How to Size Any Market From Scratch
How to Size Any Market From Scratch
- Understanding a market is one of the most important skills in strategy, investing, and consulting. If you cannot size the opportunity, you cannot make a decision.
- Think of TAM, SAM, and SOM like matryoshka dolls. Nested layers that help you move from "everyone who could theoretically buy this" to "what we can actually capture."
- There are two approaches: top-down and bottom-up. Top-down is faster because it is first-principles driven. Bottom-up takes longer because it is a detailed, building-block view of real market participants. In practice, you often need both.
- A market size without a view on where it is going is only half the picture. Headwinds and tailwinds are essential to understanding the future market size and growth opportunity — you need a stronger perspective than just today's snapshot.
- The offering: A step-by-step checklist for starting a market sizing workstream from scratch.
The first time I had to size a market for real was at McKinsey, on a due diligence for a consumer product. It was a unique opportunity because I had never really looked at that market before, so building a view from scratch — through the lens of a live transaction — was both challenging and genuinely interesting.
I felt confident because I had the backing of an entire team and firm to help me work through it. And that is kind of the point. Market sizing is not about being a genius. It is about being structured, being resourceful, and knowing where to look. Once you understand the building blocks, you can size almost anything.
I now teach this at Concordia University and it is one of the topics my students find most valuable — because it applies everywhere. Whether you are in consulting, investing, corporate strategy, or building your own business, the ability to frame and size an opportunity is foundational.
Here is how I think about it.
The easiest way to think about market sizing is through matryoshka dolls — those nested Russian dolls where each one fits inside a bigger one. The market works the same way. You start with the largest possible view and narrow down, layer by layer, until you arrive at something realistic and actionable.
- TAM — Total Addressable Market. This is the entire universe of demand. Everyone who could, in theory, buy the product or service if there were no constraints. It is the biggest doll. Think of it as: if you had 100 percent market share and infinite reach, how much revenue could exist?
- SAM — Serviceable Addressable Market. This is the realistic market. It narrows the universe down to the customers who are actually willing and able to buy at today's price point. This is where you apply what I call a "haircut" to the TAM — filtering for adoption rates, geographic reach, customer segments that actually make sense, and price sensitivity. The SAM gives you a much more grounded picture of the opportunity.
- SOM — Serviceable Obtainable Market. This is the portion of the SAM that your company can realistically capture in the near term, given your current resources, competitive position, and go-to-market capabilities. If the SAM is the realistic market, the SOM is what you can actually win right now.
People in private equity often think of this as a concentric circle exercise — pockets that fit within broader pockets. Same idea. The point is that each layer forces you to get more precise about who the customer really is, what they are willing to pay, and how much of that is actually addressable.
TAM is conceptually clear for most people. Where it gets interesting — and where many people trip up — is in defining the SAM. What does "serviceable" actually mean for this specific product or market? Which customer segments are truly reachable? What adoption rate is realistic? This is where structured thinking and good segmentation become critical.
There are two fundamental approaches to sizing a market, and in practice, you often need both.
- Top-down sizing starts with the total theoretical population of customers — all the people or businesses who could reasonably buy the product — and estimates how much they would spend. You multiply population by price to get a TAM, then apply a haircut to get the SAM. It is generally faster because it is first-principles driven, which makes it incredibly useful for framing an opportunity quickly. The trade-off is that it can overstate the market if the inputs are not refined. Think of top-down as your first line of defence — it gives you a directional gut-check that you can build on.
- Bottom-up sizing builds the market from the ground up by summing the actual revenues or unit sales of real market participants. You deeply analyze the largest players, understand the number of customers they serve at a given price, and then estimate the rest — the long tail — using proxies like store counts, regional averages, or comparable benchmarks. This approach takes longer because it is a detailed, building-block view, but it is more robust and credible in boardroom discussions. Bottom-up is also a proper way to size the vended market — what is actually being sold and serviced today — and you can extrapolate the TAM from there. It requires creativity when reliable data is not available, but the exercise builds real conviction.
In the real world, you can always stress-test a market through a top-down approach because it is based on first principles. That makes it a natural starting point. However, in my experience, it is important to also do a thorough bottom-up analysis — time permitting — to get a detailed assessment of the opportunity, the competitive supply, and a clear view on where the market is headed. Sometimes you run both simultaneously and use them to cross-check each other. There is always a risk of false precision, but running the exercise is what builds conviction.
The more granular and well-thought-out your inputs are, the more your answer will evolve — sometimes deviating meaningfully from your first pass. That is normal. The goal is not precision to the last dollar. It is a first-principles, fundamentals-driven view of whether the market is big enough to be worth pursuing.
The total addressable population represents all of the actual consumers — individuals or businesses — that could buy the product or service in an ideal world. But not all customers are created equal.
