CHAPTERS
- 0:00 – 5:27
Epic Systems in a nutshell: the quiet giant behind MyChart
Ben and David set up why Epic matters: it runs a huge share of US healthcare workflows while staying private, avoiding marketing, acquisitions, and remote work. They preview Epic’s oddities (Verona campus, no discounts, extreme customer retention) and position Judy Faulkner as a once-in-a-generation founder.
- •Epic’s unusual operating rules: no marketing/sales motion, no discounts, no acquisitions, no VC
- •MyChart’s ubiquity and Epic’s dominance among major hospital systems
- •47 years, essentially no customer churn
- •Judy Faulkner’s private-control structure and long-term independence
- 5:27 – 15:32
Judy Faulkner’s origins: math genius, early programming, and Wisconsin’s ‘Computers in Medicine’
The story begins with Judy’s upbringing, her parents’ influence, and her early love of math. A summer job learning Fortran and the emergence of early computer science programs steer her to a PhD track at the University of Wisconsin, where she encounters one of the earliest ‘Computers in Medicine’ courses.
- •Family background: entrepreneurship and civic/peace advocacy (Nobel Peace Prize connection)
- •Self-taught Fortran and early proof of exceptional programming talent
- •Sliding-doors moment: choosing Wisconsin over Stanford
- •Dr. Warner Slack and the earliest academic push to use computers in medicine
- 15:32 – 28:51
Why medical records became a software problem: paper chaos, standards, and Medicare/Medicaid
They step back to explain what medical records were historically and why standardization attempts struggled in a paper world. The creation of Medicare and Medicaid introduces massive documentation needs and drives the economic incentive to systematize records—especially for billing and auditability.
- •Early standardization attempts (MGH, AMA/AHA) and the limits of paper
- •Doctors’ autonomy vs standardized documentation
- •Medicare/Medicaid and the rise of third-party payers requiring proof and coding
- •The billing/documentation feedback loop that makes EHRs existential
- 28:51 – 32:57
The first EHR era: COSTAR, MUMPS, and MEDITECH’s foundational influence
At MGH, COSTAR becomes an early computerized record system, and the constraints of 1960s computing force the invention of MUMPS. Neil Pappalardo spins MEDITECH out of this work, creating an early commercial EMR leader—and indirectly seeding key technical and cultural ideas that Epic later adopts.
- •COSTAR’s early scope: scheduling, clinical data, and billing
- •Why MUMPS mattered: integrated database + multi-user concurrency
- •Neil Pappalardo founding MEDITECH (still a top EMR player)
- •Epic’s later dependence on MUMPS descendants (Caché/InterSystems stack)
- 32:57 – 37:07
Chronicles is born: Judy’s ‘single database’ epiphany and Epic’s core architecture
Back in Wisconsin, Judy learns MUMPS and has the breakthrough that becomes Chronicles—a unified patient-centric database. Epic’s enduring differentiation emerges: every application (clinical, billing, specialty modules) pulls from the same core data model.
- •Longitudinal patient record as the ‘holy grail’ across departments
- •Chronicles as Epic’s single-database backbone
- •Terminals/green-screen era: database + department-built screens
- •Integrated architecture becomes the seed of later reliability and adoption
- 37:07 – 51:28
Founding Human Services Computing (1979): tiny capital, no VC, and the MEDITECH playbook
Demand spreads via physician word-of-mouth, pushing Judy to start a company almost reluctantly. Epic’s origin economics are striking: a modest bank loan and a small friends-and-family round fund the first computer, while MEDITECH’s founder mentors Judy in building standardized, developer-led operations.
- •1979 founding; early customers arrive through viral academic referrals
- •$70k loan + ~$70k equity: the only primary capital ever raised
- •Business built like a ‘software developer’s company’ (manuals, standards)
- •College recruiting and internal training become foundational
- 51:28 – 1:02:31
From Epic’s name to real traction: Resolute billing + EpicCare GUI change the game
Epic remains small for years until it attacks the true hospital pain: billing. Resolute (1987) and then the PC-era shift enable EpicCare’s GUI (1992), making the product usable broadly and tying clinical workflows directly to revenue cycle—creating a compelling, integrated suite.
- •Why early growth was slow: infrastructure wasn’t ready and billing wasn’t covered
- •Resolute launches the revenue-cycle wedge on the same database
- •EpicCare (GUI) brings EMR usability to PCs; initial focus on ambulatory
- •Integrated clinical + scheduling + billing becomes the ‘administrator’s dream’
- 1:02:31 – 1:10:28
MyChart arrives shockingly early (2000): the patient portal that goes viral
Epic pioneers web-based access with MyChart, originally from Epic Web remote physician access. MyChart becomes a huge behavioral shift: patient record access, family care management, self-scheduling, fewer no-shows, and massive consumer-side pull that strengthens Epic’s market position.
