Market Size Measurement: Stop Guessing, Start Approximating

Measuring market size means estimating the total revenue opportunity available for a product or service in a defined market, over a defined period. Done well, it tells you whether a market is worth entering, how much share is realistic to capture, and whether your growth targets are grounded in something real or invented in a spreadsheet.

Most businesses either skip this step entirely or produce a number so inflated it becomes meaningless. The fix is not a more sophisticated model. It is a more honest one.

Why Most Market Size Estimates Are Wrong From the Start

I have sat in more than a few boardrooms where a slide appeared showing a total addressable market of several billion dollars, followed by a target market share of two or three percent, presented as though two percent of a number nobody had verified was a conservative ambition. It is not conservative. It is a way of making an arbitrary revenue target look like it emerged from analysis.

The problem begins with how market size is typically sourced. A business buys or downloads an industry report, finds the headline TAM figure, applies a percentage to represent their slice, and calls it done. No one interrogates where the headline number came from, what geography it covers, what it includes or excludes, or whether the methodology bears any resemblance to how the actual market behaves.

This is not market sizing. It is citation laundering. The number looks authoritative because it came from an external source, but the logic connecting it to your actual business opportunity has not been examined at all.

Good market sizing starts with a different question: not “how big is the market?” but “how many customers could realistically buy this, at what price, how often?” That reframe changes everything about how you build the estimate.

What TAM, SAM, and SOM Actually Mean

These three terms appear in almost every pitch deck and strategy document, but they are frequently misused or conflated. Getting them right matters because each one answers a different question.

Total Addressable Market (TAM) is the total revenue opportunity if you captured 100% of the market with no competitive constraint. It is a theoretical ceiling, not a realistic target. Its value is in understanding the scale of the space you are operating in, not in setting revenue expectations.

Serviceable Addressable Market (SAM) is the portion of TAM that your product, geography, and business model can actually reach. If you sell enterprise software in English to companies with over 500 employees in North America, your SAM is dramatically smaller than your TAM. Most businesses underestimate how much TAM they need to exclude to arrive at a credible SAM.

Serviceable Obtainable Market (SOM) is the realistic share of SAM you can capture given your current resources, competitive position, and go-to-market capability. This is the number that should connect to your actual revenue targets, and it is the one most frequently inflated.

When I was running agencies and we were pitching for new business or planning growth, the discipline of working through SAM rather than TAM was what separated credible plans from aspirational ones. TAM tells you the game is worth playing. SAM and SOM tell you whether you can actually win in it.

If you want to go deeper on analytics frameworks and how measurement thinking applies across marketing functions, the Marketing Analytics hub at The Marketing Juice covers the broader landscape, from attribution to dashboards to GA4.

Top-Down vs Bottom-Up: Which Approach to Use

There are two fundamental approaches to measuring market size, and they produce very different outputs, for different reasons.

Top-down sizing starts with a macro figure, typically from an industry report or analyst estimate, and works downward by applying filters: geography, segment, price point, customer type. The appeal is speed. The problem is that each filter introduces an assumption, and those assumptions compound. By the time you reach your SOM, you may be four or five assumptions deep, none of which have been validated.

Bottom-up sizing starts with the unit of sale and works upward. How many potential customers exist? What is the average transaction value? How frequently do they buy? Multiply those out and you get a market size estimate built from observable inputs rather than inherited assumptions. It takes longer, but it produces a number you can actually defend, because every component is visible and challengeable.

In practice, the most useful approach is to run both and compare them. If they land in a similar range, you have reasonable confidence. If they diverge significantly, that divergence is itself valuable information. It tells you that one of your input assumptions is wrong, and finding which one is more useful than either number on its own.

I have seen teams spend weeks refining a top-down model and produce a figure that collapsed the moment someone asked “but how many companies actually buy this category, and how often?” The bottom-up question takes ten minutes to frame and immediately exposes whether the top-down number is credible.

