Bottom-Up Market Sizing: Stop Guessing, Start Counting

Bottom-up market sizing works by calculating your addressable market from the ground up: counting real potential customers, multiplying by realistic deal values, and arriving at a number you can actually defend in a boardroom. It is the opposite of taking an industry report figure and dividing by your assumed market share, which is a method that feels rigorous but is usually just optimistic fiction dressed in a spreadsheet.

Done properly, bottom-up sizing gives you a market estimate grounded in actual buyer behaviour, real pricing, and the constraints of your go-to-market capacity. It is harder to build than a top-down number, but it is worth the effort because it forces you to confront what your market actually looks like rather than what you wish it looked like.

I have spent a fair amount of time on both sides of this problem. Running agencies, I needed to understand the real size of the markets my clients were competing in, not the inflated figures their founders had pulled from Gartner or IDC reports. And when I was growing teams and pitching for new business myself, I needed to know whether the opportunity I was chasing was genuinely worth the pursuit or whether I was about to spend six months winning a client who would never generate meaningful revenue. Market sizing done badly wastes both.

Why Top-Down Sizing Creates a False Sense of Security

The standard approach to market sizing in most strategy decks is to find a credible-sounding industry report, quote the total market value, and then apply a percentage to arrive at your target. The logic looks clean. The problem is that the percentage is usually invented, the total market figure often includes segments you cannot reach, and the whole thing creates a number that feels authoritative but has no operational grounding.

I have sat in enough pitch reviews and planning sessions to know that top-down figures get challenged the moment someone in the room asks a simple question: how did you arrive at that number? If the honest answer is “we took the industry report and divided by ten,” the credibility of everything else on the slide tends to wobble with it.

Top-down sizing also conflates TAM, SAM, and SOM in ways that matter. Your Total Addressable Market is the theoretical maximum if you had 100% market share and no constraints. Your Serviceable Addressable Market is the portion you can realistically reach given your geography, product fit, and go-to-market model. Your Serviceable Obtainable Market is what you can realistically win in a defined timeframe. Most planning errors I see come from treating TAM as if it were SAM, which inflates opportunity assessments by multiples, not percentages.

If you want a fuller picture of the research methods that sit alongside market sizing, the Market Research & Competitive Intel hub covers the broader toolkit in one place.

How to Build a Bottom-Up Model That Holds Up

The mechanics of bottom-up sizing are straightforward. You start by defining the unit of analysis, which is usually an account, a buyer, or a transaction. You count how many of those units exist in your target market. You apply a realistic conversion rate based on your actual sales data or a defensible proxy. And you multiply by your average deal value or transaction size to arrive at a revenue figure.

The formula looks like this: Number of target accounts multiplied by conversion rate multiplied by average contract value equals your realistic market opportunity. The inputs are simple. Getting them right is where most of the work lives.

Step One: Define Your ICP With Enough Precision to Count It

A bottom-up model is only as good as the buyer definition feeding it. If your Ideal Customer Profile is “mid-market B2B companies,” you cannot count them with any accuracy. If your ICP is “B2B SaaS companies with 50 to 250 employees, a dedicated marketing function, and an annual revenue between £5m and £50m, operating in the UK and DACH markets,” you can start pulling real numbers.

This is where a structured approach to ICP scoring for B2B SaaS becomes genuinely useful, not as a theoretical exercise but as the input layer for your sizing model. The more precisely you can define the firmographic, technographic, and behavioural attributes of your best-fit customers, the more accurately you can count how many of them exist.

When I was at iProspect growing the team from around 20 people to over 100, we had to get serious about which client segments were actually worth pursuing. Not every company that could buy performance marketing services was a good fit, and not every good-fit company was reachable given our capacity and go-to-market model. The discipline of defining the ICP tightly was what made the market opportunity feel real rather than theoretical.

Step Two: Count the Accounts That Actually Exist

Once you have a tight ICP definition, you need a count. There are several ways to approach this depending on your market and your resources.

