Michael Kopko

The through-line

Kopko's consistent argument is that healthcare gets better when the people delivering care are inside the economics of it rather than fighting an insurer over them. Pearl Health takes responsibility for both the cost and the outcomes of Medicare patients in partnership with physicians and the government, which he frames as the alternative to a model where "they take the money and then they throttle what you get" [1]. That alignment premise then drives everything else he talks about: a prediction-and-prevention operating discipline, an all-in bet on building technology in-house, and a patience about timelines that he thinks most technology founders lack. The same commitment shows up in how he describes the mission more broadly, as primary care enablement at scale and democratizing access to value in healthcare [3], and in his interest in caring for patients beyond the one-on-one office visit and lowering the cost of care without sacrificing quality [2].

The second, newer preoccupation is what artificial intelligence does to the shape of a company. He is struck that firms are "hitting revenue rates with smaller talent bases than we've ever seen before," citing roughly a thousand people at Anthropic and OpenAI approaching ten billion in revenue, and concludes bluntly: "we're changing human productivity" [1].

On the aligned model versus insurance

Pearl helps physicians, hospital systems and nurses manage Medicare patients, but takes on the full cost and the outcomes rather than simply supplying a tool [1]. Kopko is direct about why that matters: the familiar version of cost accountability is insurance, "and people don't like insurance companies" because the incentive is to restrict what patients receive, a friction he says "has led to some really crazy things as of late" [1]. The alternative he describes as an aligned or accountable care model, one where "the people that are taking care of you are understanding of the cost and the consequences of the cost" [1]. He accepts this is a departure from how the system has worked, calling it "pretty radical," while noting Pearl is not alone in pursuing it, and he argues the evidence supports it: "you just get better care when your caretakers are in the entire equation of what needs to be delivered versus kind of at odds with the insurance company" [1]. The same logic extends to the practical problems he wants to solve for practices, including overburdened physician teams and the question of panel size [2].

On predicting and preventing

The intellectual core of the business is a claim about predictability in senior care. With six years of operating history and access to more than a million patients' worth of data, Kopko says Pearl can forecast what will happen to older patients with reasonable accuracy, and he insists this is not a hard problem in the way people assume: "death, taxes, and Medicare is pretty much a guarantee in everybody's life you make it to 65. Most things are predictable. There are not proverbial traffic accidents, and if it is predictable, it is broadly preventable" [1]. The operating discipline follows from that sentence: "How can we predict and then prevent through automation?" [1]. The stated results are lower costs, longer life, and avoided harm, which he calls the secret sauce, alongside scale figures of a quarter of a million seniors growing toward half a million, three and a half billion dollars of Medicare expense under management, and roughly one percent of traditional Medicare [1].

On Medicare's future and why value-based care is the affordability answer

Asked whether Medicare will still exist, he says yes without hesitation, though "it may have different rules and different elements to it" [1]. He expects pressure on the budget to produce debate about eligibility age and co-pays, but argues the structural fix is elsewhere: "value-based care, what we call accountable care, is the easiest way to make it more affordable longer term cuz we're saving money. We're constantly pulling 3 to 4% of the costs out" [1]. Underneath the fiscal argument is a values argument. Americans "don't like the idea of our seniors being unable to get healthcare" in their final years, and he does not expect that to change over the next ten or twenty years [1].

On selling to physicians

His view of the physician sale is shaped by having been Pearl's first salesperson. The opening line he used was deliberately self-deprecating: "If I pull out an app, you can hit me with my phone" [1]. He is careful to say this is not technophobia. Doctors "deeply value technology," but health technology is "the most oversold category, you know, in the country in many ways," with new solutions arriving daily [1]. The resolution is to invert the pitch: "you have to focus on what their outcomes are, not what you're trying to sell," which means talking about their business, their practice and their community's health first, then aligning technology to that [1]. He is honest that this makes sales slower and harder, and that behavior change and engagement came over five to six years rather than immediately: "Very few solutions kind of disrupt overnight" [1].

On the structural obstacles to selling technology into healthcare

He breaks enterprise resistance into three parts. First is point solution fatigue, and he is sympathetic to it: "Nobody wants to add yet another vendor into the mix," and larger organizations "have walls built to protect against that," including IT review, SOC, HIPAA, finance and compliance. Their culture is to preserve what works and run an RFP when something is genuinely needed [1]. Second is security, which he says goes beyond a checkbox. Passing SOC 2 and HiTrust is the easy part; the real question is "who are you? Who are your people? How long have you been doing this?" because "one data breach and you can ruin an organization" [1]. Third is interoperability. He observes that some EHR vendors want to play nicely and some do not, and is careful to frame it as strategy rather than malice: for Epic and Cerner, "their business model is not to say, 'How do I work with the next startup and integrate with them?'", whereas Athena and eClinicalWorks are more forward on it [1]. His practical advice is to start where integration is possible while working out how to partner with the large installed systems, "because they're installed. They're not going anywhere" [1]. Regulation adds its own tax: taking in patient data through a regulated process "adds to some of the expenses of what we do, and it also slows the speed a little, but that's acceptable because we're dealing with human lives" [1].

