Healthcare market-sizing needs a reset
- Chris Loumeau
- 5 hours ago
- 8 min read
Practically every organization in every sector of healthcare routinely needs market-sizing models to guide high-stakes strategic decisions, such as where to invest and where to be cautious. The problem: In the rush to check all the planning boxes, it is all too easy for leaders to accidentally end up with—and even bet the strategy on—'models' that are not really models at all, but simply numbers. Numbers with unexamined, opaque, and sometimes wildly inaccurate underlying assumptions, inputs, and scenario choices. The end result, formatted neatly in Excel and/or Powerpoint, ends up carrying far more authority than it deserves. These "houses of cards" are putting the success of the whole strategy at risk.
In fact, this failure mode is increasingly common in healthcare strategy work today, particularly in areas like AI, automation, and services markets, because in these spaces, the “market” is rarely a clean product category. Adoption is shaped less by product features and more by the impacted workflows, competing budgets, and the organizational risk tolerance—creating opportunities for massively over- or under-estimated uptake trajectories. As a result, the savvy healthcare executive should no longer just ask, “How big is the TAM, SAM, or SOM?” They should be taking whatever answer they get back and ask, “What would have to be true for these numbers to be real—and how exposed are we to tenuous assumptions underneath them?”
Those are the right questions. And they’re exactly why many healthcare market sizing efforts need a reset. Let's take a look at most common errors we tend to see in market-sizing models—and talk through the actions I think are needed to get better results.
Quick context check-in: The TAM/SAM/SOM funnel
To level-set, some key market-sizing terms upfront. It's key to understand that market sizing models are built like a pyramid, and the three main layers are TAM, SAM and SOM.
• Total Addressable Market (TAM) is the base layer: The full economic picture (or universe) of potential demand for a given product or service.
• Serviceable Addressable Market (SAM) is the middle layer: It narrows that universe to the portion an organization could realistically serve, given its market positioning and focus.
• Serviceable Obtainable Market (SOM) narrows it further to the point on the top: The share that could plausibly be captured, given competition, buyer behavior, and existing constraints.
When done well, this progression should bring clarity to critical decision-points, not obscure them. But when leadership cannot trace a TAM, SAM, or SOM estimate back through its underlying assumptions, inputs, and scenario choices. That's when it's time to sound the 'potential house of cards' alert.
.
Why market sizing models break down the moment users start asking real questions
To understand why so many market sizing models fall apart, it helps to look at what happens when leaders actually start utilizing them to guide organizational decision-making—not just presenting them in leadership discussions.
Flimsy TAM models tend to fail in (one or all of) the same three ways:
The assumptions are hard to find. The model generates a number, but the root assumptions that feed it aren’t clearly spelled out. They’re embedded in Excel formulas, glossed over as “reasonable,” or never surfaced at all. As a result, the TAM, SAM, or SOM figure looks authoritative, but it’s difficult to explain (or defend) once questions start coming.
The market is defined too broadly to be useful. Instead of starting with what and why buyers actually purchase (from a workflows perspective), many models begin with an overly-broad market label and work backwards. That approach ignores how decisions are really made inside healthcare organizations, where spending is tied to specific needs, workflows, and budgets—not abstract market categories.
The final numbers aren't tied to how money actually flows. The model may describe the benefits of a product or service to some degree, but it isn’t built around key considerations like pricing logic, adoption friction, or competitive positioning. Without that real-world grounding, the final numbers reflect possibility more than plausibility.
This is why so many healthcare market sizing conversations right now are devolving into internal philosophical arguments. When a TAM or SOM estimate isn’t connected to operational and financial reality, teams can’t meaningfully test it. If the internal debate shifts from “Are our core assumptions right?” to “Do we like where this number landed?”—that's when the model stops being a dynamic strategic tool.
Why 'the market' is often the wrong place to begin
Once you realize how delicate many, if not most, market sizing models actually are, the solution becomes clearer: The problem is often the starting point. A traditional, top-down market-sizing approach begins with a very large market number and works backward from there. While that can be useful for basic healthcare market orientation, it unfortunately often bakes in model assumptions that don’t hold true once you get closer to real-world decision-making. Broad categories tend to mask important differences. But buyer behavior isn’t uniform. And industry adoption is shaped by far more than technical need.
A more consistently reliable approach builds from the bottom up. Instead of beginning with “the market,” it begins with the work that actually happens inside the organizations that make up your target audience. It asks what specific tasks are being performed, what problems those tasks create, and where time, money, or risk is already being managed today (and how effectively). From there, it ties potential economic value to specific operating budgets—labor, utilization, margin, or revenue—that have real owners and real constraints.
This shift in approach forces clarity early. It makes budget control explicit. It surfaces workflow friction before enthusiasm takes over. And it keeps the model anchored in how healthcare organizations actually operate, not how markets are labeled. It's not about being cautious for caution’s sake. It’s about building something that leadership can fully stand behind once real questions start getting asked.
The market-sizing building blocks that hold up
The section above explains how models collapse. In this section we'll walk through a simple checklist for building one that doesn’t. What separates sturdy market-sizing models from flimsy ones is simpler than it looks—it’s in evaluating what has to be true for a model to survive real scrutiny. At execution, most defensible market sizing work comes down to a small number of concrete decisions, whether teams label them explicitly or not.
