Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for management consultants

Interview Guide and Insight Coding Agent

Produce evidence-backed themes from interview notes with contradictions and gaps made visible

AnalysisAutomationManagement
See how it worksGet access
Interview Guide and Insight Coding Agent: what goes in, what the agent does and what you get

What it does

After 30 stakeholder interviews a consultant has a pile of notes and a deadline, and the themes in the final deck depend on which quotes were remembered. This agent codes the notes the same way every time. It reads each note set and tags statements against a theme list built from the interview guide, adding new themes when several people raise something unexpected. It checks that every theme has evidence from several sources and from different groups, so one loud voice does not become a finding. It searches for contradictions between interviewees and suggests follow-up interviews to resolve them. After new notes come in, it recodes and updates counts. The consultant approves the findings. Edge case: a theme comes from only two people at the same level. The agent marks it as thin and asks for a third source.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueYes, continueApprovedNoNo 1 STARTS WHEN New interview notes uploaded 2 USES A TOOL Read the notes and extract statements with speakerand role 3 DOES Tag each statement against the theme list 4 DOES Add new themes when 3 or more people raise somethingunplanned 5 CHECKS THE RESULT Does each theme have at least 3 sources from 2groups? If not: Mark it thin and suggest a follow-up interview,then recode. Back to step 3. 6 DOES Search for contradictions between interviewees 7 DOES Suggest follow-up questions and interviewees forgaps 8 CHECKS THE RESULT Do new notes change any theme count? If not: If counts changed, recode all notes and updatethe summary. Back to step 3. 9 YOU APPROVE Consultant approves the findings 10 RESULT Theme table with quotes, counts and open questions
Read the steps as a list
  1. New interview notes uploaded
  2. Read the notes and extract statements with speaker and role
  3. Tag each statement against the theme list
  4. Add new themes when 3 or more people raise something unplanned
  5. Does each theme have at least 3 sources from 2 groups?If not: Mark it thin and suggest a follow-up interview, then recode. Back to step 3.
  6. Search for contradictions between interviewees
  7. Suggest follow-up questions and interviewees for gaps
  8. Do new notes change any theme count?If not: If counts changed, recode all notes and update the summary. Back to step 3.
  9. Consultant approves the findingsThe agent waits here for your OK.
  10. Theme table with quotes, counts and open questions

How it decides

A theme counts as a finding only with at least 3 sources from at least 2 groups. Statements that conflict are listed side by side, never averaged.

  • A theme needs 3 sources from 2 groups to count
  • Opposite views are kept as a contradiction with names removed
  • A topic mentioned by senior people only is flagged as potentially biased
  • A new theme needs 3 mentions before it joins the list

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Minimum sources and groups per theme (default 3 and 2)
  • Initial theme list
  • Anonymity level of quotes
  • Format of the findings table
  • Number of follow-up interviews to suggest

What keeps you in control

It always asks you first

  • Consultant approves the final themes before they go to the client

Hard limits

  • Never names an interviewee in client-facing output unless allowed
  • Quotes must be exact, never rewritten

It stops when

  • Done: all themes meet the evidence rule and the consultant approves
  • Stop: fewer than 10 interviews coded, report that evidence is too thin

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensAfter 18 interviews the agent finds Slow approvals in 11 notes across 4 teams, which passes. Data quality appears in only 2 notes, both from finance, so it is marked thin. It suggests two follow-up interviews in sales. After the new notes, 5 of 7 mention it, and the theme is promoted. The consultant approves.

More agents for management consultants

Analysis · Quality control

Market Sizing Cross-Check Agent

A market size backed by two independent estimates, with the gap explained and every assumption sourced.

10 steps2 checks6 jobs
Reporting · Management

Engagement Workstream Status Rollup Agent

Deliver one consistent, checked status report each week without manual stitching.

10 steps2 checks7 jobs
Quality control · Management

Recommendation Evidence Chain Agent

Make every recommendation in the deck traceable to data the team can show.

10 steps2 checks7 jobs
Document handling · Automation

Lessons Learned Capture and Reuse Agent

Make past lessons show up in new project plans and prove they worked.

11 steps2 checks6 jobs