Use a Claude Project as a working notebook for a new physical business, turning questions into versioned planning artifacts while keeping financial, legal, spatial, and customer assumptions tied to evidence and expert review.
Source episode · 06:00
How I AI: Building a Real-World Business with Andrew Mason & Nabil Hyattwith Claire Vo
Nabeel and Andrew use a Claude Project as a central business workspace, turn simple questions into mission, budget, pricing, permit, landlord, competitive, and persona documents, then feed reviewed artifacts back into the project as context for later decisions.
A versioned business plan with mission, customer hypotheses, operating model, budget, scenarios, risks, permits, decisions, and open questions, plus source links and named owners for validation.
Anthropic AI assistant
Step by step
Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.
4 steps
Create one Claude Project for the business with a short canonical brief, source library, decision log, current assumptions, and an archive for superseded work. Minimize personal, contractual, and financial data in the model workspace.
Ask simple planning questions, but require the answer to become a structured artifact with sources, assumptions, options, numbers to validate, and an owner. Review each artifact before adding it to canonical project context.
Help plan [business] in [location] for [customer and goal]. Produce a draft [mission, operating plan, budget, pricing plan, timeline, or landlord deck]. Separate sourced facts, calculations, assumptions, options, risks, and open questions. Cite current primary sources. End with decisions required and the person or expert who should validate each one.
Use Claude to extract obligations and questions from permits, ordinances, leases, vendor terms, and insurance documents, not to deliver a final legal conclusion. Compare its output with the current document and a qualified professional.
Analyze this document for planning purposes. Extract exact clauses or sections related to [topic], dates, fees, obligations, ambiguities, and questions for [lawyer, accountant, agency, insurer, or landlord]. Quote only brief necessary excerpts with page references. Do not provide a final legal conclusion.
Develop customer segments around observable needs and behaviors, then connect each segment to an offer, event, price hypothesis, capacity assumption, and research plan. Replace invented proportions with interview, signup, and usage data as it arrives.
Propose customer hypotheses for [business] using these local sources and observations: [evidence]. For each, state the job, current alternative, likely trigger, barriers, offer hypothesis, price hypothesis, and evidence needed. Do not invent market share or demographic facts. Design a small test for each segment.
Turn an idea into a PRD, user stories, and a plan.
After the steps
How to recover when the loop fails and where human judgment helps.
Recover
Keep building
Design a weekly meta-workflow that analyzes your interactions with your AI assistant to make it smarter. This system learns your writing style from edits, suggests new automations for repetitive tasks, and helps you filter valuable AI techniques from hype.
Create a daily AI workflow that summarizes yesterday's meetings, identifies urgent tasks, and proactively learns internal company terms it doesn't understand. This keeps your AI assistant's knowledge current and highly relevant to your work.
Build a recurring AI task that automatically gathers updates from your calendar, Slack, and notes to suggest weekly priorities and prepare you for upcoming meetings. Start every Monday with a clear, comprehensive plan without the manual effort.
Join 100,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.