I build sourcing maps for healthcare deal teams. The last few months it has been pain management in South Carolina, dental nationally, primary care in California, ambulatory infusion in Texas, and a handful of other specialty verticals. The pattern that holds across every one of those engagements is the same - and it is not about the tools, the sequence, or the headcount on the BD team.
It is about what gets built between the raw target list and the first phone call.
Jonathan Babcock at Compass Equity Group recently walked through his team’s sourcing process on the Deal Sourcery podcast. 700 conversations a year. 13-step outreach sequence. 80% direct calls.
He said something most people skipped past:
“From the onset of that 13-step process, it’s conveying credibility, domain knowledge, the depth of all the research that we’ve done – showing them it’s worth taking 15 minutes out of their day.”
Those tools give you a company name and a phone number. They do not give you credibility, domain knowledge, or depth of research. That part comes from work that happens before the first touch.
Most firms doing healthcare M&A sourcing skip that part.
A PE investor I spoke with recently described his firm’s sourcing setup. They pay a search firm to build target lists, get approval, and run the outreach.
His honest read on it:
“They find good targets. A lot of times they just aren’t able to get engagement out of those targets.”
His best guess on how they build the lists:
“I have no visibility into that. I’m sure they’re trolling CMS.”
That’s the problem in one sentence. CMS plus a target list plus outreach. No intelligence in between.
The firm isn’t broken. It found an add-on that’s closing this week. But the yield is low, the engagement is thin, and the fund can’t tell why because they can’t see the methodology. The SPS Origination Benchmark found the median PE firm covers just 17.6% of its target market deal flow. The other 82% never crosses the desk. That gap is not a tools problem.
The issue isn’t the process. It’s what the process is running on.
Healthcare is the hardest vertical for generic data tools. Four reasons:
The “company” is a medical practice. Grata gives you the practice name and maybe a LinkedIn headcount. It doesn’t give you the number of physicians in the practice, their specialties, partnership structure, or whether the founding partner has trained a successor.
The decision-maker is a physician. Not a professional operator. Physicians care about patient continuity, team preservation, clinical autonomy. The credibility hook that earns a 45-minute call isn’t “we’ve done 14 deals in your space.” It’s knowing the senior partner trained at Cleveland Clinic, built the practice from three physicians to eleven over 15 years, and has watched three PE-backed competitors open in the same geography in the last 12 months. That context isn’t in any subscription.
The valuation requires data no generic tool carries. Payer mix, collections per physician, Medicare vs. commercial concentration, ancillary revenue, RVU proxies. Without these you can’t estimate EBITDA. Without EBITDA, you can’t rank the list. Without ranking, you’re doing equal outreach to tier-one targets and dead ends.
The timing signals are healthcare-specific. The 61-year-old founding partner with no visible successor. A lease expiring in 14 months. Three PE-backed competitors entering the same market in one quarter. These signals exist and they matter. They don’t show up in a Grata filter.
Without the intelligence layer, here’s what actually happens in a Compass-style 13-touch process in healthcare:
Touch 1: cold. Touch 4: still cold. Most people quit at touch 4.
It’s not that 4 isn’t enough. It’s that the first 4 touches didn’t build enough credibility to earn a 5th. There was no signal that you know anything specific about this practice or this physician. No reason to stop and engage.
That’s what Babcock means by conveying credibility. He’s not sending 13 generic follow-ups. He’s sending industry reports, conference references, research materials. He does the validation and refinement work before the first touch. Each contact is informed.
You can’t do that from a CMS pull and an Apollo sequence.
Here is what it looks like on a real target.
I just delivered a sourcing map for a sponsored group going into South Carolina pain management. Fifty practices screened. Twenty in the universe. Nine Tier 1.