Pain Miner
Find real problems instead of guessing ideas.
SKILL.md in the open agentskills.io standard — works directly in Claude, ChatGPT/Codex, Cursor, Copilot & more.
Pain Miner is a voice-of-customer skill for pain mining: it finds the real problems, wishes, objections, buying motives and switch/churn reasons behind a product, industry, audience, brand, competitor or your own customer data — and turns them into product, marketing, sales and content decisions. Method-grounded (Jobs-to-be-Done, The Mom Test, Value Proposition Canvas, Kano, RICE), it preserves the customer's real language (verbatim, verified), labels every finding by confidence (backed/plausible/hypothesis/weak), and prioritizes instead of listing. Honest by design: hypothesis ≠ evidence, no vocal minority as truth, no fear-mongering.
What this skill does
- Finds real pain points, wishes, objections, buying motives and switch/churn reasons from reviews, communities, search data, competitors or your own customer data
- Ten modes: Quick Pain Scan · Deep Pain Research · Competitor Pain Mining · Review Mining · Search-Intent Mining · Product · Marketing · Sales-Objection · First-Party-Data · Interview Guide (Mom-Test/switch-interview validation)
- Four-layer engine (source router → evidence miner → pain intelligence → output transformer) with two-pass analysis
- Frames every pain as a job-to-be-done + a Force (Push/Pull/Anxiety/Habit) — separates usage from switch/churn pains
- Preserves and verifies the customer's real language (verbatim) — no invented or "Frankenstein" quotes
- Labels every finding by confidence (backed/plausible/hypothesis/weak) + states the sample and denominator
- Prioritizes by Pain Score + RICE/ICE + Kano and delivers a "when NOT to build" verdict
- Turns pains into hooks, claims, USPs, FAQ, feature ideas, roadmap, ad/newsletter/YouTube ideas
- Detects switch reasons (breakup), love signals, workarounds and indirect (silent) pains
- Stays legally clean: APIs/exports over scraping, no bypassing ToS, anonymizes your own data
Description
Pain Miner is a skill for voice-of-customer analysis and pain mining — it stops you building landing pages, features or campaigns on gut feel by making visible what people really struggle with, in their own words. The guiding idea: don't guess ideas, don't flatter features, don't invent problems — find the real language, spot recurring problems, understand objections, cluster wishes, surface buying motives. It runs a four-layer engine: the **Source Router** picks source *classes* (first-party data, reviews, communities, search intent, competitors, tech and social sources) instead of hard-coded URLs; the **Evidence Miner** collects signals and keeps the verbatim; the **Pain Intelligence Engine** codes and clusters (per source first, then across sources, so nuance survives), assigns pain type, emotion, role and buyer-journey stage and scores frequency, severity, buying intent and confidence; the **Output Transformer** turns it into usable deliverables. The backbone is real method: **Jobs-to-be-Done** (every pain is a blocked job, not a missing feature) plus the **Four Forces** (Push/Pull/Anxiety/Habit — how it separates usage pains from switch/churn pains), the **Mom-Test filter** (past behavior and real spend count; opinions, hypotheticals and compliments are discounted; feature requests are translated back to the underlying pain), **Value Proposition Canvas** (jobs/pains/gains) with a **Kano** tag (must-be/performance/delighter), **RICE/ICE** prioritization and **Braun & Clarke thematic analysis** (a theme = an interpretation plus 2–4 real quotes). Pain Miner preserves and **verifies** the customer's real language — no invented or stitched-together "Frankenstein quotes" — labels every finding by confidence, is bias-aware (vocal minority, selection, survivorship, recency) and always states the sample and the denominator. Ten modes cover the whole path: Quick Pain Scan, Deep Pain Research, Competitor Pain Mining, Review Mining, Search-Intent Mining, Product, Marketing, Sales-Objection, First-Party-Data and an Interview Guide mode that turns hypotheses into Mom-Test/switch-interview questions. Every run ends usable — prioritized pains with verbatims, the requested outputs (pain matrix, objection matrix, personas, buyer-journey map, hooks, USPs, FAQ, feature ideas, roadmap) and a "when NOT to build" verdict plus a concrete next validation step. Legally and ethically clean: prefers official APIs/permitted exports over scraping, never bypasses logins/paywalls/ToS, anonymizes your own customer data (handing privacy questions to Privacy Check) and never works manipulatively. Pain Miner is the upstream voice-of-customer layer for Pain-to-Punchline, Deep Research, Audience Architect and SEO Briefing.
Examples
What it does not do
- Doesn't treat a single review as market truth — labels confidence and states the sample honestly
- Doesn't invent pains, numbers, quotes or sources; marks hypotheses and asks to validate
- Doesn't manufacture fear, false urgency or unbacked promises — no manipulative pain marketing
- Doesn't bypass logins, paywalls, scraping protection, rate limits or review caps; no scraping claims
- Doesn't rate or expose individuals from uploaded data — analyzes aggregate patterns
- Doesn't dump an unranked wall — it prioritizes and will say "don't build this"
Compatibility & tech
- Tested (internal)
- 4 scenarios
- Recommended runtime
- Strongest available model; for backed pains one with web/deep research or your own customer data (uploads).
