A leadership meeting is coming up. Someone wants a market size number for the board deck, the annual plan, or the next funding conversation. They don’t want a lecture on methodology. They want a figure that looks credible under pressure.
That’s where most market sizing work breaks down. Teams pull a broad industry report, lift the biggest category number they can find, apply a thin relevance filter, and call it TAM. It looks polished until someone asks basic questions. Which customers are in scope. What pricing assumption sits underneath the revenue estimate. How competition, geography, product limitations, or buying behaviour change the answer.
A useful market size equation isn’t just arithmetic. It’s a chain of evidence. If the model can’t survive scrutiny from finance, product, sales, and the CEO in the same room, it isn’t finished.
The practical standard is higher than “reasonable”. You need a number that is explainable, adjustable, and tied to observable market behaviour. That matters even more in noisy categories where competitors shift packaging, pricing, and positioning faster than annual planning cycles can keep up.
Table of Contents
- Why Your Market Size Slide Is Probably Wrong
- Core Concepts TAM SAM and SOM Explained
- Choosing Your Calculation Method
- A Worked Example of a Bottom-Up Calculation
- Common Pitfalls and How to Avoid Them
- Refining Estimates with Verified Competitor Intelligence
- Frequently Asked Questions on Market Sizing
Why Your Market Size Slide Is Probably Wrong
Most weak market size slides fail for a simple reason. They optimise for magnitude, not defensibility.
A team finds a large category number, labels it TAM, and hopes nobody asks how much of that category the company can serve. In practice, executives do ask. The hardest questions usually aren’t about the final figure. They’re about the assumptions buried inside it.
Three failure modes show up again and again:
- Category inflation. A broad software or analytics category gets treated as if every pound of spend is relevant to one product.
- Revenue mismatch. The model assumes an average contract value that doesn’t match how buyers purchase.
- Static assumptions. The estimate ignores competitor movement, changes in packaging, and shifts in who owns the budget.
That last point matters more than many analysts realise. A market size equation is often built once and presented as if it were durable. It rarely is. If rivals move upmarket, launch a lower-priced tier, or narrow their messaging to a different buyer, your assumptions can age fast.
A market size model should answer “why this number” before it answers “how big is the market”.
That’s why generic market research reports often disappoint operators. They’re useful for orientation. They’re weak as a final answer. A board deck needs something closer to an audit trail.
If you’re new to the discipline, a solid grounding in competitive intelligence fundamentals for 2026 helps because market sizing and competitor analysis shouldn’t sit in separate boxes. The number only becomes credible when it reflects market conditions.
Core Concepts TAM SAM and SOM Explained
Teams often use TAM, SAM, and SOM as if they were interchangeable. They’re not. Confusing them leads to bad planning, weak forecasts, and awkward executive reviews.
Think in filters not in one big number
The cleanest way to think about market size is as a series of filters.
TAM is the broadest layer. It represents the total revenue opportunity if every relevant customer that could use a solution like yours adopted it.
SAM narrows that set to the customers your business can serve with its current product, route to market, geography, and operating model.
SOM narrows it again to the share you can plausibly win in a defined period, given current competition, resources, and execution constraints.
Each layer answers a different strategic question.
| Layer | What it answers | Typical use |
|---|---|---|
| TAM | How large is the full opportunity space | Category framing, investor context |
| SAM | Which part of that space fits our business today | GTM planning, segment prioritisation |
| SOM | What can we realistically capture | Forecasting, hiring, territory planning |
What each layer means in practice
In UK SaaS, this distinction is especially important because category-level numbers can look healthy while the reachable opportunity is much narrower. An Antler overview of TAM, SAM, and SOM notes the UK SaaS market reached £16.3 billion in 2025, and that the competitive intelligence sub-segment remains underserved because manual tracking dominates 68% of CI leads. The same source also notes £45,000 average ACV for mid-market SaaS from Beauhurst 2025 data. Those figures are useful context, but they still don’t tell you your market.
A practical reading looks like this:
- TAM says the category exists and is commercially meaningful.
- SAM says whether your product can serve the buyers inside that category right now.
- SOM says whether your plan is believable.
Practical rule: If your TAM sounds exciting but your SAM and SOM are vague, you don’t have a market model. You have a category reference.
For PMM and CI teams, market sizing often improves when competitor dynamics are mapped properly. A structured competitive environment analysis framework for B2B strategy teams helps separate broad market demand from the narrower segment you can address.
The hierarchy also keeps stakeholder conversations cleaner. Finance can challenge the obtainable share. Product can challenge fit. Sales can challenge adoption timing. Nobody has to argue about one overloaded number pretending to answer every question.
Choosing Your Calculation Method
There isn’t one universal market size equation. There are several valid approaches, and each answers a different problem. Good analysts choose the method that matches the decision.
One proven starting point for UK B2B SaaS is the classic equation Total potential customers × Average revenue per customer = Market size. An Attest guide on calculating market size as a scale-up describes it as foundational, and cites the UK software market at £52.4 billion in 2023 with 12.4% CAGR from 2018 to 2023. For UK CI software, applying a 5 to 10% adoption rate across 1.2 million SMEs and 6,000 enterprises yields a TAM of £1.2 billion to £2.4 billion annually. The same source says UK startups using this equation pre-pitch saw 30% higher funding success from 2022 to 2024.