For B2B markets, LinkedIn Sales Navigator, ZoomInfo, Cognism, and similar platforms let you filter by firmographic criteria and export a rough account count. These numbers are not perfect, but they are grounded in something real. Companies House data, industry association membership lists, and trade directories can supplement or cross-check platform data for specific verticals.

For consumer markets, the approach shifts. You are counting people rather than accounts, which means drawing on census data, ONS population statistics, panel data, or survey-based estimates. The principle is the same: you want a count of real individuals who match your buyer profile, not a percentage of a broad demographic category.

Search engine marketing intelligence can also be a useful proxy here. Keyword search volume data gives you a signal of how many people are actively looking for solutions in your category. It does not give you a precise account count, but it does tell you something about the scale of active demand, which is a different and often more commercially relevant number than theoretical addressable market.

I saw this play out clearly when I was at lastminute.com. We launched a paid search campaign for a music festival and generated six figures of revenue within roughly a day from a relatively simple setup. The reason it worked was not sophisticated targeting. It was that we had a clear picture of who was searching, what they were searching for, and what the conversion economics looked like. The market size, in that context, was the volume of people actively expressing intent. That is a more actionable number than any industry report figure.

Step Three: Apply Realistic Conversion Assumptions

This is where bottom-up models most often go wrong. The temptation is to apply an optimistic conversion rate because the resulting number looks better. The discipline is to use a rate you can defend with reference to actual data.

If you have existing sales data, use it. Your historical lead-to-close rate, adjusted for the quality of the new segment you are sizing, is far more reliable than an industry benchmark. If you are sizing a market before you have any sales history, look for proxies: comparable companies in adjacent markets, conversion data from pilot campaigns, or reference points from Forrester’s work on lead definitions and conversion benchmarks.

The point is not to find a number that makes the model look attractive. It is to find a number that reflects how buyers in this market actually behave. If your category has a long sales cycle and a high evaluation cost, a 2% conversion rate from prospect to customer might be realistic. Plugging in 15% because it makes the TAM look exciting is how companies end up with pipeline forecasts that never materialise.

Step Four: Anchor to Real Transaction Values

Average contract value or average transaction value is the multiplier that turns your account count into a revenue figure. Like conversion rates, this number needs to come from reality rather than aspiration.

For established products, use your actual ACV data, segmented by the ICP tier you are sizing. For new products or new markets, use pricing research, competitive intelligence on comparable products, or willingness-to-pay data from customer conversations. Qualitative research methods like focus groups can surface pricing anchors and value perceptions that are genuinely useful inputs here, particularly when you are entering a market without existing transaction data.

One thing I have learned from running agencies across multiple sectors is that average deal value is almost always more variable than it first appears. The mean can be dragged up by a handful of large accounts in ways that make the typical deal look bigger than it is. Median ACV is often a more honest input for market sizing purposes.

Where Bottom-Up Sizing Connects to Competitive Intelligence

A market size figure in isolation is only partially useful. What gives it commercial meaning is understanding how that market is currently distributed: who is winning it, at what price points, through what channels, and with what retention economics.

This is where bottom-up sizing connects directly to competitive intelligence work. If you know your SAM is 4,000 accounts and you know your three main competitors each have roughly 400 to 600 customers, you can start to build a picture of market concentration, white space, and realistic displacement rates. That is a much more useful planning input than a raw market size number.

Grey market research can be particularly valuable here. Scraping job postings, analysing G2 and Capterra review volumes, tracking LinkedIn employee counts at competitor firms, and monitoring their hiring patterns all give you signals about where competitors are growing and where they are not. These are not perfect data sources, but they are grounded in observable behaviour rather than self-reported claims.

I have used this kind of intelligence work when assessing whether to invest in a new vertical or a new geography. The question is never just “how big is the market?” It is “how much of this market is genuinely available to us, given who is already there and how entrenched they are?” Bottom-up sizing gives you the first number. Competitive intelligence gives you the second.

