Business Idea Brainstorming

Overview

Help the user turn a raw business idea into a refined concept through natural collaborative dialogue. Explore the idea from multiple angles, challenge assumptions, and refine the value proposition before moving to structured intake.

Process

digraph brainstorming {
    "Understand the raw idea" [shape=box];
    "Explore the problem space" [shape=box];
    "Explore target audience" [shape=box];
    "Explore business model options" [shape=box];
    "Propose 2-3 angles" [shape=box];
    "User picks direction" [shape=diamond];
    "Summarize refined concept" [shape=box, style=filled, fillcolor="#e8f5e9"];
    "Invoke idea-intake" [shape=doublecircle];

    "Understand the raw idea" -> "Explore the problem space";
    "Explore the problem space" -> "Explore target audience";
    "Explore target audience" -> "Explore business model options";
    "Explore business model options" -> "Propose 2-3 angles";
    "Propose 2-3 angles" -> "User picks direction";
    "User picks direction" -> "Propose 2-3 angles" [label="none fit"];
    "User picks direction" -> "Summarize refined concept" [label="chosen"];
    "Summarize refined concept" -> "Invoke idea-intake";
}

The Process

Ask questions one at a time via AskUserQuestion. Prefer multiple-choice where possible.

Phase 1: Understand the Raw Idea

Start with one open-ended question:

• "Tell me your business idea in a few sentences — what is it and why does it matter?"

Listen. Don't jump to structure yet.

Phase 2: Explore the Problem Space

One question at a time:

• Is this a problem you've experienced personally?
• How are people solving this problem today? (Options: Manual workaround, Existing tools that don't fit well, They don't — they just live with it, Other)
• How painful is this problem? (Options: Mild inconvenience, Significant frustration, Costs real money/time, Mission-critical blocker)
• What triggers the moment someone realizes they need a solution?

Phase 3: Explore Target Audience

• Who has this problem the most? (Options: Individuals/consumers, Small businesses, Mid-market companies, Enterprises, Specific profession/role, Other)
• Can you describe your ideal first 10 customers?
• How do these people currently find solutions? (Options: Google search, Word of mouth, Industry publications, They don't look — they've given up, Other)

Phase 4: Explore Business Model Options

Based on what you've learned, propose 2-3 realistic business model options:

• Present each with one sentence explaining why it fits
• Include trade-offs (e.g. "Subscription gives recurring revenue but requires continuous value delivery")
• Ask which resonates most

Phase 5: Propose 2-3 Angles

Synthesize everything into 2-3 concrete product/positioning angles. For each angle:

• One-liner: What it is in one sentence
• Key bet: The main assumption that must be true
• Advantage: Why this angle could win
• Risk: The biggest concern

Present all angles and ask the user to pick one (or combine elements).

Phase 6: Summarize Refined Concept

Present a short summary of the refined idea:

## Refined Business Concept

**Idea:** [one sentence]
**Problem:** [what it solves]
**For:** [who]
**How it makes money:** [model]
**Key bet:** [the assumption that must hold]
**Unique angle:** [what makes this different]

Ask: "Does this capture what we want to validate? Anything to adjust before we start the formal intake?"

After Brainstorming

Once the user confirms the refined concept:

1. Invoke business-validator:idea-intake to begin structured data collection
2. The refined concept gives idea-intake a head start — many questions will already be partially answered

The terminal state is invoking idea-intake. Do NOT invoke any research skill directly from brainstorming.

Key Principles

• One question at a time — don't overwhelm
• Multiple choice preferred — faster for the user
• Challenge assumptions — "why do you think X?" is a valid question
• Stay curious, not critical — the goal is to refine, not to judge
• No research yet — this is pure dialogue, no WebSearch
• Keep it to 5-10 minutes — enough to refine, not so much that it delays the pipeline