Beyond the Spreadsheet: Data-Driven Site Selection for Healthcare Startups
You don't need years of patient data to pick a first site. Start with a sharp Ideal Customer Profile, use public federal and state data to find where that population clusters (your TAM), and answer four questions: how many patients you need, where your ICP concentrates, how far they'll realistically travel, and what market share you need. Aim for about 5%; if you'd need more than 10% to be profitable, it's the wrong area. Don't over-build the V1 model.
Trying to launch your first clinic and feeling overwhelmed by site selection with limited data is a common, frustrating challenge. Founders get bogged down in endless spreadsheets and complex models, which ultimately delays the care their patients need.
The 0-to-1 reality
Most healthcare startups won't have years of historical sales or patient demographics to feed a sophisticated quantitative model. You're operating from a 0-to-1 perspective, but the need for a data-backed site decision is just as critical. The answer is a strategic blend of targeted quantitative analysis (especially your Total Addressable Market) and critical qualitative factors.
How do you do site selection without historical data?
Define your Ideal Customer Profile (ICP)
Before you touch a spreadsheet, get clear on who you're trying to serve, with precision. Beyond basic demographics, consider:
- Specific health needs
- Socioeconomic factors
- Typical behaviors
- Cultural nuances
That sharp definition becomes your compass for navigating the data.
Pinpoint your TAM with publicly available data
Once your ICP is clear, find where those demographics cluster. You don't need proprietary data to start; reliable public sources published by federal and state agencies will get you there.
High-risk senior primary care: focus on areas with a high concentration of the 65+ population on Medicare and Medicaid.
Pediatric dentist: target areas with commercially insured 5-to-19-year-olds who live and go to school within your estimated service radius.
This focused demographic review is powerful. It identifies promising areas with the highest probability of success and, just as importantly, lets you confidently rule out areas that won't align with your care model and patient-acquisition goals.
Why does over-building the V1 model backfire?
A common trap is building an overly complex initial model. It nearly always leads to analysis paralysis and derails speed to market. Two reasons it backfires:
At this stage, your TAM and market-share estimates are educated guesses. Throwing more variables into a V1 model doesn't make those projections more accurate; it just adds noise and complexity to what is still a foundational assumption.
An overly precise model assumes the perfect physical space matching your criteria actually exists in the exact location the model points to. The reality of commercial real estate is far more nuanced.
What questions should drive a 0-to-1 site decision?
When you need to move fast, get clear on the answers to four fundamental questions:
- Patient acquisition. How many patients do we realistically need to acquire to reach the profitability metrics we've communicated to investors?
- Customer concentration. Where is the highest concentration of customers that truly fit our ICP?
- Travel radius. How far do we genuinely believe people will travel for our specific services? Be realistic about patient convenience.
- Market share. What percentage of market share within an estimated service area do we need to capture to hit profitability?
Pursuing areas where you'd need significantly higher share (say, more than 10%) to be profitable is an extremely challenging uphill battle for growth.
Your quantitative analysis is meant to confirm one thing: that a sufficient concentration of your ideal customers exists within your estimated service area to support profitability.
Smart, data-driven site selection gets you into the right neighborhood. But your care model and your team's ability to deliver high-quality, patient-centric care are what actually drive unit-level growth and profitability. The data points the way; the care delivers the results.
Key takeaways
- You don't need proprietary data to pick a first site; a sharp ICP plus public federal and state data produces a defensible decision.
- Define the ICP before the spreadsheet: health needs, socioeconomics, behaviors, and cultural nuance, not just demographics.
- Use public data to find where your ICP clusters (your TAM) and, just as valuably, to rule areas out.
- Answer four questions: patients needed for profitability, where the ICP concentrates, realistic travel radius, and required market share.
- Target roughly 5% market share. If you'd need more than 10% to be profitable, it's the wrong area, and don't over-engineer the V1 model.
Frequently asked questions
How do you choose a clinic site without historical patient data?
Lead with an Ideal Customer Profile, then use public federal and state data to locate where that population clusters, your Total Addressable Market. From there, answer four questions: how many patients you need for profitability, where your ICP concentrates, how far patients will realistically travel, and what market share you need to capture. That blend of a sharp profile and public data produces a defensible 0-to-1 decision without proprietary history.
What public data can you use for healthcare site selection?
Reliable demographic and population data published by federal and state agencies is enough to start, no proprietary dataset required. You're looking for where your ICP concentrates: for a senior primary care model, areas dense with 65+ residents on Medicare and Medicaid; for a pediatric dentist, areas with commercially insured 5-to-19-year-olds inside your service radius. The goal is both to find promising areas and to rule out ones that don't fit.
What market share should a new clinic target?
About 5% is a solid benchmark for a new organization within its estimated service area. If your model requires significantly more, say north of 10%, to reach profitability, that's a warning sign: it will be an uphill battle for growth, and it usually means the trade area doesn't hold enough of your ideal customers to support the model.
What is an Ideal Customer Profile (ICP) in healthcare?
An ICP is a precise definition of the patient you're built to serve, going beyond age and income to include specific health needs, socioeconomic factors, typical behaviors, and cultural nuances. It's the first step in site selection because it becomes the filter for every piece of data that follows, telling you which markets to pursue and which to rule out.
How do you avoid analysis paralysis in site selection?
Keep the first model simple. At the 0-to-1 stage your TAM and market-share figures are educated guesses, and piling on variables adds noise rather than accuracy. Avoid assuming the perfect space exists exactly where the model points; commercial real estate is messier than that. Use the analysis to confirm a sufficient concentration of ideal customers exists, then move, rather than perfecting a spreadsheet.
Get Data-Driven Site Selection Right
Don't let analysis paralysis delay your mission. I help healthcare founders navigate site selection with the right blend of quantitative rigor and practical reality, whether it's your first site or your next.
Schedule a Conversation