← all hypotheses

Public Deal Claim Stress-Test for Micro-PE and Family Office Screening

ranked [S] filter 8.5/15 spread ±0.5 signals: 3 independent
What is this?
A buy-side screening product that ingests only public or buyer-owned materials about lower-mid-market deals: CIM excerpts already shared to the buyer, management presentations, lender decks, broker teasers, press releases, and market claims copied into a structured worksheet rather than full confidential memos. AE converts each investment-relevant claim into explicit commitment and evidence classes, then pressure-tests for premise-conclusion severing, concession laundering, cosmetic confidence, and temporal/transmission blindness. The output is not a rewritten sell-side memo; it is an internal red-team brief for the buyer: which claims are robust enough to underwrite, which require confirmatory diligence, and which should be excluded from IC discussion unless independently verified. The feedback loop is rebuilt around fast external signals: missed deadlines, revised guidance, withdrawn claims in later materials, lender pushback, exclusivity dropouts, and post-LOI repricing triggers captured from deal process updates and subsequent document versions. This uses AE's reality-grading without requiring sellers to upload sensitive internal work product or manually maintain a long-lag outcome log.
Why did we consider it?
AE has a credible, differentiated wedge in lower-mid-market buy-side screening because it turns seller-controlled deal claims into an auditable red-team brief with fast reality signals, while avoiding the confidentiality and workflow friction that usually blocks adoption.
What breaks?
  • Signal Opacity: LMM deal updates (lender pushback, broken LOIs) are private, unstructured, and highly relational, making automated <24h ingestion impossible.
  • Engine Starvation: Without fast, objective reality signals, the AE's core 508-prediction autopsy loop breaks down into a manual, high-lag data entry chore.
  • Buyer Redundancy: Discounting sell-side claims is the fundamental job of a PE associate; Micro-PEs won't pay £100k+ for automated skepticism.
What did we learn?
Still in evaluation (phase: ranked). No verdict yet.

Filter scores

Five axes, each scored 0-3. Three independent runs by different model perspectives. Median shown.

AxisWhat it measures
data moatDoes this product accumulate proprietary data that compounds?
10x model testDoes a better model make this more valuable, or redundant?
fast feedback loopsCan outputs be graded against reality in <30 days?
solo founder feasibleCan a solo operator build and run this without a team?
AI providers cant eat itDo hyperscalers have structural reasons NOT to build this?
Composite median: 8.5 / 15. Graduation threshold: 9.0. IQR across runs: 0.5.

Evidence

Signal A — Primary source

The framework combines data from several sources to ensure decisions are made holistically considering data related to private equity deals (Unquote data), finances (Fame) and management (Companies House).

Signal B — Competitor with documented gap

Existing AI tools like CIM parsers and deal screeners (e.g., WorkWise AI Deal Screener, V7 Go) ingest CIMs/teasers/pitch decks for extraction, scoring, and risk flagging but struggle with non-standard formats, cannot judge add-back legitimacy or qualitative claims (premise-conclusion severing, concession laundering), require human review for reality-grading, and lack outputs like internal red-team briefs categorizing claims as robust/verify/exclude using external signals.

Signal D — Demand proxy

{"found":true,"summary":"Reddit/HN discussions highlight pains in micro-PE/family office deal screening: poor CIM quality (overvalued, hidden owner involvement), difficulty accessing proprietary deal flow without networks/funds, manual first-pass CIM review time sink, challenges for small players in lower mid-market without brokers' pocket listings.","sources":["https://www.reddit.com/r/buyingabusiness/comments/1pzme50/micro_pe_mgt_group","https://www.reddit.com/r/private_equity/comments/1q3xz8k/preparation_1_month_prior_to_starting_as_a_junior","https://www.reddit.com/r/cofounderhunt/comments…

Evaluation history

WhenStagePhase
2026-04-22 20:40evidence_searchranked
2026-04-22 19:30evidence_searchranked
2026-04-22 18:40evidence_searchranked
2026-04-22 17:50evidence_searchranked
2026-04-22 17:00evidence_searchranked
2026-04-22 16:10evidence_searchranked
2026-04-22 15:20evidence_searchranked
2026-04-22 14:30evidence_searchranked
2026-04-22 13:40evidence_searchranked
2026-04-22 12:50evidence_searchranked
2026-04-22 12:00evidence_searchranked
2026-04-22 11:10evidence_searchranked
2026-04-22 10:20evidence_searchranked
2026-04-22 09:30evidence_searchranked
2026-04-22 08:40evidence_searchranked
2026-04-22 07:50evidence_searchranked
2026-04-22 07:00evidence_searchranked
2026-04-22 06:10evidence_searchranked
2026-04-22 05:20evidence_searchranked
2026-04-22 04:30evidence_searchranked
2026-04-22 03:40evidence_searchranked
2026-04-22 02:50evidence_searchranked
2026-04-21 11:30evidence_searchranked
2026-04-21 11:00evidence_searchranked
2026-04-21 10:30evidence_searchranked
2026-04-21 09:50evidence_searchranked
2026-04-21 09:10evidence_searchranked
2026-04-21 08:40evidence_searchranked
2026-04-21 08:10evidence_searchranked
2026-04-21 07:20evidence_searchranked
2026-04-21 06:50evidence_searchranked
2026-04-21 06:20evidence_searchranked
2026-04-21 06:00filter_scorescored
2026-04-21 05:50filter_scorescored
2026-04-21 05:40filter_scorescored
2026-04-21 05:30evidence_searchargument
2026-04-21 05:20audience_simulationargument
2026-04-21 05:10red_team_killargument
2026-04-21 05:00steelmanargument
2026-04-21 04:50genesisargument