That’s useful. It still doesn’t mean it’s always the best method.
Top down when speed matters
Top-down sizing starts with an external market number, then narrows it by relevance.
It’s the fastest route to an initial estimate. It works well when you need category context, a rough order of magnitude, or a first-pass board slide.
Its weakness is precision. The more layers of reduction you apply after the headline figure, the easier it is to hide weak assumptions.
Use top down when:
- You need orientation fast and a directional view is acceptable.
- The category is established and external definitions are relatively stable.
- You’ll validate later with more operational data.
Bottom up when the number must hold up
Bottom-up sizing starts with your buying unit. Accounts, users, transactions, contracts, or seats. It then rolls up from that foundation.
This is the method I trust most for operating plans because every variable can be inspected. You can ask whether the target account count is sensible, whether expected purchases are realistic, and whether the annual spend matches your packaging.
A useful comparison framework for adjacent strategy work is in this guide to choosing the right competitive analysis framework.
The bottom-up route is usually slower, but it produces cleaner debates.
Here’s a short explainer before the final method:
Value based when willingness to pay drives the case
Value-based sizing asks what the problem is worth to the buyer, then estimates how much of that value your pricing can capture.
This is useful when the category is emerging or when budgets are justified by cost reduction, risk avoidance, or revenue impact rather than by an established software line item.
It’s also the easiest method to abuse. If the value story isn’t grounded in real buyer behaviour, the model becomes speculative quickly.
A practical comparison looks like this:
| Method | Best for | Main strength | Main weakness |
|---|---|---|---|
| Top down | Fast category framing | Quick and easy to communicate | Often too broad |
| Bottom up | GTM plans and operating models | Defensible and auditable | Slower to build |
| Value based | New categories or ROI-led sales | Captures buyer economics | Can become assumption-heavy |
The strongest models often triangulate. Start top down for context. Build bottom up for planning. Use value-based logic to test pricing realism.
A Worked Example of a Bottom-Up Calculation
A bottom-up model gets stronger when the analyst makes the buying mechanics explicit. That means naming who buys, how often they buy, and what they spend in a period.
The basic form is straightforward. Number of Target Users × Purchases Expected × Average Annual Spend.
For UK competitive intelligence software, a Valona explanation of market sizing gives a concrete version of that equation. It estimates 15,000 to 20,000 B2B CI or PMM operators in UK SaaS firms with more than 50 employees, uses 1.2 subscriptions per year, and £12,000 average ACV, producing a TAM of £216 million to £288 million for 2026. The same source says bottom-up work avoids the 25 to 40% overestimation common in top-down analysis.
Start with the buying unit
Assume you’re sizing a fictional B2B SaaS product used by PMM and CI teams. Don’t start with “all UK SaaS companies”. Start with the unit that purchases.
That might be:
- A PMM or CI operator inside a qualifying account
- A team subscription bought annually
- A multi-seat contract tied to a defined workflow
The variable choices matter because they determine what your model can defend later.
A clean build sequence looks like this:
Define the target user clearly
If the product serves PMM and CI workflows, count those operators rather than all knowledge workers.Set the purchase expectation
If contracts are annual but churn creates some replacement or expansion behaviour, document how you handle that.Apply annual spend carefully
Use the pricing structure that matches the target segment, not a blended number that hides enterprise outliers.
Build the model so someone else can audit it
A strong model should let another analyst reproduce your work without asking what you meant by each cell.
Here is a simple audit-friendly example.
| Variable | Data Source / Assumption | Value | Calculation Step |
|---|---|---|---|
| Target users | UK B2B CI/PMM operators in relevant SaaS firms | 15,000 to 20,000 | Starting population |
| Purchases expected | Annual subscription pattern with replacement effect | 1.2 | Apply frequency |
| Average annual spend | Average contract value | £12,000 | Apply revenue per purchase |
| Estimated TAM | Roll-up of above variables | £216M to £288M | Target users × purchases expected × average annual spend |
The maths is not the difficult part. The difficult part is defending the inclusion and exclusion logic.
If you can’t explain why a customer is in the model, remove them until you can.
Here, adjacent competitive research becomes useful. A careful competitor website analysis workflow often reveals whether vendors are targeting the segment your spreadsheet assumes. Pricing pages, buyer language, case-study patterns, hiring pages, and product packaging all help test whether your target account definition reflects market reality.
The practical discipline is simple:
- Document every assumption in plain language.
- Separate observed inputs from judgement calls so stakeholders know what’s factual and what’s interpretive.
- Keep ranges visible when uncertainty is real.
- Write down exclusions because those are often where the best executive questions come from.
A market size equation becomes credible when the spreadsheet shows its working, not just its conclusion.
Common Pitfalls and How to Avoid Them
The fastest way to lose confidence in a market model is to present it as more precise than it really is.
The easy mistakes
Some mistakes are obvious only after you’ve seen them a few times in reviews.