- •Origin: Sumit Rama’s challenge project from Epic Web
- •MyChart launches in 2000—far earlier than expected for HIPAA-grade web apps
- •Patient convenience and family management drive adoption
- •Operational benefits for hospitals: self-scheduling, waitlists, reduced no-shows
- 1:10:28 – 1:33:12
Kaiser Permanente’s RFP (2003): the deal that makes Epic the gold standard
Kaiser’s system-wide ‘nervous system’ replacement becomes a defining contest versus Cerner. Epic wins by insisting on a single integrated system, proving scalability via rigorous modeling, and refusing equity demands—turning a small vendor into the category’s most trusted choice.
- •Kaiser’s scale and why EMR is really the enterprise operating system
- •Epic vs Cerner: integrated architecture vs conglomerated suite complexity
- •Excel transaction-flow modeling as an 11th-hour credibility win
- •Refusing warrants/equity and still winning cements Epic’s negotiating posture
- 1:33:12 – 1:52:15
The Verona campus and the ‘software factory’ culture: whimsy + ruthless execution
Epic scales hiring and builds a massive, themed campus inspired by Microsoft’s Redmond but with Disney-like design. They codify values in ‘Ten Commandments’ and institutionalize intense training, immersion in clinical settings, fast bug-fixing, and customer-centric implementation discipline.
- •Verona farmland campus built to attract elite grads and keep them in-person
- •Ten Commandments: independence, quality, commitments, frugality
- •Process discipline: developers fix their own bugs; minimize time-to-test
- •Implementation as high-stakes logistics; flat org, few titles, strong internal standards
- 1:52:15 – 2:12:21
Customer obsession as strategy: dedicated support teams, ‘BFFs,’ and standardization leverage
Epic’s go-to-market is inverted: almost no outbound sales, heavy post-sale support, and customer-driven roadmaps. Dedicated technical teams per module and a single ‘BFF’ per customer create deep lock-in, while Epic’s opinionated standardization reduces implementation risk and compounds interoperability advantages within the Epic network.
- •Minimal sales function; teams are developers, implementers, and tech specialists
- •Per-customer, per-product support teams + one BFF accountability owner
- •Customer grading/report cards and benchmarking to drive success
- •Opinionated ‘standard build’ philosophy and the ability to say no to misfit customers
- 2:12:21 – 2:34:27
HITECH & Meaningful Use (2009): stimulus that digitizes healthcare—and creates backlash
The Great Recession fuels massive incentives for EHR adoption, pushing hospitals from ~single-digit penetration to near-universal digitization in five years. The act achieves digitization but introduces compliance-driven UX burdens, shifts work onto clinicians, and fails to meaningfully solve cross-vendor interoperability.
- •$27B+ incentives (and later penalties) force rapid EHR adoption
- •Digitization success: ~9% to ~95% hospital adoption (2009–2014)
- •Downsides: ‘death by a thousand clicks,’ burnout, regulatory workflow bloat
- •Interoperability remains weak; incentives favored use compliance over open standards
- 2:34:27 – 2:55:38
Post-MU consolidation and the government-contract contrast: Cerner’s DoD/VA quagmire vs Epic’s focus
Meaningful Use accelerates vendor consolidation, increasing complexity for competitors and reinforcing Epic’s integrated advantage. Cerner wins massive DoD and VA contracts but suffers long delays and overruns, while Epic continues winning premier health systems and becomes the default training standard in US medicine.
- •Competitor M&A accelerates (Allscripts, Cerner-Siemens) and fragments suites
- •Cerner’s DoD (2015) and VA (2017) implementations balloon in time and scope
- •Epic keeps landing marquee systems (Mayo, Partners/Harvard, Intermountain, etc.)
- •Training network effects: most med students and top academic hospitals run Epic
- 2:55:38 – 3:57:01
Epic today and what’s next: Cosmos, AI scribes, valuation, and Judy’s succession trust
They quantify Epic’s present scale (customers, users, revenue) and highlight Cosmos as the platform that finally unlocks large-scale anonymized clinical analytics. The future discussion centers on payers/pharma expansion, ambient AI reducing documentation burdens, and a purpose-trust succession plan designed to keep Epic private and independent indefinitely.
- •Scale snapshot: ~607 customers, 5.7B revenue, ~14k employees, MyChart at massive reach
- •Cosmos: ~295M patients and 15B encounters for research and ‘lookalike’ discovery
- •AI/ambient scribe partnerships (Nuance, Abridge, Suki) and the ‘EHR fades into background’ thesis
- •Succession: purpose trust, no IPO/sale, CEO must be long-tenured Epic software developer
