Your Patients Are Everywhere. Your First Clinic Can't Be.
A virtual-first company holds an advantage almost no first-time clinic operator has: data on exactly who chooses it for care. Use it. Have your clinicians segment the panel to find which patients would get better outcomes in person, map where that specific segment concentrates, then learn how that local market moves. Run the same data-first read on payer rates, regulatory load, and clinical labor. The heat map shows where your marketing worked. Your data shows where a clinic will work.
The two most common ways I see virtual-first companies pick the city for their first clinic are the highest concentration of current patients, or wherever leadership happens to live. The board agrees it's time for a physical site, someone pulls up the patient heat map, points at the brightest blob or the home metro, and says: there. I've sat in that meeting more than once. Both are the most natural place to start, and both are reliable ways to pick the wrong market.
The opposite mistake is just as common: overanalyzing broad psychographic and demographic segmentation, the kind of third-party market study I argued against in data-driven site selection. So if the heat map is too blunt and the market study is too broad, what's the right way?
You're holding something more useful than either, and it's the reason a virtual-first company can choose a first market with more precision than almost any operator building de novo: you already know exactly who chooses you for care. The work is turning that knowledge into a location.
This is the fourth piece I've written on the virtual-to-physical transition. The earlier ones covered what physical care buys you and why your virtual playbook won't transfer. This one covers the decision that comes before any of that: where the first clinic goes.
What advantage does a virtual-first company bring to this decision?
An operator building de novo picks a market on third-party demographics, a market study, and instinct. You don't have to. You already know the age, the conditions, the acuity, and the geography of every patient who has elected you for care. That panel is the most accurate market research you will ever own, and most companies leave it sitting in a database while they argue over a map.
Your patient data is the asset. The first clinic is what you build on top of it. Start with who already chooses you, not with where the dots look densest.
Which of your patients would do better in person?
This is a clinical question before it is a real estate question, and it is where your clinicians earn their seat at the table. Put the panel in front of them and have them segment it: which patients, by condition, acuity, and complexity, would see materially better outcomes from in-person care? The diabetic whose management has plateaued on video. The patient who needs hands-on monitoring, a procedure, or labs that a screen has been compromising. That group is your in-person population.
It is a subset of the panel, and it is the only count that matters for a clinic. Five thousand virtual patients in a metro might contain eight hundred who clinically belong in a room with a provider. Build for the eight hundred. A flat assumption that "15 to 30 percent will convert" gets you a number; clinical segmentation gets you the right number, and it tells you which service lines the clinic needs to offer on day one.
Where does that segment concentrate?
Now map the segment, not the whole panel. The unit is the drive shed: the 20-to-30-minute trip a patient will make to a front door. Metros don't visit clinics; drive sheds do.
Greater Dallas spans 9,000 square miles. Your 4,000 metro patients tell you nothing until you ask where your in-person segment clusters tightly enough to fill a schedule. Then ask the question that decides whether this is a business or a one-off: is that clustering a pattern you can find again in the next market, or a quirk of this one? The first market should teach you something repeatable.
How do patients move in that market?
Concentration on a map is not the same as reach on the ground, and this is where local real estate knowledge earns its place. How patients travel changes completely by market. You can't read New York the way you read Charlotte. New York is walk times, subway lines, and which neighborhood you're standing in. Charlotte is highways and drive times. Suburban markets are steadier, but every one of them has its own quirks: the river nobody crosses, the two towns that never mix, the interchange that adds 20 minutes at 5pm.
The question that tells you the most: where does your segment already go to get care? People build healthcare habits around routes they already drive, the highways and landmarks and town names they use to decide what's close and what's a trek. A location on one of those routes works. A location that asks patients to break a habit to come see you fights gravity from day one.
How do you pressure-test the market beyond patients?
Your patient data tells you who and where. Three more reads tell you whether the market will support the clinic, and you run each the same way: with your own data, not assumptions. Score every surviving market on these before you tour a single space.
Payer relationships and in-person rates
Your contracts live state by state, and in-person reimbursement is a different schedule than virtual. Pull your own claims data: where does your payer mix pay well for in-person care, and where would a clinic see the same patients for less? If you hold value-based contracts, check whether the utilization targets are even reachable without a physical site.
Regulatory load
State licensure, facility codes, and clinic classification rules vary enough to swing your opening date by four to six months. Two markets identical on patient data can sit a half-year apart on time-to-open. Price that delay in burn, not in patience.
Clinical labor in one zip code
Your virtual team was hired from anywhere. The clinic needs MAs, front desk, and phlebotomy within commuting distance of one address, and the hospital system down the road hires the same people. Check local wage data and posting volume before you commit, not after.
The same lens, every time
Patients, payers, regulatory, labor: each is a data question with an answer sitting in your systems or a public dataset. Run all four and a list of eight candidate metros usually collapses to two or three, where you weigh demand against speed against cost. No market wins all four.