Treating TAM as the plan
TAM is strategic context. It isn’t the near-term target. When teams use it that way, forecasts become fantasy.Mixing incompatible sources
One source counts companies. Another counts users. A third reflects a different geography or segment. The model still totals neatly, but the underlying populations don’t match.Ignoring buying frequency
Some categories need a demand and frequency lens before value. A Clifton Private Finance guide to the market size formula describes the variant Potential customer demand × Frequency of sale, and applies it to UK SaaS using 150,000 PMM/CI leads with four intel needs per year. The same source says post-Brexit market sizing requirements for more than 5,000 bids increased use of these equations by 40% among consultancies, and that 65% of UK B2B firms using frequency-adjusted sizing achieved 22% faster GTM in 2023.Leaving competition out entirely
A market may be large, but your obtainable share depends on how crowded the segment is and whether rivals own the buyer narrative.
The fix is usually process not maths
Most of these issues are solved with discipline rather than a better formula.
Use a review checklist:
| Check | What to ask |
|---|---|
| Scope | Are all inputs talking about the same market and buyer? |
| Timeframe | Are the assumptions aligned to one planning period? |
| Revenue logic | Does spend reflect actual packaging and pricing? |
| Competition | Have we stated why this share is obtainable? |
| Evidence trail | Can another person trace every important assumption? |
Markets rarely punish teams for being conservative. Leadership usually punishes teams for being unable to defend the number.
If a model feels fragile, it usually is. Tighten the definitions, reduce the number of speculative variables, and make every assumption inspectable.
Refining Estimates with Verified Competitor Intelligence
A market size model is often treated as a one-time project. That’s a mistake. The assumptions start ageing the moment competitors change what they sell, how they package it, and who they target.
Static models expire quickly
Most annual market sizing exercises are snapshots. They’re useful for a moment, then reality moves.
A rival introduces a lower-priced tier. Another shifts messaging from enterprise to mid-market. A third starts hiring in a region you assumed was underpenetrated. None of those changes automatically rewrite your TAM, but they can change your SAM and SOM quickly.
That’s why competitor intelligence matters to market sizing, not just to battlecards or sales enablement. You’re not looking for chatter. You’re looking for public competitor movement that changes the assumptions inside the model.
The most useful signals tend to be observable and reviewable:
- Pricing changes that alter the practical entry point for buyers
- Packaging changes that expand or narrow who can buy
- Messaging shifts that show a move toward a different ICP
- Hiring patterns that suggest regional or segment expansion
- Proof changes such as new case studies, new compliance language, or new product-page positioning
What to update when competitors move
When a verified signal appears, update the model at the assumption layer, not by forcing the output.
For example:
| Competitor movement | Market model variable to review |
|---|---|
| New lower-tier plan | Average annual spend and reachable segment size |
| Enterprise-only packaging | Exclusions in SAM and expected win scope |
| Regional sales hiring | Geographic serviceability assumptions |
| New feature cluster for a buyer group | Target user count and segment relevance |
| Shift in homepage or pricing-page language | ICP definition and obtainable share logic |
This makes the market size equation dynamic rather than decorative. It also makes your executive conversations easier. Instead of saying “we revised SOM because the market changed”, you can say which public movement changed, when it changed, and which assumption was updated as a result.
Good market models don’t chase headlines. They absorb verified changes into documented assumptions.
For teams that want this level of evidence quality, the standard to aim for is an inspectable chain from source to interpretation. A useful reference point is Metrivant’s approach to verified competitor signals, where public movement is detected first and interpreted after verification. That trust boundary is what keeps updates defensible.
Without that discipline, “dynamic market sizing” becomes a softer phrase for reactive guesswork.
Frequently Asked Questions on Market Sizing
How often should a market model be updated
Update the full model on a planning cadence, but review assumptions whenever meaningful competitor movement appears.
In practice, the core structure shouldn’t change every week. The variables inside it might. Pricing, packaging, buyer focus, and geography are the assumptions most likely to need adjustment.
What if the category is new or poorly defined
Start narrower than feels comfortable.
Use a bottom-up frame based on known users, known workflows, and realistic spend. Then state clearly what sits outside the current model. In early categories, a smaller defensible estimate is more useful than a large speculative one.
Which is better top down or bottom up
For strategy communication, top down is fine as a first layer. For operating plans, bottom up is usually stronger.
The right answer in practice is often both. Use top down to show category context. Use bottom up to show the number you intend to operate against.
Should TAM, SAM, and SOM all be in the same deck
Usually yes, but they shouldn’t carry equal weight.
TAM gives context. SAM gives focus. SOM is where scrutiny tends to land because it connects directly to execution. If time is tight, spend the most effort on the assumptions behind SAM and SOM.
How detailed should assumptions be
Detailed enough that another person can challenge them without guessing what you meant.
Short labels like “mid-market buyers” or “enterprise pricing” aren’t enough. Name the segment logic, the buying unit, the spend logic, and any exclusions that materially change the result.
If your team needs to track competitor pricing shifts, packaging changes, messaging moves, and other public movement with an evidence chain you can defend in planning, explore Metrivant. It’s built for verified competitor intelligence so strategy, PMM, and CI teams can update assumptions with proof instead of noise.