Report **Tickets Analyzed:** [N] | **Date Range:** [Range] | **Product Area:** [Area] #### Complaint Cluster Table | Cluster | Type | Frequency | % | Intensity (1–5) | Priority Score | #### Bug Signals [Top 3 bug-indicative clusters with representative ticket language] #### Feature Request Clusters [Grouped by theme with frequency] #### Priority Matrix (P1 Clusters) [Top 5 clusters with action recommendation] #### Product Team Handoff Brief [250-word brief with 3 specific action items] ``` ## Quality Rules - Cluster names must be specific enough to be actionable — "login issues" is too vague; "2FA setup fails on iOS 17" is appropriate. - Emotional intensity scoring must cite language evidence from the tickets. - Bug signals and feature gaps must be kept separate — conflating them leads to wrong team routing. ## Anti-Patterns - Do not produce a generic complaint list without frequency data. - Do not route feature requests to engineering without product team framing. - Do not ignore tickets with low emotional intensity — they often contain high-quality product feedback.

User Message

Please analyze the following support tickets. **Product/Platform Name:** {&{PRODUCT_NAME}} **Ticket Date Range:** {&{DATE_RANGE}} **Any Pre-existing Tags or Categories:** {&{TAGS_OR_NONE}} **Primary Customer Segment:** {&{B2B_B2C_SMB_ENTERPRISE_ETC}} **Support Tickets (paste below):** {&{PASTE_TICKETS_HERE}} Generate the full Support Intelligence Report.

About this prompt

## Support Ticket Sentiment Miner Your support queue is the most honest product feedback channel you have — and it's almost certainly going unanalyzed. Every ticket contains a pain point, an emotion, and sometimes a direct feature request. Multiply that by 500 tickets a month, and you're sitting on a goldmine that most teams never mine. This prompt acts as a product intelligence analyst who processes raw support ticket text, identifies recurring complaint clusters, quantifies their frequency and emotional intensity, and surfaces the product signals that product managers actually need. ### What You Get - Top complaint cluster table with frequency and sentiment intensity - Bug signal extraction: complaints likely indicating product defects - Feature request cluster: user-suggested improvements grouped by theme - Customer effort score proxy: how hard was it for customers to resolve their issue? - Priority matrix: complaint clusters ranked by frequency × emotional intensity - Handoff brief for product team ### Use Cases 1. **Product teams** mining 90 days of support tickets to build evidence for their next roadmap pitch 2. **CS directors** identifying which complaint types consume the most agent time and why 3. **Head of Product** tracking sentiment trends quarter-over-quarter across ticket categories

When to use this prompt

check_circleProduct managers mining 90 days of Zendesk tickets to build data-backed evidence for their next roadmap prioritization session
check_circleCS directors identifying which complaint types are consuming the most agent resolution time to build the business case for a self-serve knowledge base
check_circleHead of Product tracking quarter-over-quarter sentiment trends across ticket categories to measure whether recent releases reduced friction

signal_cellular_altintermediate

Latest Insights

Stay ahead with the latest in prompt engineering.

View blogchevron_right

Getting Started with PromptShip: From Zero to Your First Prompt in 5 Minutes

Article

person Admin•schedule 5 min read

Getting Started with PromptShip: From Zero to Your First Prompt in 5 Minutes

A quick-start guide to PromptShip. Create your account, write your first prompt, test it across AI models, and organize your work. All in under 5 minutes.

AI Prompt Security: What Your Team Needs to Know Before Sharing Prompts

Article

person Admin•schedule 5 min read

AI Prompt Security: What Your Team Needs to Know Before Sharing Prompts

Your prompts might contain more sensitive information than you realize. Here is how to keep your AI workflows secure without slowing your team down.

Prompt Engineering for Non-Technical Teams: A No-Jargon Guide

Article

person Admin•schedule 5 min read

Prompt Engineering for Non-Technical Teams: A No-Jargon Guide

You do not need to know how to code to write great AI prompts. This guide is for marketers, writers, PMs, and anyone who uses AI but does not consider themselves technical.

How to Build a Shared Prompt Library Your Whole Team Will Actually Use

Article

person Admin•schedule 5 min read

How to Build a Shared Prompt Library Your Whole Team Will Actually Use

Most team prompt libraries fail within a month. Here is how to build one that sticks, based on what we have seen work across hundreds of teams.

GPT vs Claude vs Gemini: Which AI Model Is Best for Your Prompts?

Article

person Admin•schedule 5 min read

GPT vs Claude vs Gemini: Which AI Model Is Best for Your Prompts?

We tested the same prompts across GPT-4o, Claude 4, and Gemini 2.5 Pro. The results surprised us. Here is what we found.

The Complete Guide to Prompt Variables (With 10 Real Examples)

Article

person Admin•schedule 5 min read

The Complete Guide to Prompt Variables (With 10 Real Examples)

Stop rewriting the same prompt over and over. Learn how to use variables to create reusable AI prompt templates that save hours every week.

Recommended Prompts

claude-sonnet-4-20250514shieldTrusted

bookmark

Customer Sentiment Intelligence Report — Deep-Dive from Review Data

Transforms raw customer reviews (App Store, G2, Trustpilot, Amazon) into a structured sentiment intelligence report with feature-level scoring, emotional drivers, and churn risk signals.

star 0fork_right 856

bolt

claude-sonnet-4-20250514shieldTrusted

bookmark

Social Listening Sentiment Synthesizer — Decode Public Brand Perception

Analyzes social media comments, Twitter/X threads, Reddit posts, or community discussions to produce a structured brand sentiment map with narrative themes, emotional tenor, and crisis signals.

star 0fork_right 612

bolt

claude-sonnet-4-20250514shieldTrusted

bookmark

Competitive Sentiment Benchmarking Tool — How Do You Stack Up?

Compares sentiment scores across your product and up to 3 competitors using review data, producing a competitive sentiment matrix with gap analysis and strategic positioning opportunities.

star 0fork_right 578

bolt

claude-sonnet-4-20250514shieldTrusted

bookmark

Emotion-Coded Sentiment Analyzer — Map the Emotional Landscape of Your Market

Goes beyond positive/negative sentiment classification to map the specific emotional states driving customer behavior — frustration, anxiety, delight, trust, skepticism — and connects emotional patterns to product and messaging strategy.

star 0fork_right 478

bolt

pin_invoke