How we turned years of M&A podcast interviews into a cited knowledge base that dealmakers can query directly inside Claude, in the middle of a live deal.
Background
M&A Science is a Chicago-based community and education platform for mergers and acquisitions professionals, founded by Kison Patel. It is best known for the M&A Science Podcast, a deep library of long-form interviews with practitioners on diligence, deal structuring, earnouts, integration, and negotiation, years of first-hand expertise that most dealmakers never have time to fully absorb.
All of that knowledge lived in audio and long-form transcripts. There was no way for a dealmaker in the middle of live diligence to ask a specific question, like how practitioners structure earnouts for a founder-led services business, and get a direct, cited answer. Finding one meant knowing which of hundreds of episodes covered the topic, scrubbing through an hour of audio, and pulling the framework out by hand. The podcast was a powerful education asset, but it was not yet something a deal team could use in the room, and there was no bridge from that content into the AI tools dealmakers increasingly rely on day to day.
What We Did
We partnered with M&A Science as a co-marketing technology build, designing and building a custom Model Context Protocol (MCP) server, the M&A Science Playbook, that turns the podcast library into a structured, cited knowledge base queryable directly inside Claude.
We launched the MVP live at the M&A AI Summit on June 10, 2026, with M&A Science driving event promotion and Wyecliff providing the build and on-site support.
- Built a cited knowledge base by ingesting 147 podcast episodes and splitting them into 1,313 tagged transcript segments, each mapped back to its source episode, so every answer Claude gives traces to a specific citation.
- Mapped 25 M&A topics across the full deal lifecycle: financial, legal, operational, IT, and people diligence, deal structuring, LOI terms, earnouts, purchase price adjustments, Day 1 and Day 100 integration, and retention.
- Documented 38 practitioner disagreements so the system surfaces where experts genuinely disagree, on earnout metrics or valuation approach for example, instead of flattening hundreds of interviews into one manufactured answer.
- Shipped a set of Claude-native skills on top of the corpus: browse the topic library, generate a stage-aware deal briefing, build a due-diligence question bank, or work through earnout design options, all grounded in cited content.
Impact
The clearest proof came from a live user, not a lab test. At the Summit, John Palusci, VP of Strategic Finance at Schweiger Dermatology Group, used Claude connected to the M&A Science Playbook to run financial diligence on an active deal, then presented the results on stage.
- Financial diligence review that normally took about a week by hand came back in one to two hours, roughly 35x faster, with about 85% first-pass completion.
- Finding a topic across the library went from hours of scrubbing audio to a sourced answer in seconds.
- 318 queries were logged against the knowledge base within the first weeks of the MVP going live.
Financial diligence review in 1 to 2 hours instead of about a week
Tagged transcript segments from 147 podcast episodes, each cited
Practitioner tradeoffs surfaced instead of one generic answer
Logged against the knowledge base in the first weeks of the MVP
“We didn't just want to publish more content. We wanted the knowledge our practitioners have shared over hundreds of episodes to actually show up at the moment someone is sitting inside a live deal.”
Software: Claude, custom Model Context Protocol (MCP) server
Similar Case Studies
More work from similar industries, or other recent stories from our library.
Structured Copilot enablement for ~2,500 service staff across two license tiers
How we brought structured Copilot training to roughly 2,500 service staff across two license tiers, then layered discovery, a power-user cohort, and train-the-trainer to make it stick.
Secure AI contract and spec review that gives 200+ hours back to estimating
How we built a dedicated platform that turns 50-page GC subcontracts into prioritized redlines in minutes, on the company's own secure infrastructure.
Copilot discovery and summit enablement for 500+ insurance producers
How we shadowed four IMA producers, turned their real workflows into live demos, and got a room full of producers asking to start.
Tell us your biggest problem.
We'll show you the ROI.
Drop the problem in the box. A Wyecliff partner replies inside one business day with two ideas you can ship in 30 days. No pitch deck, no sales call required.