opportunity-solution-tree
Acerca de
Esta habilidad construye un Árbol de Oportunidades y Soluciones para estructurar el descubrimiento de producto, mapeando un resultado deseado hacia oportunidades del cliente, soluciones potenciales y experimentos comprobables. Se basa en el marco de Teresa Torres y es ideal para cuando un equipo no está seguro de qué construir a continuación o necesita priorizar múltiples ideas en competencia. Úsalo para clarificar espacios de funcionalidades complejas antes de crear un PRD, pero no para historias bien definidas o correcciones de errores.
Instalación rápida
Claude Code
Recomendadonpx skills add avelikiy/great_cto -a claude-code/plugin add https://github.com/avelikiy/great_ctogit clone https://github.com/avelikiy/great_cto.git ~/.claude/skills/opportunity-solution-treeCopia y pega este comando en Claude Code para instalar esta habilidad
Documentación
Opportunity Solution Tree (OST)
Structures product discovery by connecting a desired outcome → customer opportunities → solutions → experiments. Prevents jumping to solutions before validating the problem space.
Based on Teresa Torres, Continuous Discovery Habits (2021).
The 4-level structure
┌─────────────────────┐
│ DESIRED OUTCOME │ ← single measurable metric
└──────────┬──────────┘
┌───────────────┼────────────────┐
┌──────┴─────┐ ┌──────┴─────┐ ┌──────┴─────┐
│Opportunity │ │Opportunity │ │Opportunity │ ← customer pain/need
│ A │ │ B │ │ C │
└──────┬─────┘ └──────┬─────┘ └────────────┘
┌──────┴───┐ ┌──────┴───┐
┌───┴──┐ ┌───┴──┐ ┌───┴──┐ ┌───┴──┐
│Sol 1 │ │Sol 2 │ │Sol 3 │ │Sol 4 │ ← possible solutions
└───┬──┘ └──────┘ └───┬──┘ └──────┘
┌────┴────┐ ┌───┴────┐
│ Exp 1 │ │ Exp 2 │ ← fast experiments
└─────────┘ └────────┘
Key principles:
- One desired outcome at a time — don't try to solve everything
- Opportunities are customer problems/needs, never solutions
- Generate ≥3 solutions per opportunity before choosing one
- Experiments are the cheapest way to validate an assumption
- The tree is a living document — update weekly as you learn
How to build an OST
Step 1 — Define the desired outcome
Confirm or help the user articulate one measurable outcome at the top of the tree.
Good outcomes:
- "Increase 7-day retention from 20% to 35%"
- "Reduce time-to-first-value from 3 days to 1 day"
- "Increase conversion from free to paid from 2% to 5%"
Bad outcomes (reject these):
- "Build a better onboarding" — that's a solution
- "Improve the product" — unmeasurable
- "Launch feature X" — that's an output
If the user can't state a metric: ask "What would need to be true for you to consider this effort a success?"
Step 2 — Map opportunities from research
From customer interviews, analytics, support tickets, or NPS feedback, identify 3–7 customer opportunities (pain points, unmet needs, desires).
Frame each from the customer's perspective:
- ✅ "I struggle to understand which plan is right for me"
- ✅ "I can't find past purchases quickly"
- ✅ "I feel anxious about whether my data is safe"
- ❌ "Users need a better search" — that's a solution
Prioritise using Opportunity Score (Dan Olsen, The Lean Product Playbook):
Opportunity Score = Importance × (1 − Satisfaction)
Survey customers: rate each need on Importance (0–1) and current Satisfaction (0–1).
- High Importance + Low Satisfaction = highest score = best opportunity
- Plot on Importance vs Satisfaction chart — upper-left quadrant is the sweet spot
Step 3 — Generate solutions (diverge before converging)
For each top-priority opportunity, brainstorm ≥3 solutions from three angles:
- PM perspective: What UX/product change addresses this?
- Designer perspective: What interaction or visual change?
- Engineer perspective: What technical approach? (often the most creative)
Rules:
- Don't commit to the first idea — compare and contrast
- "Best ideas often come from engineers" — include technical solutions
- Solutions should be independent (different solutions for the same opportunity)
Step 4 — Design experiments
For the most promising solutions, design 1–2 fast experiments:
| Experiment | Assumption tested | Method | Success metric | Effort |
|---|---|---|---|---|
| <experiment name> | <what belief this validates> | <A/B test / fake door / prototype / interview> | <metric + threshold> | <1d / 3d / 1w> |
Assumption categories (prioritise in this order):
- Value: Will users want this? (most important to test first)
- Usability: Can users figure it out?
