Hire an AI customer interview pattern tracker

Everything it can hand back

Your task sets the actual outputs. The specialist picks the methods that cover them.

• Your notes sorted into facts, praise and ideas, with the share that was usable
• Buyer patterns with the number of separate people behind each, marked new, grown or unchanged
• A buyer card for each pattern, with quotes and a confidence level
• The next round's questions and who to talk to
• A call to keep interviewing or stop and act, and one change to test

How it does the job well

• Check the current policy and the customer's order or account facts before replying.
• Answer the actual question in the first two lines, then explain.
• Match the customer's tone. Acknowledge the problem once, without grovelling.
• Hand refunds outside policy, legal threats and safety issues to a person.

Mistakes it avoids

• Calling something a pattern from one person.
• Asking buyers what they would pay, what they would like or whether they would recommend you. The guide rewrites those.
• Counting numbers from interviews as statistics. They show what kinds of buyers exist, not how many.
• Running this when buying is pure price with no struggle, or when you can reach fewer than five buyers. Read their words by hand.
• Using it to learn from reviews alone. Review text is thinner than an interview.
• Promising refunds, dates or fixes the policy does not allow.

What a good result looks like

An example, not your task: You are talking to buyers every few weeks and want the learning to add up.

• ROUND. People so far, new people this round and the share of notes that were facts.
• PATTERNS. Each pattern with the number of separate people, its status and the strongest quote.
• ONE-OFFS AND SURPRISES. Single stories that fit no pattern, with what to ask about them.
• BUYER CARDS. One card per real pattern, with quotes and a confidence level.
• WHAT HOLDS BUYERS BACK. Ranked forces and one change to test, with a number to measure.
• THE CALL. Keep interviewing or stop and act, and the rule that decided it.

How the work is checked

• Every pattern is backed by two or more separate people, and one-offs are not called patterns.
• Every quote is word for word with a person ID and a date.
• No finding rests on praise, a future promise or a request for a feature alone.
• The call to stop or continue names the rule that decided it.
• States which checks passed, failed or could not run, and names missing inputs.

These checks have not been run against a model yet. Your AI uses them to check its own work.

A second opinion

Before it hands back, your AI reviews the draft the way Clayton Christensen and Eric Ries would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• Read the folder. Loads your question, people, patterns and round log. With an empty folder it plans round one and stops.
• Sort the notes. Splits each person's notes into facts, praise and ideas. If more than half of a round is not fact, it rewrites the questions instead of mapping.
• Count the usable stories. Needs at least five real purchase stories before it maps. With fewer, it plans the next round.
• Map the stories. Maps all usable stories together and compares the patterns with last round. A pattern needs two or more separate people.
• Build the cards. Writes a buyer card for each new pattern, marked low confidence while it rests on fewer than five stories.
• Decide whether to continue. Two rounds in a row with nothing new means stop and act. Otherwise it plans the next round around the gaps.

How the role works

1. Checks your task, context and available tools.
2. Picks the methods needed and prepares a work plan.
3. Carries out supported tasks, with each method's limits and review gates.
4. Checks the real deliverables and corrects supported faults.
5. Hands back files, evidence and unresolved gaps.

Your specialist's working setup

Download the base specialist without a task form or account. README.md and SETUP.md explain how to use the folder. SOUL.md defines its scope. INTAKE.md asks missing questions inside your AI. PLAYBOOK.md, SKILLS.md, BEST_PRACTICES.md, CHECKS.md and HANDOFF.md guide the work and its checked deliverables.

Complete associated skills, supporting files and licences are included. Choose only the methods needed for the actual task. Keep context, working files and accepted feedback in your AI project.

Use this specialist through MCP or the API

Call prepare_specialist with role_id customer-interview-program-keeper only. No task is required to load it. Your AI asks what job you need done, what you want back and which relevant context it can use.

For task-led hiring, call find_specialist, explain the suggested specialists and let the user accept or change the team.

Complete ZIP: Download base specialist. API: GET /api/v1/specialists/customer-interview-program-keeper/pack. The host AI supplies its own model, tools and execution; a pack does not connect accounts or run a task.

Inputs and helpful context

The specialist starts from the task brief. Only active methods require their inputs.

For the primary method, when used

• What you want to learn, in one line
• Notes from each person in your latest round, with an ID and a date
• What you sell and who buys it (optional)
• Who you can reach for the next round (optional)

Helpful context (optional)

• A list of recent buyers you can contact
• Your reviews and tickets, to find who to interview

Tools and access

• Claude Code: It can keep the interviews folder, read the record and plan each round.
• A call or meeting recorder: Record with consent and keep exact words, which the notes check needs.
• Any good chat AI: Writing the question list and rewrites is text work any strong chat AI does with this skill.
• A voice recorder: Recording, with consent, keeps the exact words the next step depends on.
• Any good chat AI: Sorting and tagging notes is text work that any strong chat AI does well with this skill.
• A shared spreadsheet or index cards: Sorted notes are easy to filter by mark when each item has its own row or card.
• Any good chat AI: Reading notes and mapping stories is text work any strong chat AI does with this skill.
• A spreadsheet or board: The forces table is easier to sort and share in a spreadsheet.
• Any good chat AI: Any current chat AI can group stories and write the cards. Paste the stories as plain text, with dates.
• Shopify admin: Its order export gives you dates and items bought together, which sharpen the timeline.
• Any good chat AI: Ranking blockers and drafting fixes is text work any strong chat AI does with this skill.
• Your analytics: You need a baseline figure to judge the test.

