ExpertsFMEpic Systems (MyChart)
Acquired · April 21, 2025
5 Selected Lessons:
heuristic · 2 case · 1 framework · 1 priority · 1
Epic's Seven Powers today: ranked sources of competitive durability
Source
Acquired — Epic Systems (MyChart) (April 21, 2025) · at 3:08:51 ▸ · at 3:12:07 ▸
When this applies
when analyzing why any scaled incumbent sustains differential returns over competitors
The takeaway
Durable advantage rarely comes from one power alone; look for businesses that stack several of the seven powers simultaneously.
Why it works
Multiple reinforcing power sources (switching costs, scale economies, network economies, branding, process power) compound so that losing any single one wouldn't be fatal, making the incumbent robust to many kinds of attack.
One level up
The strength of a competitive position is best judged by counting how many independent power sources reinforce it, not by the size of any single one.
Where it stops holding
This stops working as an analysis tool when a business has only one power source, which can be competed away by a sufficiently motivated rival.
Detail
hierarchy:
- Switching costs (biggest: EMR is the central nervous system of a hospital, failure means lives/revenue lost)
- Scale economies (only possible to fund broad product suite with huge customer base)
- Network economies (interoperability across hospitals; standardized training of doctors/nurses who move between institutions)
- Branding (trusted by most prestigious institutions, 'no one gets fired for buying Epic')
- Process power (proprietary training, MUMPS/Cache-based development skills that don't transfer elsewhere)
rationale: During the early 'takeoff phase' Epic arguably had no power at all -- it won by surviving on a small profitable niche while slowly building the full suite; only once it became the scale incumbent did these seven powers compound.
In their words
I mean, the most obvious one is switching costs. I mean, this is the biggest switching cost piece of software ever.
Bootstrapped software company via talent-as-capital
Source
Acquired — Epic Systems (MyChart) (April 21, 2025) · at 38:35 ▸ · at 43:08 ▸
When this applies
when a founder has a rare, cornered technical skill and wants to avoid dilution/control loss from investors
The takeaway
If you possess a truly scarce skill, you can bootstrap by using that skill as your capital, trading dilution and loss of control for slower, self-funded growth.
Why it works
A generational technical talent can personally produce output that would otherwise require a large funded team, substituting labor-capital for financial capital and avoiding the governance strings attached to outside investors.
One level up
Access to a cornered resource (rare talent) can substitute for access to financial capital in building a company.
Where it stops holding
This stops working when the founder's skill is not rare enough to single-handedly produce a competitive product, or when the market requires speed that only outside capital can buy.
Detail
procedure:
- Identify a founder with a rare, demonstrated technical skill (e.g., Judy's generational programming ability)
- Raise minimal seed capital just to cover hard assets (e.g., $70K for a minicomputer)
- Avoid venture capital to avoid being pressured to bring in 'grown-up' business executives
- Grow slowly, staying a small team for years while refining the product
benchmarks: {"total capital raised":"$70K equity + $70K debt over 47 years, reaching $5.7B revenue"}
rationale: Building software is not necessarily capital-light, but a generational programmer can substitute for the capital an army of engineers would otherwise require.
In their words
They had kind of a cornered resource as a startup. You know, Microsoft did in Bill Gates and Paul Allen, too. But Epic totally did too in Judy. Like she could write really, really great software. They didn't need to go hire an army of programmers.
Rarely raise prices aggressively on locked-in customers
Source
Acquired — Epic Systems (MyChart) (April 21, 2025) · at 1:54:06 ▸ · at 2:57:19 ▸
When this applies
when a business achieves high switching costs and could extract more price from captive customers
The takeaway
A company may choose to under-monetize a captive customer base to make switching away less appealing, rather than maximizing short-term price extraction.
Why it works
The hosts suggest that keeping price increases minimal (below inflation) removes much of the customer incentive to rebid, which may help preserve the switching-cost moat rather than eroding it through resentment-driven re-evaluation.
One level up
Long-term dominance sometimes requires forgoing available short-term profit to keep the defection threshold out of reach.
Where it stops holding
This stops working when competitors offer a genuinely disruptive alternative that makes the switching cost worth bearing regardless of your pricing restraint, or when leaving money on the table starves reinvestment.
Detail
rule: Avoid raising prices aggressively on existing locked-in customers, even when you have pricing power.
threshold: Epic's average yearly price increase is about 2% across the board, below inflation.
rationale: I think they want it to be an irrational and irresponsible decision for a healthcare system not to use Epic -- echoing Bezos's framing of AWS.
In their words
They basically never raise prices once you're a customer. They do, but their average yearly increase is about 2% across the board.
Bundle future products for free to prevent customer shopping
Source
Acquired — Epic Systems (MyChart) (April 21, 2025) · at 1:54:06 ▸ · at 1:57:21 ▸
When this applies
when an incumbent risks losing adjacent spend to point-solution competitors
The takeaway
Pre-announce and bundle upcoming adjacent functionality into the core contract so customers wait for you rather than shop elsewhere.
Why it works
Promising free future functionality within an existing enterprise agreement removes the incentive for a customer to evaluate and adopt a third-party point solution, even if the incumbent's version isn't ready yet.
One level up
Incumbents can defend adjacent markets not by being first, but by making switching to a challenger pointless before the challenger even has a chance to land.
Where it stops holding
This stops working when customers have urgent, immediate needs that your bundled future feature can't meet in time, forcing them to adopt a competitor regardless.
Detail
rule: When a new adjacent product category emerges, announce your own forthcoming free version to existing customers before building out a competitor's market share.
threshold: Applies when you already have the contractual relationship and the customer has nothing to lose by waiting (zero-cost switching to your version later).
rationale: You're certainly not going to go around shopping for some new thing when you know that it's just going to come for free to you.
In their words
It's this kind of amazing bundling strategy that certainly reminds me of the Microsoft episode.
Epic's failed Philips partnership created lasting aversion to partnerships
Source
Acquired — Epic Systems (MyChart) (April 21, 2025) · at 1:26:47 ▸ · at 1:29:47 ▸
When this applies
when a larger partner proposes a co-development/distribution deal that looks appealing but creates dependency on something outside your control
The takeaway
Be wary of partnerships that require ceding control of distribution or product to another company, since misaligned incentives usually surface as costly failures.
Why it works
Outsourcing distribution or joint development to a partner creates an uncontrollable dependency; when incentives aren't perfectly aligned or execution falters, the smaller party absorbs the cost and risk while gaining little durable advantage.
One level up
Organizations build lasting identity and strategy from their failures as much as their successes; a single costly partnership failure can permanently shape a company's risk appetite.
Where it stops holding
This stops working when the partner's incentives are genuinely and durably aligned with yours (e.g., true joint ventures with shared economics and control), or when you lack the resources to build and sell independently.
Detail
situation: Philips proposed licensing Epic's full IT system to resell under its own brand to radiology/international customers, promising expanded distribution.
decision: Epic invested significant resources (hiring in the Netherlands, building out the product) to support the partnership.
outcome: The partnership folded within a year or two of launch, becoming 'a really expensive detour' and creating lasting internal 'scar tissue' against partnerships.
lesson: This single failure shaped Epic's enduring strategy of staying focused on what it can control, going direct to customers, and avoiding integration partnerships -- contributing to its reputation for being closed/non-interoperable.
In their words
The whole thing ends up folding. And within a year or two of coming out, it was a really expensive detour. And the company develops this intense scar tissue for partnerships.