- Modes
- Quick Pain Scan (fast first read) · Deep Pain Research (sources, clusters, confidence) · Competitor Pain Mining · Review Mining · Search-Intent Mining · Product Mode (feature prioritization, RICE, MVP) · Marketing Mode (hooks, claims, landing page) · Sales-Objection Mode (objection matrix) · First-Party-Data Mode (your own data, anonymized) · Interview Guide Mode (Mom-Test / switch-interview validation)
- Inputs
- Product / industry / software / brand / audience / competitor / idea (text) · Reviews, comments, forum threads (pasted) · Your own data: support tickets, CRM notes, churn reasons, NPS, surveys, chat logs · Existing landing page / feature list / roadmap · Keyword / search-query lists
- Output format
- Structured Markdown report (verdict + data basis, top pains with verbatim + confidence, pain matrix, wish list, objection matrix, emotional triggers, persona/buyer-journey mapping, product/marketing/content opportunities, prioritization + "when NOT to build", next validation); four depths (Compact/Standard/Deep/Export) with matrices, heatmaps and copy/briefing export blocks.
- Subcategory
- Voice of customer & pain mining
- License
- Proprietär
Security profile
Runs entirely on your machine with your own AI — no external runtime, no running costs.
Only instructions, templates and references — no executable scripts.
Accesses external sources / the network in normal use (e.g. live pages, search/data APIs).
What you get
- pain-miner-1.0.0/11 files
- .claude-plugin/marketplace.json
plugins/pain-miner/skills/pain-miner/11 files
- SKILL.md
- manifest.json
references/9 files
- first-party-data.md
- interview-guide.md
- mining-plays.md
- outputs-and-templates.md
- pain-miner-checklist.md
- pain-miner-method.md
- pain-taxonomy.md
- scoring-and-evidence.md
- source-router.md
- .agents/skills/pain-miner/→ universal — same content (Codex, Cursor, Copilot, Gemini, Windsurf, Cline)
- LICENSE.txt
Installation
Also works as a chat prompt
No AI tool? Paste it into Claude, ChatGPT or Gemini and use the method right away.
This skill ships no scripts, so the prompt carries its full method.
Works best when your chat has web access.
Installing is the full version — it triggers automatically, runs its scripts and loads references as needed. As a chat prompt you drive the method by hand.
Unlock to copy the ready-to-paste prompt — then in “My Skills”.
Reviews
No reviews yet – be the first.
Note
Research capability depends on the AI platform. Without web search Pain Miner works from hypotheses and supplied data; with web search, deep research or your own data the results get much stronger. It doesn't replace real customer interviews or legal advice — it prepares, prioritizes and flags what still needs validation.
Changelog
- v1.0.001.07.2026[object Object]
Frequently asked questions
What does Pain Miner do?
Pain Miner is a voice-of-customer skill for pain mining: it finds the real problems, wishes, objections, buying motives and switch/churn reasons behind a product, industry, audience, brand, competitor or your own customer data — and turns them into product, marketing, sales and content decisions. Method-grounded (Jobs-to-be-Done, The Mom Test, Value Proposition Canvas, Kano, RICE), it preserves the customer's real language (verbatim, verified), labels every finding by confidence (backed/plausible/hypothesis/weak), and prioritizes instead of listing. Honest by design: hypothesis ≠ evidence, no vocal minority as truth, no fear-mongering.
How do I find out what customers really struggle with instead of guessing?
Pain Miner is a voice-of-customer skill: it extracts real customer pains, wishes, objections, buying motives and churn reasons from reviews, communities, search data, competitors or your own data, preserves the customer's real language (verbatim), labels every finding by confidence (backed/plausible/hypothesis/weak), and turns them into product ideas, hooks, FAQ and roadmap impulses.
Does Pain Miner work without web search?
Yes. Without web search Pain Miner works from your briefing and delivers clearly-labeled hypotheses plus a search strategy and analysis templates. It's strongest with web search, deep research, file uploads or your own customer data (support tickets, reviews, NPS) — that's when hypotheses become backed pain points.
What's the difference between Pain Miner and Pain-to-Punchline?
Pain Miner finds and prioritizes the real pains (upstream research). Pain-to-Punchline takes a known pain and turns it into a punchy ad line. Pain Miner is the voice-of-customer layer before it, and hands its findings to Pain-to-Punchline, Deep Research, Audience Architect, SEO Briefing or Privacy Check.
Does Pain Miner invent problems or quotes?
No. Pain Miner never invents pains, numbers, quotes or sources. Every statement gets a confidence tier, quotes must exist verbatim in a real source (no stitched-together 'Frankenstein quotes'), a single review is never treated as market truth, and with no data everything is clearly marked as a hypothesis.
Is Pain Miner manipulative or fear-based marketing?
No. Pain Miner uses real customer problems to communicate more clearly, helpfully and relevantly — never to manufacture fear, false urgency or unbacked promises. Marketing output is always derived from backed pains, never hallucinated.
Can Pain Miner just scrape reviews and forums?
Pain Miner uses web search, official APIs, permitted exports, public info or your data — and prefers APIs/exports over scraping. It never bypasses logins, paywalls, scraping protection, rate limits or review caps. Reviewer names are personal data (GDPR); for the privacy assessment it hands off to Privacy Check (Datenschutz-Check).
Which AI tools does Pain Miner work with?
Claude (Projects & system prompt) · ChatGPT (Custom GPT & Custom Instructions) · Any LLM · Strongest with web search / deep research / file uploads / connectors · Native UI (tables, buttons/forms) where available, Markdown fallback otherwise
How do I use Pain Miner?
Pain Miner is a SKILL.md in the open agentskills.io standard: install it with one command (npx) or download it and add it to your AI tool — Claude (Projects), ChatGPT (Custom GPT), Cursor, Copilot, Gemini CLI and more. No code needed.