Case Study
Why a Growth Equity Fund Replaced PitchBook and Harmonic with Crustdata to Build Their Deal Sourcing Tool
Company
A Growth Equity Fund
use case
Deal Sourcing Tool
company overview
This growth equity fund manages over $2B in AUM and invests in B2B software companies from pre-seed through growth stage. Their investment team focuses on identifying technical founders at the earliest possible stage, often before the founder has incorporated a company or appeared in any database.
scale
$2B+ AUM, 40+ investment professionals
The fund's sourcing edge depends on reaching founders before competitive processes begin. But their entire workflow was manual, expensive, and reactive. Associates spent weeks reviewing profiles school by school, while the databases they paid for only surfaced founders after the window to build a relationship had already closed.
The fund targets technical founders, often PhDs or researchers commercializing their work out of top universities. Finding them required brute-force manual effort.
Associates reviewed alumni networks one school at a time, working through Cambridge, Imperial, ETH Zurich, and TUM in sequential batches of hundreds of profiles each
Each batch took days to review, filter, and cross-reference against the fund's existing pipeline, with no way to automate the screening or avoid reviewing the same profiles repeatedly
The process was too slow to run across more than a handful of universities at once, which meant the fund's geographic coverage was limited by how many schools associates could manually work through in a quarter
The fund invests at pre-seed and seed, where founders often have no company record in any database yet.
PitchBook and Crunchbase track companies that already have press mentions, funding records, or established profiles. By the time a founder appears in these databases, multiple firms have already reached out
The signals that matter at the earliest stage, a researcher leaving their university position, a technical lead quitting a large company, a profile updated to "building something new," happen weeks or months before any traditional database picks them up
The fund was consistently arriving at founders after the competitive dynamic had already formed, reducing their ability to build relationships on their own terms and secure favorable allocation
The fund paid for PitchBook, Harmonic, and additional enrichment providers, but none of them supported the internal tooling the team wanted to build.
PitchBook charges per seat with no API access, making it impossible to feed company or founder data into the fund's internal deal tracking systems programmatically
The fund's technology team wanted to build automated sourcing pipelines, but every vendor they used required manual export, CSV upload, and re-entry into internal tools
Paying $20K per seat for a tool that required manual workflows on top of it meant the fund was spending on data access and still spending associate time on data entry
The fund knew what signals predicted a founder was about to start a company. They had no infrastructure to detect those signals at scale.
Profile changes, job departures, stealth company formation, and first key hires all happen on professional profiles and social media before they surface in any structured database
Associates monitored these signals manually by checking profiles one at a time, which meant they only caught signals for founders they already knew about, not founders they had yet to discover
The fund had backed companies in the past by spotting that a researcher had quit their job and asking if they were starting something. They wanted to systematize that intuition, but no vendor they used could deliver real-time profile change alerts at the scale they needed
Crustdata gave the fund the data infrastructure to replace manual sourcing with automated, thesis-driven pipelines that run across every target university and geography simultaneously.
The fund now runs queries like "PhD founders from top-10 European universities who changed their profile in the last 30 days and list stealth or building something new" using Crustdata's People Search API with 60+ filters and nested boolean logic
What previously took associates weeks of university-by-university review now runs as a single API call that returns structured results across every target school, geography, and founder archetype at once
The Company Search API with 95+ filters handles the company-side screening, letting the fund identify early-stage companies by headcount growth, funding stage, geography, and industry without relying on PitchBook or Crunchbase for discovery
When the fund identifies a potential founder, the enrichment response reflects who that person is today, not a quarterly database snapshot.
Crustdata's People Enrichment API returns 90+ datapoints per request, including current role, work history, education, skills, and profile metadata, sourced live rather than pulled from a pre-built cache
Associates no longer cross-reference multiple sources to verify whether a founder has actually left their previous role or updated their focus area. The API response answers those questions in a single call
The direct request-response model, with no ticket systems or multi-step retrieval, means the fund's internal tools can enrich a founder profile programmatically as part of an automated pipeline rather than requiring manual lookup
The fund set up automated monitoring through Crustdata's Watcher API to catch the earliest founder signals without manual checking.
Webhook-based alerts fire when target profiles change: a researcher quits their university position, a technical lead updates their title to "co-founder," or a stealth company posts its first job opening
Every Monday, associates receive a curated list of people across all target universities and companies who triggered a signal in the past week, delivered directly into the fund's internal systems
The alert logic is fully customizable. The fund weights signals differently by sector, geography, and founder archetype, creating a detection system that reflects their specific investment thesis rather than a generic platform's definition of what matters
The fund's deal sourcing shifted from manual and reactive to automated and predictive, with measurable impact across associate time, data spend, and deal flow quality.
University-by-university review that consumed weeks of associate time now runs as automated queries across every target school and geography simultaneously, freeing associates to spend their time on relationship building and due diligence rather than profile review
The fund expanded their geographic coverage from a handful of universities per quarter to continuous monitoring across 20+ institutions, without adding headcount
Associates now spend the first hour of Monday mornings reviewing a curated signal feed rather than spending entire weeks assembling it manually
The fund eliminated their PitchBook subscription and consolidated from three data vendors into one API-first provider, removing $20K/seat licensing costs that came with no programmatic access
One API now handles company discovery, founder search, enrichment, and monitoring, which means the technology team maintains one integration instead of managing manual exports across multiple platforms
The savings funded the development of the fund's internal deal sourcing tool, turning a cost center (data subscriptions) into an investment in proprietary infrastructure that compounds over time
The fund now identifies founders at the earliest detectable signals: profile changes, job departures, stealth company formation, and first key hires, weeks before these founders appear in PitchBook or receive inbound from other investors
Building relationships at the stealth stage, before the founder has even incorporated, leads to stronger partnerships, better terms, and higher allocation when the founder eventually raises
The sourcing advantage compounds over time as the fund's automated pipelines cover more universities, more geographies, and more founder archetypes than any manual process could sustain