Opportunity Mapping: Deep Guide to the Opportunity Solution Tree (OST)

Do not jump directly to solutions without solving the problem. Opportunity Mapping is a visual thinking tool that forces you to explore the "problem space" first before entering the "solution space."

Core Framework: OST (Opportunity Solution Tree)

Opportunity Mapping structures your thinking into four levels:

1. Outcome: The business goal we want to achieve (derived from north-star-metric).
2. Opportunities: User pain points, needs, or desires validated through customer-discovery.
3. Solutions: Specific features or ideas to address the above opportunities.
4. Experiments: Lean tests to validate solution hypotheses (derived from lean-experimentation).

Checklist

1. Set Outcome — Define a single measurable business metric. Do not pursue multiple unrelated outcomes simultaneously.
2. Map Opportunity Space — Based on factual evidence from research. Classify scattered user feedback into a hierarchical opportunity tree.
3. Generate & Filter Solutions — Propose 2-3 mutually exclusive alternative solutions for each selected opportunity, avoiding the "one-choice" trap.
4. Identify Assumptions — Before implementing a solution, list all desirability, feasibility, and viability assumptions that must be met for it to work.
5. Design Experiments — Design lean experiments for high-risk assumptions.

Process Flow

graph ost_flow {
    node [shape=box, fontname="Arial"];
    Outcome [label="Desired Outcome", style=filled, fillcolor=lightblue];
    Opp1 [label="Opportunity A (Evidence-based)"];
    Opp2 [label="Opportunity B (Evidence-based)"];
    Sol1 [label="Solution A.1"];
    Sol2 [label="Solution A.2"];
    Exp1 [label="Experiment A.1.1"];

    Outcome -- Opp1;
    Outcome -- Opp2;
    Opp1 -- Sol1;
    Opp1 -- Sol2;
    Sol1 -- Exp1;
}

Expert Principles

1. Opportunity vs. Solution

• Opportunity: "It's hard for me to find the right size for me." (✅)
• Solution: "Build an AI size recommendation plugin." (❌) -> Always describe the problem at the opportunity level, not the feature.

2. Continuous Discovery

An Opportunity Map is not a one-time document; it should grow and be restructured with every customer-discovery interview.

3. Prioritization

Do not try to solve all opportunities on the tree. Choose your current focus based on "potential contribution to the Outcome" and "sufficiency of evidence."

Anti-Patterns

• Reverse Engineering: Inventing an "opportunity" just to justify a feature you've already decided to build. (❌)
• Island Branches: A solution on the tree that cannot be traced back to any parent opportunity, or an opportunity that cannot connect to the final result. (❌)
• Lack of Depth: Opportunity descriptions that are too broad (e.g., "improve experience"), making it impossible to derive specific solutions.

Delivery Standards

The structure of the Opportunity Map is recommended to be recorded via Mermaid or Dot diagrams in docs/pmpowers/specs/opportunity-map.md.