How to Build a Bottom-Up Market Size Estimate

The mechanics are straightforward. The discipline is in being honest about what you know versus what you are assuming.

Step 1: Define your customer unit precisely. Not “SMEs” or “consumers.” A specific profile: companies with 50 to 250 employees in the professional services sector in the UK, or adults aged 30 to 55 who own a home and spend on home improvement. The more specific your definition, the more credible your count will be.

Step 2: Count the universe. Use publicly available data sources: census data, Companies House, trade association membership figures, government business registers, industry bodies. You are looking for the total number of units that match your customer definition. This will be an estimate, and that is fine. Name your source and note the margin of uncertainty.

Step 3: Apply a penetration rate. Not every potential customer will buy. What percentage of your universe is genuinely reachable, has the problem you solve, and has the budget to act on it? This is where honest thinking matters most. A penetration assumption of 40% in a category where market leaders hold 15% is not credible. Base your penetration rate on analogous markets, competitor data, or your own sales conversion history.

Step 4: Apply average revenue per customer. What does a typical customer spend, per year, on this category? Not on your product specifically, but on the category. This is your unit of market value. If customers spend on average £5,000 per year on the category and there are 100,000 reachable customers, the market is £500 million at full penetration.

Step 5: Sense-check against known reference points. Are there public companies in your space with disclosed revenues? Are there trade body estimates you can cross-reference? Does your bottom-up figure sit in a plausible relationship to those reference points? If your estimate is three times larger than the combined disclosed revenue of the three largest players in the market, something is wrong.

Where to Find Market Size Data Without Paying for Analyst Reports

Expensive analyst reports are not the only source of credible market data. In many cases, they are not even the best source, because their methodologies are opaque and their definitions rarely match your specific market.

Public company filings are underused. If there are listed companies in your sector, their annual reports, investor presentations, and earnings calls contain market size estimates, growth rate commentary, and competitive positioning data that has been reviewed by auditors and investor relations teams. It is more reliable than most bought reports.

Government statistical agencies publish sector-level output and expenditure data. In the UK, the ONS publishes detailed industry breakdowns. In the US, the Census Bureau and Bureau of Economic Analysis cover most sectors. These are not always perfectly aligned with commercial market definitions, but they provide a credible anchor.

Trade associations frequently publish annual industry surveys with market size data. The methodology varies, but the figures are typically based on member reporting rather than modelled estimates, which gives them a different kind of credibility.

Search volume data is a useful proxy for demand in consumer markets. If you can quantify the volume of people actively searching for a category of solution, you have a demand signal that is independent of supply-side estimates. SEMrush’s thinking on data-driven marketing covers how search data can be used as a commercial input, not just a content planning tool.

Competitor pricing and customer count data, where publicly available, lets you reverse-engineer market revenue. If a competitor has 5,000 customers at an average contract value of £10,000 per year, they are generating £50 million in revenue. If they hold an estimated 20% market share, the market is approximately £250 million. That is a market size estimate built from observable inputs, not inherited from a report.

The Assumption Problem: Why Precision Is the Wrong Goal

One of the most persistent errors in market sizing is presenting the output as though it were a measured fact rather than a structured estimate. A market size figure expressed to three significant figures, produced from a chain of assumptions, is not more accurate than a range. It is just more confident-looking.

When I judged the Effie Awards, one of the things that separated strong entries from weak ones was how entrants handled uncertainty. The strongest papers named their assumptions, quantified the range of plausible outcomes, and were honest about what they did not know. The weakest presented single-point estimates with false precision and collapsed under questioning.

The same principle applies to market sizing. A range of £200 million to £350 million, with named assumptions driving each end of the range, is more useful than a single figure of £275 million that implies a level of certainty you do not have. Forrester has written about the dangers of treating analytical outputs as unquestionable, and that caution applies here. The model is not the market. It is a structured way of thinking about the market.

Present your assumptions explicitly. Label each one. Note which assumptions have the most leverage on the final number. If your penetration rate assumption moves from 10% to 15%, what does that do to the output? Sensitivity analysis of that kind is not a sign of weakness in your model. It is a sign that you understand it.

