field string | type string | description string | allowed_or_example string |
|---|---|---|---|
question_id | string | Stable identifier composed from the dimension prefix and row number | BI-001 |
dimension | string | One of the 12 DDScore review dimensions | Business Idea; Offering; Team; Market; Competitors; Technology & IP; Scalability; Legal & Regulatory; Exit; Presentation; Financials; Fundability |
claim_type | string | The claim or decision-evidence topic tested by the question | market definition |
falsification_question | string | A question framed around evidence that could contradict the current claim | What result would make the stated benefit materially smaller? |
evidence_to_request | string | The minimum evidence fields an outside reader would request | baseline; window; sample; exclusions; raw result |
contradiction_signal | string | An observable signal that conflicts with the current claim | The result disappears when failed cases are included |
decision_state_if_unresolved | string | State retained until the question is answered | open question |
founder_repair_action | string | A bounded material or evidence repair action | Publish the baseline and exclusions beside the claim |
stage_note | string | How to interpret the question without treating early-stage evidence as late-stage evidence | At pre-seed, direct observations can be stronger than a broad survey |
Startup Investor Falsification Question Bank
A founder's materials are easier to review when each important claim has a question that could disprove it. This open CSV resource gives founders 48 outside-reader questions across the same 12 dimensions used in a structured private-company review.
Each row records the claim type, a falsification question, evidence to request, a contradiction signal, the state to retain when the question is unresolved, a founder repair action and a stage note. There are four questions for every dimension:
- Business Idea
- Offering
- Team
- Market
- Competitors
- Technology & IP
- Scalability
- Legal & Regulatory
- Exit
- Presentation
- Financials
- Fundability
Files
startup-investor-falsification-question-bank.csv: 48 reusable questions.data-dictionary.csv: field definitions and allowed values.LICENSE.md: CC BY 4.0 attribution terms.
Intended Use and Limits
Use the bank before an investor meeting, data-room review or material rewrite. Start with the exact sentence an outside reader will see. Pick the closest question. Record the evidence that answers it and keep contradictions visible.
These questions do not verify a claim, assess source quality, calculate a company score, predict funding, provide investment advice or replace founder judgement, investor judgement or full due diligence. An unanswered question is a review state, not proof that the underlying claim is false. The resource contains no real company or personal data.
DDScore Context
DDScore turns submitted private-company materials into a structured 0–100 Due Diligence Score and full written report across 12 dimensions. It checks relevant claims against current public sources, shows confidence and gaps and supports an outside first pass.
Disclosure: I work on DDScore at Playful Pixels Oy.
Method Note
The dimension names and product boundary come from the current DDScore Wiki. The 48 questions are original review prompts. They are not sampled from companies, investor decisions or private materials. Version 1.0 was reviewed on 2026-07-20.
License
CC BY 4.0. Attribute Playful Pixels Oy and link to this dataset card.
- Downloads last month
- 19