Skip to main content

Learning Roadmap

How to Become a AI Jobs-to-be-Done Analyst

A step-by-step, phase-based learning path from beginner to job-ready AI Jobs-to-be-Done Analyst. Estimated completion: 5 months across 4 phases.

4 Phases
20 Weeks Total
Medium Entry Barrier
Intermediate Difficulty
← AI Jobs-to-be-Done Analyst Overview Interview Prep →
Your Progress 0 / 4 phases

Progress saved in your browser — no account needed.

  1. Foundations - JTBD Theory & AI Literacy

    4 weeks
    Goals
    • Master JTBD framework theory including forces of progress, job maps, and switch interviews
    • Build working literacy in modern AI capabilities (LLMs, embeddings, agents, RAG)
    Resources
    • "Competing Against Luck" by Clayton Christensen
    • "When Coffee and Kale Compete" by Alan Klement (free online)
    • DeepLearning.AI short courses on LLMs and LangChain
    • OpenAI Cookbook and Anthropic documentation
    Milestone

    You can conduct a structured JTBD interview, write job stories, and explain LLM capabilities and limitations to a non-technical audience.

  2. Applied Research & Workflow Analysis

    6 weeks
    Goals
    • Learn to synthesize qualitative research into quantified opportunity scores
    • Practice workflow decomposition to identify AI automation and augmentation insertion points
    Resources
    • "Demand-Side Sales 101" by Bob Moesta for interview depth
    • Dovetail research repository tutorials
    • Opportunity scoring template by Tony Ulwick (Strategyn)
    • Real-world AI product teardowns on Lenny's Newsletter and Product Hunt
    Milestone

    You can run an end-to-end JTBD research sprint: interview, synthesize, score opportunities, and produce an AI opportunity brief.

  3. AI Feasibility & Prompt Experimentation

    6 weeks
    Goals
    • Build basic Python scripts to test LLM performance against job outcomes
    • Learn to evaluate AI outputs using relevance, safety, latency, and cost metrics
    Resources
    • Python for Data Analysis by Wes McKinney (selected chapters)
    • LangChain documentation and quickstart guides
    • Weights & Biases evaluation tracking tutorials
    • Anthropic prompt engineering interactive tutorial
    Milestone

    You can build a lightweight evaluation harness to test whether current AI can fulfill a specific job story, and communicate results to engineering.

  4. Strategy, Prioritization & Portfolio Work

    4 weeks
    Goals
    • Learn to present AI opportunity portfolios to executive stakeholders
    • Build 2-3 portfolio-ready case studies demonstrating JTBD-to-AI-feature pipelines
    Resources
    • "Inspired" by Marty Cagan (product strategy chapters)
    • Case study templates from Reforge or Product School
    • Practice presenting to product communities (Mind the Product, Lenny's community)
    Milestone

    You can lead a cross-functional AI product discovery initiative, present prioritized opportunities with supporting data, and land interviews for JTBD analyst roles.

Practice Projects

Apply your skills with hands-on projects. Ordered by difficulty.

JTBD Research Sprint for a SaaS AI Feature

Beginner

Conduct 5-8 JTBD switch interviews with users of a popular SaaS tool (e.g., Notion, Slack, or Figma). Synthesize findings into job stories, map unmet needs, and write an AI opportunity brief recommending one AI feature grounded in your research.

~25h
Jobs-to-be-Done framework theory and interview methodologyUser research synthesisTechnical writing for AI product specifications

AI Capability Audit Against User Jobs

Intermediate

Select a product category (e.g., email clients, CRM, design tools). Map the top 5 user jobs, then evaluate current AI capabilities from OpenAI, Anthropic, and open-source models against each job. Build a capability-job fit matrix and identify the highest-value AI opportunity.

~30h
AI capability landscape assessmentJob mapping and job story creationPrioritization frameworks

LLM Prototype Evaluation Harness

Intermediate

Build a Python-based evaluation harness using LangChain and an LLM API that takes job stories as input and tests whether AI-generated outputs fulfill the stated job outcome. Include automated scoring, test case management, and results tracking.

~35h
Basic Python scripting for data extraction and LLM prompt testingAI evaluation metrics literacyWorkflow and process decomposition

Competitive AI Teardown Report

Beginner

Select two competing products with AI features (e.g., Grammarly vs. Writer, Copilot vs. Cursor). Conduct hands-on testing, map each product's target jobs, evaluate AI quality, and produce a detailed teardown report with strategic recommendations.

~20h
Competitive analysis of AI product strategiesStakeholder communicationAI capability landscape assessment

End-to-End AI Product Opportunity Portfolio

Advanced

Conduct a comprehensive JTBD research program for a real or simulated company: interview 15+ users, build job maps for 3 personas, score 10+ AI feature opportunities, prototype the top opportunity with an LLM, and present a prioritized roadmap with ROI projections to a mock executive audience.

~60h
Jobs-to-be-Done framework theory and interview methodologyPrioritization frameworksAI product strategy

Agentic Workflow Design from JTBD Analysis

Advanced

Identify a multi-step user job that could be fulfilled by an AI agent. Decompose the job into sub-tasks, design the agent architecture using LangChain, define guardrails and human-in-the-loop checkpoints, and build a working prototype that demonstrates autonomous job completion.

~45h
Workflow and process decompositionAI capability landscape assessmentBasic Python scripting for LLM prompt testing

Ready to Start Your Journey?

Prep for interviews alongside your learning — it reinforces every concept.

Practice Interview Questions Explore More Careers