Official methodology · facts checked 2026-08-30

How does OpenSeed review a business plan?

OpenSeed does not ghostwrite a plan or issue an acceptance verdict. It compares claims with evidence across multiple perspectives and turns the findings into revision priorities and next validation actions for the founder or team.

1. Confirm the input boundary

Reviews can start from pasted text, a guided form, or DOCX, Markdown, and TXT files. A text-based PDF may be available when signed in and enabled. Scanned documents and unsupported formats have separate limitations. The latest program rules and investor requirements must always be checked against the original source by the user.

2. Separate claims from evidence

Statements such as “the market is large,” “customers want this,” or “the team can execute” are separated from figures, sources, experiments, and experience. The review then checks whether numbers and assumptions agree across sections without losing the location of the supporting evidence.

3. Cross-check multiple evaluation roles

The current product roster contains 17 evaluation roles: 7 core roles and 10 professional roles. Professional roles apply according to relevance. The Chief report composer and conditional global reviewer are separate and are not included in this count.

  • Problem and customer — Checks whether the target problem, customer assumption, and supporting evidence connect.
  • Market and product — Checks market sizing, competition, differentiation, and the value the product actually delivers.
  • Team and execution — Checks roles, relevant capability, milestones, resources, and the validation plan.
  • Finance and revenue — Checks whether price, cost, revenue, and cash-flow assumptions reconcile with their evidence.
  • Policy and risk — Checks program or investor requirements, operational risks, and questions that remain unanswered.

4. Turn findings into revision priorities

  1. Mark gaps and questions — Separate strengths, unsupported claims, risk signals, and follow-up questions.
  2. Prioritize revisions — Turn the findings into the revisions and next validation actions worth doing before submission.

The output is reference feedback. See the sample report, then revise the document and decide on the final submission yourself.

Start by use case

Limits of the methodology

AI analysis is not legal, tax, accounting, investment, or other professional advice. It does not guarantee program acceptance, investment, funding, loan approval, or revenue. The user remains responsible for factual accuracy, the final submission, and the decision.