Real Problems Are Everywhere.
The Evidence Is Fragmented.
People describe real pain, failure, workarounds and unmet needs every day across forums, reviews, communities and support discussions.
The problem is not the absence of evidence. The problem is turning scattered evidence into structured, traceable problems worth investigating.
PDE does not start with ideas. It starts with evidence.
Why Discovering Real Problems Is Hard
The internet contains evidence. It does not automatically contain structured problem intelligence.
Who We Serve
Founders & Independent Builders
Understand what problem may be worth validating before building.
Product Teams
Turn fragmented feedback into repeated problem patterns.
Researchers & Market Analysts
Structure emerging evidence into traceable problem intelligence.
Operators & Business Owners
Understand where real workflows and existing solutions are failing.
What PDE Does
Every conclusion follows the same chain. No step may be skipped.
PDE also exposes unknowns — what is not yet proven matters as much as what is supported.
What PDE Does — and Does Not Do
We Do
- Collect traceable human evidence
- Structure problem signals from real sources
- Group repeated patterns across sources
- Define atomic problems with actor, context, desired result, blocking step
- Separate problem existence from commercial proof
- Expose unknowns and missing proof
- Allow opportunity reasoning only after evidence gates
We Don't
- Generate startup ideas from nothing
- Guarantee market size
- Guarantee willingness to pay
- Guarantee product-market fit
- Predict winners
- Replace customer interviews
- Turn weak evidence into certainty
- Create fake commercial proof
- Hide unknowns
What We Believe
Evidence First
Every conclusion starts with real human evidence — not ideas, not prompts, not AI speculation.
Problem First
Structure the problem before discussing solutions. A clear problem is already half-way to a decision.
Product Last
Products come after evidence, signals, clusters and commercial proof — never before.
Unknown Is Valid
A null field is an honest "the evidence does not state this" — not a zero, not a guess.
Traceability Matters
Every conclusion must trace back to its source. If it cannot, it is an assertion — not a finding.
Problem ≠ Opportunity
A problem can be real and well-documented without being commercially attractive. These are separate gates.
AI Must Stay Bounded
AI can structure, classify and summarize evidence. It cannot invent evidence, fabricate demand, or create opportunities.
Why Traceability Matters
A Problem must trace back to its source. If a conclusion cannot be traced to evidence, it should not be treated as fact.
Where AI Fits
AI can help
- Structure evidence
- Classify source types
- Summarize bounded evidence
- Group related signals
AI cannot
- Invent evidence
- Fabricate demand
- Manufacture certainty
- Create unsupported opportunities
PDE vs Idea Generators
Idea Generator
- Starts with
- Prompts
- Produces
- Possible ideas
- Weakness
- May sound plausible without proving a real problem
PDE
- Starts with
- Real evidence
- Produces
- Structured problems and validation status
- Strength
- Shows what is supported and what remains unknown
PDE is not trying to be more creative. It is trying to be more grounded.
Current Stage
PDE is currently in an early public research stage. The methodology and research infrastructure are being built before scale.
The public site currently contains 1 structured problem with 3 signals and 3 evidence records. More will be added as the evidence base grows — not before.
Why Start Small?
PDE prefers precision over coverage. A smaller evidence base with full traceability is more useful than a large database of weak or invented opportunities. Every problem on PDE is built to be examined — not browsed.
See How PDE Works in Practice
Metered Automation Cost Predictability
Small operators on per-task-metered automation platforms face unpredictable costs, invisible consumption, and fail-closed pauses that halt production even after paid upgrades.
View the problem →