On the ten-year mindset

Kopko thinks the biggest founder error in health technology is temperamental. Healthcare does not produce social-app growth curves "because we're at human lives are at stake," and buyers deliberately bias toward conservatism: "They'd rather miss some opportunity than make a mistake that leads to someone's demise or injury or disease" [1]. Many entrepreneurs and engineering teams are simply not wired for that, and he says he framed Pearl accordingly from the start: "this is going to be a 10-year proof point," followed by the question of whether you become a real platform in years ten to twenty [1]. He pairs this patience with an optimism about value: "very few things that don't produce value can sustain," and while the time span can be frustrating, "if you're delivering real value creation, it's hard to keep you down" [1].

He also identifies domain knowledge as the second common failure mode, alongside the sheer cost and difficulty of getting a first product born and then bought. The buying cycle is "multi-year, it's multi-nodal, so you got to get a lot of people to say yes," and most businesses never clear it [1]. Founders arriving from defense or other industries carry a general appreciation of regulation but not the specifics, and he sees first-time health entrepreneurs struggle because they lack "norms, rules of the road, domain expertise" [1].

On building the full stack in-house

Pearl built its own platform rather than assembling third-party tools, working with AWS, cloud and AI inputs, but keeping ownership of the stack [1]. The organizational expression of that choice is scale: scrum teams paired directly with a product group form an R&D unit that is now about seventy percent of the company, which he describes as "an all-in bet on technology as the way through" [1].

On engineers, AI and productivity

He rejects the idea that AI diminishes strong engineers. They "will do even better, will be compensated even more, will be more productive, and will be able to develop and build things faster," even if what counts as a great engineer looks different than it did ten or twenty years ago [1]. The cost of building has fallen with it: where a first launch used to run into millions of dollars, he now puts it in the hundreds of thousands [1]. He also notes that hiring engineering talent was very hard in Pearl's early period, marking a shift in the company's experience across two distinct epochs [1].

On distributed work without attendance rules

Pearl has offices in New York, Boston and Seattle and Kopko says he enjoys seeing colleagues in them, but there is no requirement to be there. He deliberately avoided mandated days in office and tracking: "I love seeing people. I love working with people, but just never wanted to go into a rule-based system where we kind of take attendance" [1]. The one hard constraint is regulatory. Most staff are in the United States because Medicare data broadly cannot leave the country [1].

On what gets harder at scale

Now past product market fit and sustainable unit economics, he says the open question is "are you a category-winning platform?", which he defines as clients loving you, economics lining up, talent that perpetuates itself and a brand that speaks for itself [1]. Two things keep him up. One is pace: "are you doing enough? Are you keeping up with the pace of innovation and change" [1]. The other is talent transition, where "what got you here won't necessarily get you there" and the skills that carried the first startup years are no longer dominant, forcing senior hires that "can go well, that can go badly" [1]. He still describes the company in startup terms emotionally while conceding it has become a growth stage business, valuing the startup roots for "the ability to be able to build quickly and to have a lot of ownership" [1].

On staying in the market and leading through change

Maintaining product market fit, in his account, is not an analytical exercise conducted from a distance. Citing Mark Cuban's insistence on grinding, he argues that being in the market, running pitches, listening to the team, reading the Slack channels and paying attention to losses is what tells you where the organization has to go [1]. Those inputs produce moments of insight where "we're going to have to change. If we want to keep being what we want to be, this is no longer good enough" [1]. He is realistic about how such moments land, since "Nobody really really wakes up saying, 'I can't wait for things to change today'" [1]. That leaves the leadership problem as one of articulation and retention: explaining the change clearly enough to keep the right people through it. He explicitly rejects the prophet posture, saying it is not "a from the mountaintop type job where you just come down and say I figured out where the world's going" [1], and prescribes a mix of reflective distance and market presence, with humility as "dose number one in the morning. Just be grateful for what we've got and then you got to earn the keep today" [1].

Takeaways

• The aligned or accountable care model puts providers inside the cost equation rather than opposite an insurer, which he argues produces better patient outcomes than the insurance approach of taking the money and throttling what patients get [1].
• Senior care is largely forecastable, and his operating rule follows directly: "if it is predictable, it is broadly preventable," so the discipline is predicting and then preventing through automation [1].
• Selling technology to physicians requires leading with their outcomes, practice and community health rather than a demo, because health technology is the most oversold category they encounter [1].
• Enterprise healthcare resistance breaks into point solution fatigue, security diligence that goes far beyond SOC 2, and EHR interoperability where some vendors have no business reason to integrate with startups [1].
• Healthcare demands a ten-year proof-point mindset, because buyers deliberately prefer missing an opportunity to making an aggressive mistake with human lives [1].
• Pearl built its full stack in-house on AWS with cloud and AI inputs, and its R&D unit is now roughly seventy percent of the company, an all-in bet on technology [1].
• AI raises rather than lowers the value of strong engineers, and it has cut the cost of a first product launch from millions to hundreds of thousands [1].
• Value-based care is the practical route to Medicare's long-term affordability, pulling three to four percent of costs out, and he does not expect the program to disappear [1].