How Market Size Connects to Marketing Planning

Market size is not an academic exercise. It connects directly to how you allocate budget, set targets, and assess performance. If you do not know the size of the market you are operating in, you cannot know whether your growth is coming from market expansion or share gain, and that distinction matters enormously for strategy.

When I was growing an agency from around 20 people to over 100, one of the most clarifying questions we asked was not “how do we grow?” but “how big is the market we are actually competing in?” The answer changed what we thought was possible, who we thought our real competitors were, and which client segments were worth prioritising. Market size thinking is strategy thinking.

It also connects to budget justification. If your SOM is £50 million and you are currently generating £5 million in revenue, you have significant headroom. That headroom justifies investment. If your SOM is £8 million and you are already at £5 million, the conversation shifts entirely. You are not in a growth market. You are in a share consolidation or category expansion play, and the marketing strategy should reflect that.

For teams building out their analytics capability more broadly, the Marketing Analytics section of The Marketing Juice covers how to connect market-level thinking to campaign measurement, attribution, and performance reporting.

Market size data also helps you avoid the common trap of optimising for metrics that look impressive but are disconnected from commercial reality. If you are growing impressions in a market that is contracting, you are not winning. Understanding market context is what separates performance reporting from performance theatre. Unbounce’s breakdown of how to make marketing analytics meaningful touches on this, noting that metrics only matter when they are anchored to a commercial outcome.

Common Errors That Undermine Market Size Estimates

Confusing TAM with SAM. A global market figure applied to a business that operates in three countries with a mid-market product is not a useful starting point. The gap between TAM and SAM is often 80 to 90 percent, and pretending otherwise produces targets that have no relationship to reality.

Using the same figure across different decisions. The right market size for an investment case is not the same as the right market size for a budget allocation decision, or a competitive analysis. Each use case has a different level of precision required and a different definition of “the market.” Treating one number as universally applicable creates confusion downstream.

Not updating estimates. Markets change. A market size estimate built three years ago on pre-pandemic data, or before a major regulatory change, or before a category-defining competitor entered, is not the same market. Treating stale figures as current is a form of wilful imprecision. Build a review cadence into how you use market size data, just as you would with any other planning input.

Ignoring the demand side. Most market sizing focuses on the supply side: how many companies sell in this space, what are their revenues. But market size is in the end a demand-side question. How many buyers exist, what problem are they trying to solve, and how much are they willing to pay? Supply-side proxies are useful, but they can mislead in markets where unmet demand is significant or where informal competition is not captured in company revenue data.

MarketingProfs has a piece on the cost of skipping rigorous preparation in analytics that, while focused on web analytics, makes a point that applies directly here: the failure to define what you are measuring before you measure it is where most analytical errors begin.

What a Credible Market Size Output Looks Like

A credible market size estimate has five components. Not a single number, not a slide with a large figure and a small-print disclaimer, but a structured output that can be interrogated.

A defined market scope. Geography, customer type, product category, time period. If the scope is not defined, the number is not defined.

A named methodology. Top-down, bottom-up, or both. What data sources were used. What assumptions were applied at each step.

A range, not a point estimate. With a base case, an upside case, and a downside case. Each driven by a specific assumption changing, not by arbitrary percentage adjustments.

A sensitivity table. Which assumptions have the most leverage on the output? If your penetration rate assumption is wrong, how wrong does the market size get? This is the most useful part of the analysis for decision-making, and it is almost always omitted.

A stated purpose. What decision is this estimate informing? That context shapes how precise the estimate needs to be and which assumptions matter most. A market size estimate for an investment case needs different rigour than one built to frame a campaign brief.

Forrester’s writing on how to structure analytical outputs for decision-making is relevant here. The discipline of connecting an analytical output to a specific decision, rather than producing analysis for its own sake, is what separates useful measurement from measurement theatre.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions