investor-fit-research

Build investor and fund shortlists from Crawlora public PitchBook profiles and stored investor, fund, and company records. Use for fundraising research based on observed strategy, geography, stage clues, and portfolio evidence with incomplete public coverage explicitly preserved.

Access and scope

Set CRAWLORA_API_KEY to your Crawlora key. Prefer the listed Crawlora MCP tools when connected; otherwise use the bundled scripts/crawlora.sh, which sends x-api-key to https://api.crawlora.net/api/v1. Read reference/endpoints.md for exact methods and parameters. Check application code as well as HTTP status; payloads are inside data. Stop on authentication errors, back off on 429, and retry a transient upstream failure once. Bound requests by the requested scope and credit budget. Preserve source IDs, URLs, source dates, and retrieval/crawl times separately.

Build a source-attributed fundraising shortlist for a specified company/round. Establish business model, sector, geography, stage, round size, desired investor type, and explicit exclusions. A shortlist is research evidence for qualification, not confirmation of available capital or willingness to invest.

Discover firms, funds, and relevant portfolio clues

• Discover investor and fund filter values through their dataset facets. Search names/descriptions and exact returned geography/type/strategy/status values. Headquarters geography is not the same as an investment mandate. A full-text match on a sector/stage phrase is a lead, not a verified mandate filter.
• Resolve returned profile IDs and kind before item/live detail. Investors (manager firms), funds, and companies have different namespaces; keep each identity separate. Live profile routes require exactly one of id or url, even though those fields are individually optional in the schema.
• Read stored item records with their crawled_at, then refresh shortlisted investor/fund/company profiles only as needed. A missing stored item does not establish the firm is absent, closed, or unwilling to invest.
• Use public descriptions, overview fields, and visible investment/company rows as attributed evidence. PitchBook tables are public previews: total can exceed the number of visible rows. Blank check sizes, stages, dry powder, performance figures, or relationships stay unknown; do not reconstruct them from profile popularity or aggregate portfolio counts.
• Match company names/IDs and actual relationship rows before inferring a portfolio link. Similar firm names and fund-family branding are not proof that a particular fund made the investment. A past investment does not prove current focus, lead-investor appetite, exclusivity, or conflict policy.

sh

scripts/crawlora.sh /datasets/pitchbook-investors/facets facet=investor_type
scripts/crawlora.sh /datasets/pitchbook-funds/facets facet=fund_strategy
scripts/crawlora.sh /datasets/pitchbook-investors/search q="climate" page=1 page_size=10
# Resolve a returned investor/fund ID and kind before live detail:
# scripts/crawlora.sh /pitchbook/investor id="$INVESTOR_ID"

Qualify fit and present the shortlist

Build a criterion matrix with supported, contradicted, and unknown states for sector/business fit, geography, stage, size, strategy, and relationship clues. Show the source excerpt/field, date, and entity behind each state. Unknown criteria are not silently zero; a critical unknown keeps a candidate conditional. Disclose weights if a numeric ranking is requested. Do not rank by exits or portfolio counts as a proxy for fit or performance.

Return firm/fund identity, evidence-backed fit, relevant visible investments, public business contact channels if supplied, unresolved questions, and a bounded rejected/conditional list. A source's 'active' status or vintage year is not verified current fundraising, uncommitted capital, or an open application process. Preparing a shortlist does not authorise contacting investors, sending company information, registering for paid data, or making investment decisions.