Should your first clinic go where leadership lives?
There's a filter nobody puts on the slide: where the leadership team lives. I've watched it silently override the data.
Here's the fair version of the argument. A first clinic is a prototype, and you only build your first clinic once. Prototypes need daily attention: walking the space during construction, sitting in the waiting room during week two, watching where the flow breaks. A clinic you can drive to gets fixed faster than a clinic you fly to.
So treat proximity as a tiebreaker, not a filter. If your HQ city passes the data tests, proximity is a legitimate advantage and you should weight it. If your HQ city fails them, putting the clinic there means you've optimized for your commute instead of your care model. And the model is what the next two years of fundraising will be judged on. Name the tradeoff out loud in the room. Distance costs management attention. A weak market costs the thesis.
Do you choose the market or the site first?
Once a market survives the data, the work changes shape: state, then metro, then drive shed, then site. Only at that last step does this become a conventional first-clinic project: space program and second-gen versus build-out decisions, structure decisions like medtail versus ground-up, and the LOI. I've watched teams collapse these steps, start touring spaces in week one, and end up reverse-justifying a market because the broker found a charming endcap.
Example: One virtual-first group I advised had its largest state panel in California and its leadership in Austin. The heat map said Los Angeles. The data said otherwise. Their clinicians segmented the panel and found the in-person population in Texas skewed toward exactly the conditions the clinic was built to serve. Their California payer book reimbursed in-person poorly relative to their Texas contracts, and the California licensing path added roughly five months. Austin had two-thirds the raw panel, the right clinical segment concentrated in one drive shed, and contracts that paid for in-person care on day one. The smaller blob won, and the clinic reached breakeven inside a year.
How does your first market shape every site after it?
One more thing the heat map can't see. Your first market is also a curriculum: the segmentation model, the payer mechanics, the staffing approach, and the buildout playbook you develop there become the template site #2 inherits. A first market that's an outlier, with an unusually favorable payer arrangement or a regulation-light environment you never plan to repeat, teaches you lessons that don't travel.
So pick the market that proves the model, not the one that flatters the map. The brightest blob is where your marketing worked. The right market is where your in-person segment, your payer contracts, the state's rulebook, and the local labor pool all say yes at the same time, and where the way you found it will work again in the next city.
Key takeaways
- A virtual-first company's biggest edge in choosing a first market is its own data: you already know who elects you for care, which no de novo operator does.
- Have clinicians segment the panel to find who would get better outcomes in person. That subset, not the whole panel, is the demand a clinic serves.
- Map where that segment concentrates inside a 20-to-30-minute drive shed, then read how the local market moves — travel patterns vary completely between New York, Charlotte, and a suburb.
- Run the same data-first read on payer relationships and rates, regulatory load, and clinical labor; an eight-metro list usually collapses to two or three.
- Treat HQ proximity as a tiebreaker, not a filter, and remember the first market becomes the template every later site inherits.
Frequently asked questions
How should a virtual-first healthcare company choose its first physical clinic location?
Start with your own patient data, the advantage no de novo operator has. Have your clinicians segment the panel to find which patients would get materially better outcomes in person, map where that segment concentrates inside a 20-to-30-minute drive shed, and learn how patients move in that local market. Then run the same data-first read on payer rates, regulatory load, and clinical labor. The right market clears all of these; the brightest blob on the heat map usually clears none.
How do you know which virtual patients need in-person care?
Treat it as a clinical question first. Put the panel in front of your clinicians and segment by condition, acuity, and complexity to find the patients whose outcomes improve with hands-on care, monitoring, procedures, or labs that virtual visits compromise. That segment is your true in-person demand. A flat conversion assumption gives you a number; clinical segmentation gives you the right number and tells you which service lines to open with.
What is a drive shed, and why does it matter more than the metro?
A drive shed is the 20-to-30-minute drive-time polygon around a clinic's front door. Patients don't visit a metro; they visit whatever they can reach in a reasonable trip. Greater Dallas spans 9,000 square miles, but a single drive shed inside it might capture only a few hundred of your in-person segment. Map the segment against drive sheds, and account for how travel patterns differ between dense urban, suburban, and highway-driven markets.
Should I open my first clinic near company headquarters?
Only if your HQ city already passes the data tests. Proximity is a legitimate tiebreaker because a first clinic is a prototype that needs daily, in-person attention. But if the HQ city fails on patient segment, payer, regulatory, or labor grounds, putting the clinic there optimizes for your commute instead of your care model — the thing your next raise will be judged on.
Picking Your First Physical Market?
I help virtual-first healthcare organizations turn their own patient data into a first-market decision before the broker tours start: clinical segmentation, drive-shed mapping, payer geography, and the site strategy that follows.
Schedule a Strategy Session