Take a luxury hospitality brand as an example. It would be misleading to treat every guest as a single, uniform type. In reality, they serve distinct high-volume customer segments: leisure travellers booking for vacation, corporate travellers attending conferences with high frequency but lower nightly spend, group and event travellers bringing larger tickets, and a small but lucrative segment of celebrities and VIPs with very high willingness to pay.
Defining the right segments requires structured thinking. Segments should be MECE — mutually exclusive, collectively exhaustive — meaning no overlap but complete coverage. To build these segments, you pull from primary sources like 10-Ks, investor day decks, and earnings call transcripts. You supplement with expert insight from industry professionals who understand customer behaviour on the ground. And you triangulate with third-party reports from firms like IBISWorld, Statista, or major consulting firms.
Keep the customer segments you uncover top of mind for later steps. When sizing the SAM, you will need to understand which segments are ready and willing to adopt the offering. And when assessing market trends, segmentation matters too — economic tailwinds like rising corporate travel budgets will affect one segment differently than a recession cutting leisure spend.
When sizing a market, knowing how big it is today is only half the picture. You need a strong perspective on where it is going — understanding the future market size and growth opportunity is essential. Will demand and price grow, shrink, or stay flat over your investment horizon? In most diligence or strategy contexts, we take at least a five-year lens.
You can think about this through two lenses.
- Tailwinds are macro forces that grow demand and expand the market over time. These are the structural or cyclical trends working in the market's favour. Examples of tailwinds include things like population growth, rising health awareness, regulatory support for an industry, new technology adoption, and shifts in consumer behaviour like post-COVID revenge travel. When identifying tailwinds, you want to distinguish between those that are structural and long-lasting versus those that are temporary or cyclical.
- Headwinds are macro forces that contract demand or put downward pressure on the market. These are the risks and challenges that could slow growth or shrink the opportunity. Examples of headwinds include things like economic recessions, new regulation or compliance burdens, the emergence of substitutes, changing consumer preferences away from the product, rising wage pressure across industries, and technology disruptions like Apple's privacy changes impacting digital advertising. Headwinds are just as important as tailwinds because they shape the risk profile of the opportunity and help you build a balanced perspective.
You can bring this to life by forming a directional view: will the market grow a lot, grow a little, stay flat, shrink a little, or shrink a lot? Think about velocity too — 10 percent or more growth is a very different story than 2 percent, even if both are technically "growing."
Some trends can be quantified directly using hard data. Population growth comes from census and federal data. Pricing trends come from inflation indices. Industry growth rates often appear in analyst reports from IBISWorld, Statista, or major consulting firms.
Other trends require triangulation and multiple data points to validate — things like shifts in consumer behaviour post-COVID, or competitive dynamics driven by new entrants affecting pricing power and market share. And some trends require you to develop conviction using analogous markets and precedent cases. For example, predicting the adoption of wearable fitness devices might be informed by looking at smartphone adoption curves.
The ultimate goal is to build a well-reasoned narrative — not just state a growth rate, but explain why the market will look different in three to five years and whether that makes the opportunity more or less attractive.
The first step in any market sizing exercise is getting smart on the space. That means pulling together credible sources. Here is where I typically start.
- Earnings calls and quarterly filings. These are gold. Official filings give you a clean view of revenue, costs, and earnings. Earnings calls give you the narrative — how leadership frames performance, what the market actually cares about, and where analysts are probing. Platforms like Qartr and Seeking Alpha make these incredibly accessible and are generally free to use.
- Annual and quarterly filings (10-Ks, annual reports). Operational metrics hidden in the text, full financial snapshots, risk sections that highlight the challenges management actually worries about, and segment-level breakdowns that expand your understanding. Public filings can generally be downloaded in Excel or Word, which makes the analysis process much simpler.
- Third-party reports. Top consultancies, PE firms, and investment banks publish quarterly or annual sector reports with market sizing, growth forecasts, and key drivers. Trade unions and industry stakeholders often have hard-to-find operational data. Specialized research firms offer deep but sometimes inconsistent coverage. And broad platforms like Statista or IBISWorld are useful for quick directional framing but should always be validated against higher-credibility sources.
- Press releases and company sites. Useful for triangulating specific events or milestones. Hidden gem: company blurbs often include operational or financial metrics worth capturing.
- Investor Days. Gold mines when available. Companies showcase their performance, metrics, and competitive moats. Not every firm hosts them, but when they do, it is a unique window into strategy and positioning.
- AI tools. Great for brainstorming driver trees, sizing approaches, and locating credible sources. Still early days — never treat outputs as hard data. Always double-check before using insights.
Here is the step-by-step checklist I follow — and teach my students — when starting a market sizing workstream from scratch.
One last thing. Market sizing is not about getting to the perfect number. It is about building a structured, defensible view of whether an opportunity is worth pursuing — and being able to explain why. The frameworks are simple. The discipline is in the details. And the more you practice it, the faster and sharper you get.