- Feasibility: Can we build it?
- Viability: Does the business case work?
Cheap experiment types:
- Existing product: A/B test, fake door, prototype, user interview, data analysis
- New product: XYZ hypothesis ("At least X% of Y will do Z"), landing page, concierge MVP
Step 5 — Visualise and document
Write docs/discovery/OST-<outcome-slug>.md:
# Opportunity Solution Tree: <Outcome>
**Desired outcome**: <metric> from <current> to <target> by <date>
**Last updated**: <date>
## Opportunity map
| # | Opportunity | Importance | Satisfaction | Opportunity Score | Priority |
|---|------------|-----------|-------------|-------------------|---------|
| A | <customer need> | 0.8 | 0.3 | 0.56 | 1st |
| B | <customer need> | 0.7 | 0.6 | 0.28 | 3rd |
| C | <customer need> | 0.6 | 0.2 | 0.48 | 2nd |
## Solutions for top opportunities
### Opportunity A: <name>
| Solution | Description | Experiment |
|---------|-------------|-----------|
| Sol A1 | <description> | <experiment> |
| Sol A2 | <description> | <experiment> |
| Sol A3 | <description> | <experiment> |
## Active experiments
| Experiment | Assumption | Status | Result |
|-----------|-----------|--------|--------|
| <name> | <assumption> | Running / Done | <result or pending> |
## Learning log
- <date>: Discovered <insight> from <source>. Killed <solution> / promoted <opportunity>.
Integration with /prd
Once an opportunity is validated and a solution is chosen:
→ Run /prd with the validated opportunity as the problem statement
→ The OST's Opportunity Score data feeds directly into PRD §3 (Success Metrics) and §4 (Target Users)
Anti-patterns
❌ Opportunity = solution in disguise: "Users need a search bar" is a solution. "Users can't find past purchases" is an opportunity.
❌ Skipping divergence: Picking the first solution for each opportunity. Always generate ≥3 before choosing.
❌ Experiments that take >1 week: If it takes longer than a week to learn, it's not an experiment — it's a feature.
❌ Updating the tree once: OST is a continuous practice. Update weekly as you learn.
❌ Too many outcomes: One outcome per tree. If you have multiple outcomes, run multiple trees or pick the highest priority.
Repositorio GitHub
Habilidades relacionadas
content-collections
MetaEsta habilidad proporciona una configuración probada en producción para Content Collections, una herramienta centrada en TypeScript que transforma archivos Markdown/MDX en colecciones de datos con tipado seguro mediante validación Zod. Úsala al construir blogs, sitios de documentación o aplicaciones Vite + React con mucho contenido para garantizar seguridad de tipos y validación automática de contenido. Abarca todo, desde la configuración del plugin de Vite y compilación MDX hasta la optimización de despliegue y validación de esquemas.
polymarket
MetaEsta habilidad permite a los desarrolladores crear aplicaciones con la plataforma de mercados de predicción Polymarket, incluyendo la integración de API para operaciones y datos de mercado. También proporciona transmisión de datos en tiempo real a través de WebSocket para monitorear operaciones en vivo y actividad del mercado. Úsela para implementar estrategias de trading o crear herramientas que procesen actualizaciones de mercado en tiempo real.
creating-opencode-plugins
MetaEsta habilidad ayuda a los desarrolladores a crear complementos de OpenCode que se conectan a más de 25 tipos de eventos, como comandos, archivos y operaciones LSP. Proporciona la estructura del complemento, las especificaciones de la API de eventos y los patrones de implementación para módulos en JavaScript/TypeScript. Úsala cuando necesites interceptar, monitorear o extender el ciclo de vida del asistente de IA de OpenCode con lógica personalizada basada en eventos.
sglang
MetaSGLang es un framework de alto rendimiento para el servicio de LLM que se especializa en generación rápida y estructurada para JSON, expresiones regulares y flujos de trabajo de agentes utilizando su caché de prefijos RadixAttention. Ofrece una inferencia significativamente más rápida, especialmente para tareas con prefijos repetidos, lo que lo hace ideal para salidas complejas y estructuradas, y conversaciones multiturno. Elige SGLang sobre alternativas como vLLM cuando necesites decodificación restringida o estés construyendo aplicaciones con uso extensivo de prefijos compartidos.