Functions and deliverables

• Customer interview pattern tracker: Turn real buying stories into shared patterns. Track each round and plan the next interviews or a change to test.
  • Your notes sorted into facts, praise and ideas, with the share that was usable
  • Buyer patterns with the number of separate people behind each, marked new, grown or unchanged
  • A buyer card for each pattern, with quotes and a confidence level
  • The next round's questions and who to talk to
  • A call to keep interviewing or stop and act, and one change to test
• Customer purchase interview script writer: Plan a buying story interview for five to eight recent buyers. You get a timeline script, probes and questions to avoid.
  • Who to talk to and a short request note
  • An opening to set the frame
  • Seven stops of timeline questions with probes
  • Questions about what they used before and what they dropped
  • Questions to avoid, with three of yours rewritten
  • A plan for capturing exact words
• Customer conversation notes sorter: Sort raw customer interview notes into facts, compliments and feature ideas. Get a follow-up question for every weak note.
  • Every note tagged with a symbol and marked as fact, compliment, fluff or idea
  • The follow-up question for each weak note
  • A summary of the real signals (pains, goals, obstacles, workarounds and money)
  • A judgement of each person as a customer, a complainer or not a customer
• Customer switching journey mapper: Turn interview notes from recent buyers into switch stories. See what pushed and held each person back, and the patterns across them.
  • One six-line story per person, each line tied to a note
  • A table of the four forces for each person
  • What each buyer fired, and what tipped the choice
  • A check on whether they used it again
  • Clusters with counts, one-offs and surprises
  • A list of what to ask next
• Customer needs profile builder: Share purchase stories and reviews. Get customer profiles based on their needs, buying triggers and actual quotes.
  • One avatar card per job group, with the job in one situation sentence and the circumstance it happens in
  • Practical, emotional and social sides of the job, the trigger moment and the buying timeline
  • The four forces for each avatar, what they stop using to buy, and their own words, each with a quote
  • A "not for" line per avatar and a list of claims in the quotes that need proof
• Customer switching barrier planner: Rank what stops buyers switching, get two fixes for each, and pick one small change to test first. Built for a buyer switch friction fix.
  • A ranked list of habits and worries with the sums
  • Each worry sorted by type and placed in the path
  • Two fixes for each, with cost and speed
  • Help for getting rid of the old way
  • A line saying who it is not for
  • One small change to test, with its measure and keep rule
• Brand profile builder: Share your website. Get a reusable profile covering your products, customers, voice and approved claims, with sources.
  • A folder of eight plain files, from a start file for your AI to a claims list and a change log
  • A tag on every fact, and every claim marked approved, pending or banned
  • A two-sentence diagnosis of your biggest gap, and a list of what's missing and who can answer it
  • A short slice for your next step, if you name one
• Clayton Christensen: Explore why customers buy using Clayton Christensen’s recorded approach. Start with a real situation, the struggle and what the buyer used before.
  • Your problem worked on the way Clayton Christensen would: their questions, their method, their push-back
  • The answer or the work, in Clayton Christensen's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step
• Eric Ries: Turn a product plan into testable guesses using Eric Ries’s public approach. Start with the riskiest assumption and measure what customers actually do.
  • Your problem worked on the way Eric Ries would: their questions, their method, their push-back
  • The answer or the work, in Eric Ries's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step

Tasks you can use it for

• You are talking to buyers every few weeks and want the learning to add up.
• Your interviews keep giving compliments and no clear reasons for buying.
• You want to know when you have heard enough.

Limits to check before starting

• You buy on price alone, with no struggle or doubt: read your buyers' words by hand instead.
• You can reach fewer than five buyers: read their words by hand, as there is not enough to map.
• You only have reviews: read them yourself, since review text is thinner than an interview.

Best way to use this role

Give it one clear outcome, the actual files and a definition of done. Keep the work in one AI project so it can reuse checked context and earlier files.

• Start a chat with the agent. In Claude Code, add it as a skill or save it in .claude/agents, then tell it what you want to learn.
• The first time, it plans your first round. Run the interviews yourself.
• Paste each person's notes with an ID and a date, and remove names and contact details.
• Read the patterns and the call. Run the next round it plans, or make the change it suggests.
• Run it again after each round. It keeps the record in an interviews folder.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/customer-interview-program-keeper for its role instructions and pinned sources.

This is a prepared AI role. No task has run. Your AI carries out authorized work using its available tools and reports the actual deliverables and checks.