Build and maintain Opportunity Solution Trees — visual artifacts that map outcomes to opportunities to solutions to assumption tests. Use when creating a new OST, updating an existing tree, validating tree structure, or organizing discovery work. The OST is a living document that evolves as teams conduct interviews and run experiments. Triggers when user asks to 'create an OST', 'build opportunity solution tree', 'update the tree', or 'organize our opportunities'.
Create and maintain Opportunity Solution Trees that guide continuous discovery and product decisions.
A visual tree with four layers:
Shift teams from being told "build these features" to figuring out "what should we build to achieve this outcome?"
The tree adds structure to an otherwise messy, wide-open problem.
[OUTCOME]
|
+--------------+--------------+
| | |
[OPPORTUNITY] [OPPORTUNITY] [OPPORTUNITY]
| | |
+---+---+ +---+---+ +---+---+
| | | | | | | | |
[S] [S] [S] [S] [S] [S] [S] [S] [S]
| | |
[TEST] [TEST] [TEST]
Key principles:
The root of your tree. A single, measurable business/product goal.
[Outcome]: [Metric] from [current] to [target] by [date]
Why it matters: [Business impact]
How we'll measure: [Specific metric definition]
Customer needs organized by experience map phases.
Group opportunities by the customer journey:
Example: Project Management Tool
Getting Started Phase
Daily Planning Phase
Collaboration Phase
Progress Tracking Phase
Opportunities should decompose vertically:
[Big Opportunity]: "I can't manage my time across calendars"
|
+-- [Medium]: "I can't see all my meetings in one place"
| |
| +-- [Small]: "I have to check 3 different apps"
|
+-- [Medium]: "I miss meetings because they're not synced"
|
+-- [Small]: "My phone doesn't alert me for work calendar events"
Each level should be:
Ideas to address opportunities. MULTIPLE solutions per opportunity.
Compare and contrast to:
Minimum: 3 solutions per opportunity you're pursuing
For opportunity: "I can't see all my tasks across projects in one view"
Solution 1 (Aggregation): "All Tasks" view that shows every task from every project in a master list
Solution 2 (Smart filtering): Dynamic filters that remember your most common task queries
Solution 3 (Dashboard): Customizable dashboard with task widgets you can arrange
Solution 4 (Search-first): Powerful search that surfaces tasks as you type your intent
[Solution]: [Brief description of the idea]
How it addresses the opportunity: [Connection to customer need]
Assumptions:
- [Key assumption 1]
- [Key assumption 2]
- [Key assumption 3]
Status: [Idea | Testing | Validated | Invalidated | Shipped]
Experiments to validate (or invalidate) solutions before building them.
Every solution has underlying assumptions. Make them explicit.
Solution: "Smart filtering that remembers common queries"
Assumptions:
For each critical assumption, design a small test.
Assumption: "Users will notice and use auto-suggested filters"
Test Options:
Pick the test that:
[Assumption]: [What needs to be true]
Riskiness: [High | Medium | Low]
Test method: [Specific test type]
Success criteria: [What would prove it true]
Failure criteria: [What would prove it false]
Timeline: [How long to run]
Result: [Pending | Validated | Invalidated]
Learning: [What we discovered]
Work with stakeholders to clarify:
Create an experience map for the customer journey:
Use interviews and observation to validate this map.
From customer interviews, extract opportunities and organize by experience map phases.
Start with 20-30 opportunities to show breadth of the problem space.
You can't pursue all opportunities at once. Select 1-3 to focus on based on:
For each target opportunity, generate at least 3 diverse solutions.
For each solution you're considering, list all assumptions that must be true for it to work.
For the riskiest assumptions, design small tests you can run this week.
Execute tests, gather data, update the tree with learnings.
The OST is a living document. Update it continuously.
After customer interviews:
After assumption tests:
During team sync:
Keep the tree readable:
The #1 failure: 98% of people make this mistake.
Fix: Use the opportunity-framing skill to validate every opportunity statement.
Opportunities remain broad and vague.
Fix: Decompose vertically. Go at least 3 levels deep in your highest-priority opportunity.
Eliminates ability to compare.
Fix: Force yourself to generate at least 3 solutions before picking one.
Tree becomes a planning document, not a discovery tool.
Fix: Always have 5-10 active tests running. If not, you're not doing continuous discovery.
Build the tree once, then "start building."
Fix: Discovery and delivery happen in parallel. Update the tree weekly forever.
Opportunities based on assumptions, not customer evidence.
Fix: Conduct weekly customer interviews. Every opportunity should cite an interview.
Tree has 200 branches and no one can parse it.
Fix: Focus on current area of exploration. Archive or collapse the rest.
graph TD
A[Outcome: Increase trial conversion 12% to 18%]
A --> B[I can't see value in first session]
A --> C[I get stuck during setup]
B --> D[Solution: Interactive demo]
B --> E[Solution: Quick-win checklist]
B --> F[Solution: Template library]
D --> G[Test: Fake door test]
# Outcome: Increase trial conversion from 12% to 18%
## Opportunity: I can't see value in first session
Evidence: "I signed up but didn't know what to do first. I closed the tab." - Interview #12
### Solution: Interactive demo
Assumptions:
- Users learn better by doing than watching
- Demo can be completed in <5 minutes
- Demo shows most valuable features
#### Test: Prototype with 10 users
Status: Running
Success criteria: 8/10 complete demo and continue to product
### Solution: Quick-win checklist
[Continue...]
| Phase | Opportunity | Confidence | Solutions | Tests Active | |-------|-------------|------------|-----------|--------------| | Setup | I don't know where to start | High (12 interviews) | 3 | 2 | | Planning | I can't see all my tasks | Medium (5 interviews) | 4 | 1 |
Most important: Pick what your team will actually use and update.
OST work is done by the product trio:
All three should contribute to the tree.
Use the tree to communicate:
Much better than feature lists.
You know your OST practice is working when:
Go interview 5 customers. Opportunities will emerge.
Use the opportunity-framing skill on each statement. Ask "What problem does that solve?"
Pick the one with strongest customer evidence AND highest impact on outcome.
Interview different customer segments. Or different journey phases. Or ask different questions.
Archive opportunities you're not pursuing. Focus on current exploration area.
Show how the tree connects to outcomes they care about. Translate opportunities to business impact.
You don't have time NOT to. Discovery prevents building the wrong thing (massive time waste).
Google Workspace CLI for Gmail, Calendar, Drive, Contacts, Sheets, and Docs.
Manage Apple Notes via the `memo` CLI on macOS (create, view, edit, delete, search, move, and export notes). Use when a user asks OpenClaw to add a note, list notes, search notes, or manage note folders.
Work with Obsidian vaults (plain Markdown notes) and automate via obsidian-cli.
Use when you need to control Slack from OpenClaw via the slack tool, including reacting to messages or pinning/unpinning items in Slack channels or DMs.
Manage Apple Reminders via remindctl CLI (list, add, edit, complete, delete). Supports lists, date filters, and JSON/plain output.
Manage Trello boards, lists, and cards via the Trello REST API.
Category:productivity