A durable model moves in a clear sequence: Who controls the spend, what work is being changed, where the economic value comes from, what level of adoption is feasible, and what portion of that opportunity a specific company could realistically capture.
1) Start with who controls the budget—not who feels the pain
In healthcare, the person (or people) most affected by a problem is rarely the same person who actually approves the purchase of its solution. Especially in provider world, budget authority is often fragmented across finance, HIT, operations, clinical leadership, and governance. Treating “health systems” or “providers” as a single buyer without specifying where the budget actually sits often inflates the forecasted market size. A defensible model names the buyer in real budgetary terms, not demographic ones.
2) Be precise about the workflows being changed
Strong TAM modeling doesn’t begin with a named market category—it begins with a specific task. Broad labels like “AI,” “automation,” or “care management” don’t tell you what is actually changing inside an organization. Models hold up when they identify the specific activities being affected, how those activities are performed today, and what organizational friction is being reduced. This is what determines willingness to pay and reveals the real barriers to market adoption.
3) Tie economic value to where money already moves
Once the impacted workflow(s) is clear, the next step is connecting economic value to a specific operational unit. In healthcare, "value" is frequently linked to an array of discrete economic metrics like labor costs, avoided utilization, margin improvement, risk exposure, throughput improvement, or revenue gains. Each behaves differently, has different owners, and faces different constraints. When value is treated generically, the TAM model loses credibility—because buyers don’t make decisions against abstract “savings.”
4) Treat market adoption and vendor capture as separate questions
Even when value is real and well-defined, market adoption is not automatic. Healthcare organizations adopt third-party solutions within specific planning and contracting windows, under integration constraints, and alongside competing priorities. True market capture is narrower still, shaped by competition, procurement dynamics, and switching costs. This is another area where many market-sizing efforts falter—doing careful work on the top of the TAM funnel, then (however unintentionally) treating SAM and SOM as simple reductionist math. In actuality, this is where leadership's attention should be most focused.
Scenario discipline: What low, base, and high cases are actually for
Once the structure of a market sizing model is sound, scenario discipline becomes the real test of whether it’s useful. This is another weak point where many models can suddenly fall apart. Low, base, and high cases are often treated as labels rather than tools (e.g., “conservative,” “reasonable,” and “optimistic”) or, worse, as formally implied forecasts. That’s not what a model's scenarios are for.
In a defensible model, scenarios exist to test assumptions (not outcomes). They are designed to answer questions like: "What actually drives this market size?", "Where is the downside risk?", and "How sensitive is the result to changes in behavior, adoption, or pricing?"
Strong scenario discipline is surprisingly simple:
Change one or two assumptions at a time
Observe what meaningfully changes
Be explicit about why
When models are treated this way, something important tends to emerge: many of the individual variables don’t matter all that much, but a few of them matter a lot. That sensitivity profile is the insight. It shows leadership which of the modeled assumptions the strategy is truly betting on—and which ones are largely background noise. In that sense, scenario modeling isn’t about predicting the future. It’s about making uncertainty visible and discussable, rather than hiding it behind a single “best guess” number.
How to use the model as strategy infrastructure
Market sizing does its most important work after the model is built. Used properly, it becomes part of your strategy infrastructure. It gives leadership a shared reference point for tradeoffs, sequencing, and where to prioritize time. It’s not just a set of numbers to review and move on from.
A robust model should make it easier to answer questions like:
Which market segments deserve our early attention? Why?
What practical constraints will shape adoption? How should those constraints influence timing and rollout?
Where is pricing logic anchored in real economics versus where is it still largely aspirational?
Which modeled assumptions represent meaningful exposure? (Such that getting them wrong would force a change in strategy.)
From this perspective, sensitivity analysis isn’t a rote academic exercise. It’s a pragmatic guardrail. It helps leaders and teams avoid confusing a compelling market story with a go-to-market strategy that actually withstands scrutinized pressure.
The goal of any and all market-sizing effort is calm defensibility
The point of market sizing isn’t precision for its own sake. It’s dynamic decision support. A defensible market-sizing model doesn’t promise certainty, but instead creates shared understanding. Leaders can see what the execution strategy depends on, where vulnerabilities live, and how sensitive the true market opportunity is against existing real-world constraints. Such an approach forces conversation realignment from defending a final output to debating underlying model assumptions.
In a healthcare environment so often defined by notoriously long sales cycles, gradual adoption of new technologies, and constant ROI scrutiny, that pivot is the difference between intentional market analysis that informs execution strategy and a loosely-disciplined market assessment that merely decorates it.
Looking for help with an analytical model?
As I'm sure you've gathered if you've made it this far: Proper market-sizing is a bit of a passion of mine, and something I enjoy doing. If your team is enlisting support in building an analytical model, or wants reusable model scaffolding for this kind of market sizing discipline—especially for AI and services markets where adoption is assumption-sensitive—this is precisely the kind of project work Union does in our custom research division. If you're interested in learning more, reach out to info@unionhealthcareinsight.com.