Stress-Testing Your Model Before You Rely on It

Any market sizing model, however carefully built, contains assumptions. The discipline of stress-testing is about making those assumptions visible and understanding how sensitive your conclusions are to changes in each input.

Run your model with three scenarios: a base case using your best estimates, a conservative case where you cut your conversion rate and ACV by 30%, and an optimistic case where you increase both by 20%. If the conservative case still shows a market worth pursuing, you have a strong opportunity. If the optimistic case is the only scenario that makes the economics work, you should be asking harder questions before committing resources.

This kind of scenario discipline is also useful for understanding which input variable matters most. Sometimes the model is most sensitive to account count. Sometimes it is ACV. Sometimes it is conversion rate. Knowing which lever drives the outcome most tells you where to focus your validation effort before you commit to a market entry or a budget allocation.

A SWOT and strategic alignment analysis done alongside your market sizing model can surface the structural constraints that your spreadsheet will not capture: regulatory barriers, channel dependencies, capability gaps, and competitive dynamics that affect whether your conversion assumptions are realistic in the first place.

Using Pain Point Research to Validate Your Sizing Assumptions

One of the most common mistakes in market sizing is treating it as a purely quantitative exercise. The numbers need to be anchored in a real understanding of why buyers in this market buy, what drives urgency, and what creates friction in the purchase process.

Pain point research is not just useful for messaging. It is a validation layer for your market sizing assumptions. If your model assumes a 5% conversion rate but your buyer research reveals that procurement cycles in this sector average 14 months and require board-level sign-off, your conversion rate assumption needs revisiting. The market might be the right size, but the velocity at which you can access it is constrained in ways your spreadsheet does not reflect.

Early in my career, before I had the budget or the team to do this properly, this clicked when the hard way. I was asked to assess the opportunity in a new vertical, built what I thought was a solid model, and presented a number that looked compelling. The problem was that I had not spoken to enough buyers in that sector to understand that their procurement process made the conversion timeline almost twice as long as our existing business. The market was real. Our ability to access it in the timeframe the model assumed was not. That is a distinction that matters enormously when you are allocating headcount and budget against a market opportunity.

When to Update Your Model

A bottom-up market sizing model is not a document you build once and file. Markets change. Your product changes. Your go-to-market motion changes. Pricing evolves. Competitors enter or exit. Any of these shifts can materially affect the inputs to your model, and therefore the conclusions you draw from it.

As a rule, I would revisit the model whenever you make a significant change to your ICP, your pricing, or your go-to-market approach. I would also revisit it annually as part of the planning cycle, using updated account data and refreshed conversion and ACV figures from the previous year’s actual performance.

The goal is not to produce a perfect number. It is to maintain a living estimate that reflects current reality closely enough to support good decisions. There is a useful parallel here with how lead management disciplines work in practice: the value is not in the model itself but in the discipline of keeping it current and using it to drive consistent decision-making rather than relying on intuition or outdated assumptions.

If you want to go deeper on the research methods that feed into market sizing and competitive analysis, the Market Research & Competitive Intel hub covers primary research, competitive intelligence frameworks, and audience analysis in one place.

The Commercial Discipline Behind the Numbers

Bottom-up market sizing is in the end a discipline of honest accounting. It forces you to count real things: real accounts, real buyers, real deal values, real conversion rates. It replaces the comfortable vagueness of “the market is worth £X billion” with the more demanding question of “how many buyers can we actually reach, what will they pay, and how many of them can we realistically win?”

That is a harder question to answer. It is also a more useful one. The companies and marketing leaders who build strategy on honest market estimates, rather than inflated ones, tend to make better resource allocation decisions, set more realistic targets, and avoid the costly experience of discovering mid-year that the market they planned against does not exist in the form they assumed.

I have seen both outcomes play out across 20 years and 30 industries. The difference between them is rarely about the quality of the execution. It is almost always about the quality of the assumptions that went into the plan before execution started. Bottom-up sizing is one of the most reliable tools available for making those assumptions honest.

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