0 of 10 prompts revealed
The Language Mining Method - 5 Phase Flowchart

Quick Reference

Pick the path that matches your time and goal.

GoalRun These
Full workflow (2-3 hours)Phase 1 → 2 → 3 → 4 → 5 (in order)
Quick version (60 min)Phase 1 → 2 → 3 → 5 (skip Phase 4)
Just need quotesPhase 1 → 2
Sharpen positioningPhase 1 → 2 → 3 (focus on misdiagnosis)
Write copy from researchPhase 5 (after completing 1-3)

Phase 0: Know Your Buyer First

"Who am I mining language for?"

If you already know your ideal customer, skip this and go straight to Phase 1. But if you're fuzzy on who exactly you serve, what problem you solve for them, or why they choose you... run this first. It gives the AI your website (or asks you 5 questions) and produces a complete customer profile you can use as input for the mining prompts below.

🔒 PROMPT: Ideal Customer & Market Research Click to reveal
Ideal Customer & Market Research
## EXECUTE NOW

Run this prompt immediately. Do not ask what to do with it. Do not summarize it. Do not ask for clarification before starting. Check the inputs I have provided at the very bottom of this message and begin Stage 1 now.

---

## YOUR ROLE

You are a business strategy researcher completing Section 2 of a strategic business assessment: Ideal Customer and Market.

This is a 2-stage process. Total time: about 5 minutes.

- Stage 1 of 2: You assess what you know and ask me targeted questions to fill gaps
- Stage 2 of 2: You produce a research brief and draft answers

Do not produce the research brief or draft answers until after I respond to your Stage 1 questions.

---

## THE 7 QUESTIONS YOU ARE HELPING ME ANSWER

1. Who is your ideal customer or client?
2. What major problem does your product or service solve for them?
3. Why do customers typically choose you over competitors?
4. What emotional or psychological outcomes do your customers want?
5. What frustrations do your customers experience before they find you?
6. How sophisticated are your customers in understanding their problem?
7. What industries or niches contain the greatest opportunity for your offer?

---

## STAGE 1 INSTRUCTIONS

Check what I provided at the bottom of this prompt. Then follow the correct path.

**PATH A: Website URL provided**

Step 1. Fetch the website. Read the homepage, services or products page, and about page if accessible.
Step 2. Note any Section 1 answers I included and use them as additional context.
Step 3. Identify which of the 7 questions you can answer confidently from research alone.
Step 4. Identify the 3 gaps only I can fill. These are things no website would reveal.

Then output exactly this:

---
Stage 1 of 2: Research Complete

**Here is what I found:**
[3 to 4 bullet points directly relevant to the 7 questions. Focus on customer language, positioning signals, stated outcomes or problems. Do not summarize what the company does generally.]

**Before I write your draft answers, I have 3 quick questions.**
Each one will sharpen a specific answer I cannot get from research alone.

1. [Question] *(This helps with Q[number]: [question topic])*
2. [Question] *(This helps with Q[number]: [question topic])*
3. [Question] *(This helps with Q[number]: [question topic])*

Take your time. Short answers are fine. Reply when ready and I will move into Stage 2. Or if you would rather stop here, you have my research findings and can complete the 7 questions on your own.

---

**PATH B: No website URL provided**

Step 1. Use any Section 1 answers I included as your starting point.
Step 2. If no Section 1 answers were provided either, acknowledge that briefly and proceed.
Step 3. Ask 5 questions that cover the same ground the website would have covered. Make them specific enough that my answers will let you draft all 7 Section 2 questions well.

Then output exactly this:

---
Stage 1 of 2: Getting Started

**I don't have a website to research, so I have 5 questions.**
These cover the same ground research would have. Short answers are fine.

1. [Question] *(This helps with Q[number]: [question topic])*
2. [Question] *(This helps with Q[number]: [question topic])*
3. [Question] *(This helps with Q[number]: [question topic])*
4. [Question] *(This helps with Q[number]: [question topic])*
5. [Question] *(This helps with Q[number]: [question topic])*

Reply when ready and I will move into Stage 2. Or if you would rather stop here, that's completely fine.

---

**QUESTION RULES FOR BOTH PATHS:**
- Do not ask about anything already visible on the website or covered in Section 1
- Every question must connect directly to one of the 7 questions
- Path A: exactly 3 questions. Path B: exactly 5 questions. No exceptions.

---

## STAGE 2 INSTRUCTIONS

After I respond, produce both deliverables below in full.

---
Stage 2 of 2: Research Brief + Draft Answers

**SECTION 2 RESEARCH BRIEF**

Start with a TL;DR: 2 sentences maximum capturing who the customer is and what core problem is being solved.

Then write a 150 to 250 word synthesis covering:
- Who the customer appears to be and what they care about
- The core problem and how this business positions around it
- Customer sophistication signals and emotional motivators
- Any market positioning observations relevant to the 7 questions

---

**SECTION 2 DRAFT ANSWERS**

For each of the 7 questions, write a draft answer using these rules:
- Draw from all available context: website research if done, Section 1 if provided, my answers to your questions
- Write in first person as if I am the business owner speaking
- 2 to 4 sentences per answer
- Mark any inferred claim with [VERIFY]
- Be specific. Generic answers fail this exercise.

Format each answer as:

**Q1: [question text]**
[Draft answer]

**Q2: [question text]**
[Draft answer]

...through Q7.

Then close with exactly this line:

"That's Stage 2 of 2 complete. These are working drafts built from everything you've shared. Review each one and correct anything marked [VERIFY]. If you want to sharpen a specific answer, just tell me which one."

---

## HARD RULES

- Do not skip Stage 1 and jump to draft answers
- Do not ask more questions than specified per path
- Do not produce walls of text. Use the stage markers and structure above
- Use [VERIFY] generously. It makes assumptions transparent, not embarrassing
- If I ask to stop at any point, confirm that stopping is valid and summarize what was completed

---

======= FILL IN YOUR DETAILS BELOW BEFORE HITTING SEND =======

**WEBSITE URL**
(Your business website. Delete this line if you don't have one.)


**SECTION 1 ANSWERS**
(Paste your Section 1 answers below this line. Delete this line if you haven't completed Section 1.)


======= END =======

What You Get

A 7-question customer profile with draft answers you can use as input for Phase 1. Paste the output into Phase 1's "target buyer" field and your mining results will be dramatically more specific.

Phase 1: Adjacent Territory Mapping

"Where does my buyer's pain show up under different names?"

Your buyers don't search for your solution. They search for their problem, but they describe it differently than you do. This prompt identifies where that pain appears under different names, in different communities, using different vocabulary.

You provide: (1) Your core problem statement (1-2 sentences), (2) Your target buyer (be specific), and (3) Any known adjacent concepts.

🔒 PROMPT: Adjacent Territory Mapping Click to reveal
I need to find where my target buyers discuss their problems online—but they won't be using my terminology. Help me map adjacent territories. MY CORE PROBLEM: [Paste your problem statement] MY TARGET BUYER: [Paste your buyer description] ADJACENT CONCEPTS I'VE HEARD: [List any terms you've heard, or write "None yet"] Please identify: 1. CONCEPTUAL ADJACENCIES What academic/professional concepts describe this same problem? - Psychology terms - Business/management terms - Industry-specific jargon - Research terminology 2. BOOK ADJACENCIES What books do people read when experiencing this problem? - Business books - Self-help books - Industry-specific books - Academic texts that crossed over to mainstream For each book: Why would my buyer read this? What problem are they trying to solve? 3. COMMUNITY ADJACENCIES Where do these people gather online? - Relevant subreddits (with estimated size) - Industry forums - Professional networks (LinkedIn groups, Slack communities) - Review sites (G2, Capterra, course platforms) - Q&A sites (Quora topics) 4. EMOTIONAL ADJACENCIES What emotional states accompany this problem? - Frustration patterns - Fear patterns - Aspiration patterns What would someone search when feeling each emotion? 5. FAILURE ADJACENCIES What have they probably tried that didn't work? - Products/services - DIY approaches - Advice they've received Where would they complain about these failures? Output as a prioritized research plan: - Top 5 books to mine reviews from - Top 5 communities to search - Top 10 search queries to run - Top 5 emotional/problem phrases to look for

Human Checkpoint

  • Do these adjacencies feel right? Would my buyer actually be in these places?
  • Are there obvious communities or books missing?
  • Which 3-5 sources should we prioritize first?

Add any adjacencies you know about that AI missed.

Phase 2: Raw Language Mining

"What are the exact words real people use?"

We want RAW language. Not summaries, not articles about the problem, but actual humans expressing their actual pain in their actual words.

🔒 PROMPT: Deep Language Mining Click to reveal
Research the following sources to extract raw customer language about [PROBLEM]. SOURCES TO MINE: [Paste prioritized list from Phase 1] SPECIFIC SEARCH QUERIES: [Paste search queries from Phase 1] For each source, extract: 1. VERBATIM PAIN QUOTES - Exact phrases people use to describe the problem - Copy word-for-word, including typos and informal language - Note the source and context for each 2. EMOTIONAL LANGUAGE PATTERNS - Frustration expressions ("I'm so tired of...") - Fear expressions ("I keep worrying that...") - Shame expressions ("I feel like a fraud when...") - Hope expressions ("I just want to...") - Relief expressions ("Finally someone who...") 3. METAPHORS AND ANALOGIES - How do they visualize the problem? - What comparisons do they make? - What imagery do they use? 4. QUESTIONS THEY ASK - Exact questions posted in forums - Questions in book reviews - Questions that reveal their mental model 5. FAILED SOLUTIONS MENTIONED - What have they tried? - Why did it fail? - How do they describe the failure? 6. DESIRED OUTCOMES - How do they describe "solved"? - What would success look like to them? - What metrics/timeframes do they mention? 7. OBJECTIONS AND SKEPTICISM - What are they skeptical about? - What objections do they raise to solutions? - What makes them hesitate? FORMAT EACH FINDING AS: --- **Quote:** "[Exact verbatim quote]" **Source:** [Where found - be specific] **Context:** [What prompted this - thread topic, book being reviewed, etc.] **Category:** [Pain/Emotion/Metaphor/Question/Failed Solution/Outcome/Objection] **Usability:** [How/where this could be used in messaging] --- Prioritize: - Quotes with strong emotional charge - Metaphors that are vivid and specific - Questions that reveal misunderstandings - Failures that your solution actually addresses Minimum extraction: 30 distinct quotes across categories

Good Mining Looks Like

  • Quotes feel raw and unpolished
  • You can feel the emotion in the words
  • Sources are specific and traceable
  • Mix of pain, hope, and frustration
  • Metaphors are vivid and specific
  • Questions reveal their mental model
  • You're surprised by at least one finding

Bad Mining Looks Like

  • Quotes sound like marketing copy
  • Everything confirms what you already believed
  • Sources are vague ("from a forum")
  • All pain, no hope or aspiration
  • Generic phrases anyone could say
  • No emotional charge
  • Nothing surprising

Phase 3: Pattern Synthesis

"What surprised me? What confirmed? What's missing?"

This phase transforms raw data into insights by forcing three questions: What surprised you? What confirmed what you already believed? What's missing that you expected to find?

🔒 PROMPT: Pattern Synthesis Click to reveal
Analyze the language mining results and synthesize patterns. RAW DATA: [Paste all quotes and findings from Phase 2] MY INITIAL HYPOTHESES ABOUT THIS AUDIENCE: [What did you believe about their pain before this research?] Analyze and report: 1. THREE THINGS THAT SURPRISED ME For each: - What the finding was - Why it's counterintuitive - What it means for messaging - Specific quotes that prove it 2. THREE THINGS THAT CONFIRMED For each: - What was validated - How strong the evidence is - How to leverage this confirmation - Specific quotes that prove it 3. THREE GAPS OR MISSING ELEMENTS For each: - What I expected to find but didn't - Possible reasons it's missing - Whether to investigate further or let it go 4. DOMINANT EMOTIONAL PATTERNS - Primary emotion (what most people feel) - Secondary emotion (what underlies the primary) - Trigger moments (what activates the emotion) - Language clusters around each emotion 5. THE MISDIAGNOSIS PATTERN - What do they THINK their problem is? - What is their problem ACTUALLY? - Why do they miss the real problem? - How does your solution address the real problem? 6. THE LANGUAGE TRANSLATION TABLE | What They Say | What It Means | Your Term For It | |---------------|---------------|------------------| [Map their vocabulary to your concepts] 7. THE "NOBODY IS SAYING THIS" OPPORTUNITY - What truth exists that no one is articulating? - What would be surprising/relieving for them to hear? - How could you own this message?

Human Checkpoint

  • Do the surprises actually surprise YOU? (If not, dig deeper)
  • Does the misdiagnosis pattern ring true to your experience?
  • Is there a "nobody is saying this" insight that feels ownable?

Add your own observations: What patterns do you see that AI missed? What connects to your direct experience with clients? What contradicts what you've heard firsthand?

Phase 4: Knowledge Asset Creation

"What reusable documents should exist from this research?"

Before jumping to application, decide what reusable assets this research should produce. Not everything needs a document. Pick what you'll actually use.

Standard Asset Menu

AssetWhat It IsWhen It's Valuable
Customer Language BibleSearchable reference of all extracted language organized by categoryAlways. This is your source doc.
Translation DictionaryTwo-column mapping of their words to your wordsWhen your terminology differs from theirs
Objection EncyclopediaObjections with emotional root cause and reframe languageBefore sales conversations or page rewrites
Story FileCollection of proof points and stories with full source/contextBefore presentations or content creation
Presentation Language GuideSlide-by-slide language recommendationsBefore specific presentations
Sales Page Swipe FileOrganized headlines, bullets, and proof pointsBefore copywriting
Content Angle BankTopics and hooks derived from questions they askBefore content calendar planning
Email Sequence SeedsPain-to-solution flows based on emotional patternsBefore email marketing
🔒 PROMPT: Asset Recommendations Click to reveal
Based on the research synthesis, recommend which knowledge assets to create. SYNTHESIS RESULTS: [Paste Phase 3 output] MY IMMEDIATE USE CASE: [What do you need this research for right now?] For each recommended asset: 1. ASSET NAME 2. WHY IT'S VALUABLE for your situation 3. WHAT IT WOULD CONTAIN (specific sections) 4. ESTIMATED CREATION TIME 5. PRIORITY (Must Have / Should Have / Nice to Have) Recommend maximum 4 assets: - 1-2 "Must Have" for immediate use case - 1-2 "Should Have" for ongoing use For the "Must Have" assets, provide a detailed outline of what each section would contain.

Phase 5: Immediate Application

"How do I use this for my specific deliverable?"

Pick the template that matches what you're building. Each one takes your research and shapes it into a specific deliverable.

Quick-Start Version

The compressed workflow for when you have 60 minutes.

PhaseTimeWhat to Do
Phase 110 minRun adjacent territory mapping. Pick top 3 sources.
Phase 230 minRun language mining on just those 3 sources. Get 15-20 quotes.
Phase 310 minQuick synthesis. What surprised you? One key insight.
Phase 4skipGo straight to application.
Phase 510 minApply to immediate need using relevant template.

Quality Standards

How to tell if your mining session produced real signal.

Good Mining

  • Quotes feel raw and unpolished
  • You can feel the emotion in the words
  • Sources are specific and traceable
  • Mix of pain, hope, and frustration
  • Metaphors are vivid and specific
  • Questions reveal their mental model
  • You're surprised by at least one finding

Bad Mining

  • Quotes sound like marketing copy
  • Everything confirms what you already believed
  • Sources are vague ("from a forum")
  • All pain, no hope or aspiration
  • Generic phrases anyone could say
  • No emotional charge
  • Nothing surprising

When to Re-Run This Process

  • Every 6-12 months (language evolves)
  • When entering a new market segment
  • When a major competitor emerges
  • When conversion rates drop
  • When you notice your copy feeling stale
  • After a positioning pivot