If you are preparing a raise, the market slide is one of the few places in your deck where an investor can check your judgment rather than your enthusiasm. That is why the question rarely stops at "how big is the market?" The follow-up, sometimes asked out loud and sometimes not, is "and how much of that can you actually sell to?"
Founders often treat these as one number shown at two zoom levels: a large figure to establish the opportunity and a smaller one to look reasonable. Investors read them as answers to two different questions. The total market tells them whether the prize is worth the risk. The addressable market tells them whether you understand who will buy. The distance between the two is where they decide how much to trust the rest of your plan.
Two numbers, two questions
Total market answers a ceiling question: if this category works, how large can it get? A top-down figure from a credible industry source is usually fine here, as long as you cite it and it measures what you sell rather than an adjacent category. Few investors will argue with your ceiling. They mostly want to confirm it is large enough that a good outcome is possible.
Addressable market answers a fit question. Of everyone inside that total, who has the problem badly enough to pay to solve it, has the authority to buy, and can be reached with the product and channels you have today? That answer depends on choices you have made about customers, product scope, and go-to-market. Because it reflects your choices, it is the number that reveals how you think, and that is why investors probe it.
Total market tells an investor the prize is worth the risk. Addressable market tells them whether you know who will buy.
The gap is the argument
The common mistake is not a total market that is too large. It is an addressable number produced by multiplying the total by a convenient percentage. "If we capture 1% of a $10 billion market" yields a number without telling anyone who the customers are, why they would switch, or how you would reach them. The figure is not unconvincing because it is big. It is unconvincing because nothing connects it to a buyer.
A defensible addressable market is built from filters, and each filter is a claim about customers that can be tested:
Not everyone in the category experiences the pain your product solves, or experiences it often enough to act.
Budget authority, organization size, and procurement rules remove some of the market before a sales conversation starts.
Geography, integrations, language, and channel limit the market you can serve this year, whatever you plan for later.
Customers with a workable current solution, or high switching costs, belong in a later phase of the plan.
Investors do not expect each filter to be precise. They expect each one to have a basis. A filter you can explain with evidence moves the meeting forward. A filter you cannot explain becomes the question that takes up the rest of the meeting.
A worked example (hypothetical)
Consider a hypothetical founder building scheduling and patient-intake software for independent physical therapy practices. The figures below are round, illustrative numbers, not market data.
Suppose a published source puts annual spending on practice management software for outpatient clinics at $2 billion. Independent practices, excluding hospital-owned clinics and large chains that buy enterprise systems, account for an assumed 40% of that spending, or $800 million. Of those, the product only helps practices still running intake on paper or a general-purpose scheduler.
The founder's first draft assumes 60% of independents fit that description, which puts the addressable market at $480 million. Then the founder tests the assumption: about 15 interviews with practice owners to learn how they describe the problem, followed by a short survey of 150 independent practices. The survey finds 35% still use manual or general-purpose intake. The addressable market becomes $280 million.
The second number is smaller, and it is the stronger slide. When an investor asks "how do you know?", the founder has an answer. A sample of 150 also carries a margin of error of roughly plus or minus 8 percentage points at 95% confidence for a result near 50%, which is precise enough to separate "about a third" from "most," the distinction that matters for this decision.
The energy industry formalized this decades ago
Utility energy efficiency planning has used the same layered logic for a long time. Potential studies separate technical potential (everything physically possible), economic potential (the portion that is cost-effective), and achievable potential (what customers will realistically adopt, given program design and real barriers). Planners and regulators rarely treat the first number as the plan. Program goals are set against the third.
In my years working in and around utility programs, including at Austin Energy, where survey responses were joined with program participation, billing, and Census data, the layer that took the most care was the one that depended on customer behavior. Published data could say who was eligible. Only customers could say who was likely to act.
A startup faces the same structure with less formal vocabulary. Total market resembles technical potential. Addressable market resembles economic potential narrowed by your reach. The share you can realistically win in the next few years is your achievable potential, and that is the number your financial model should rest on.
Where research fits
Most filters can be estimated from secondary sources: establishment counts from the U.S. Census Bureau's County Business Patterns, industry association reports, public company filings. The filter that rarely can is the one investors push hardest on: how many prospective customers have the problem badly enough to act on it.
That filter calls for primary research, and the study does not need to be large. It needs a small set of interviews to define the problem in buyers' own terms, then a short survey that reaches the right segment to measure how common the problem is and what customers use today. Getting the right people in the sample matters more than the number of completes.
What to bring into the meeting
A credible market slide carries four things: one total market figure with its source; an addressable figure built from named filters; the basis for each filter; and the filter you are least sure of, with what you are doing to test it. Naming the weak link yourself reads as judgment. Leaving an investor to find it reads as the opposite.