TAM, SAM, SOM: How to Estimate Your Market Size

TAM, SAM, SOM.

Everyone talks about them, and just as many probably hate them.

Why? Because it’s often unclear what they mean, how to measure them, and, most importantly, where to find accurate, reliable data to make your calculations trustworthy.

That’s exactly what we’ll uncover in this guide.

Let’s dive in.

TAM (Total Addressable Market) 

TAM represents the entire market for the problem your offering solves, either globally or across all potential segments.

TAM is valuable for setting a bold vision and attracting investor interest. The general consensus is that VCs typically look for a TAM of over $1B to consider the opportunity substantial.

SAM (Serviceable Available Market)

SAM represents the portion of TAM you can target, given your product/service scope, geography, target segments, business model, and legal and regulatory limitations.

SAM = TAM × Market Accessibility Coefficient

Where Market Accessibility Coefficient = the percentage of the total market that you can realistically target and serve, considering both how well your product/service fits target customers and factors like geography and legal restrictions.

For example, if you're a fisherman:

• TAM is the entire ocean and all the fish in it.
• SAM is the part of the ocean you can realistically reach, given your licenses, fishing regulations, and the size of your boat.

Even if your TAM is huge, your SAM may represent only a small portion of it if:

• Your product is only available in specific regions.
• It’s designed for a limited customer segment.
• It doesn’t meet all regulatory requirements.

Consider these examples:

• If your product isn’t HIPAA-compliant, you can’t enter the U.S. healthcare market.
• Without a Spanish localization, your reach to Spanish-speaking audiences will be limited.
• If your product costs $10,000 but targets small businesses, your SAM may shrink because many can’t afford it.

SOM (Serviceable Obtainable Market)

SOM represents the portion of SAM you can realistically capture in the short term (1–3 years), considering current market dynamics, competition, your go-to-market efforts, and available resources.

SOM = SAM × Market Share Coefficient

Where Market Share Coefficient = the percentage of the market you expect to capture, based on your marketing and sales strategy, budget, and other factors.

SOM is the tactical subset of SAM, focused on what's achievable given current conditions. For example, back to the fisherman analogy:

• TAM is the whole ocean.
• SAM is the part you can reach.
• SOM is the actual fish you expect to catch based on factors like fish behavior, weather, your team's skill, and competitor activity.

Even with a substantial SAM, a limited initial marketing budget and a narrow focus will naturally constrain your SOM.

It depends on multiple factors, including:

• Marketing Strategy: 
  What is the plan for customer acquisition? Paid ads, targeted engagement on Reddit, word-of-mouth?
• Sales Efforts: 
  How effectively will your team work with clients?
• Budget: 
  How much money do you have for promotion and product improvement?
• Competition: 
  How quickly will competitors react to your moves?
• Demand: 
  Will there be demand for your product in the current economy?
• Product:
  Will your product consistently meet or exceed customer expectations in terms of features, reliability, and ease of use?
• Support:
  How effectively will your support team keep customers happy?
• Team Dynamics: 
  How well will your team collaborate, communicate, and execute across different departments to achieve shared goals? Are those shared goals clearly defined?

Why Estimate TAM, SAM, SOM?

Estimating TAM, SAM, and SOM isn't just a number-crunching exercise. Here’s why you should invest the time in understanding these metrics:

1. Strategic Decision Making
   Knowing the size of your market at different levels helps inform key business decisions, from product development to resource allocation and market entry strategies.
2. Investor Confidence
   Investors want to see that you’ve thought through the potential of your market and can realistically capture a portion of it. Having solid TAM, SAM, and SOM estimates backs up your business plan and growth projections.
3. Sales & Marketing Focus
   Understanding which segments of your market you can realistically target allows you to direct your sales and marketing efforts more effectively.

How to Estimate TAM, SAM, SOM

There are two main approaches to market size estimation: Top-Down and Bottom-Up.

Top-Down Approach

The Top-Down approach starts with broad market data and narrows it down to smaller segments using logical filters and assumptions.

How it Works:

1. Find a Big Number: 
   Start with a broad, relevant market statistic from a credible source (e.g., reputable research agencies or government data).
2. Apply Filters: 
   Narrow that number down based on factors relevant to your specific market or product. For example:
   1. Filter by industry.
   2. Filter by geography.
   3. Filter by technology type.
   4. Filter by customer size.
3. Result: 
   The remaining number is your estimated market size (often used for TAM or SAM).

Pros:

• Quick: 
  Provides a fast, rough estimate, especially useful for TAM.
• Leverages Existing Data: 
  Relies on established (though sometimes expensive) research.
• Good for Context: 
  Helps understand the overall market landscape and your potential place within it.

Cons:

• Assumption Heavy: 
  The quality of the estimate depends entirely on the validity of the filters and the accuracy of the source data.
• Can be Generic: 
  May not reflect the nuances of your specific niche and doesn’t account for how your product fits within the market.
• Risk of Overestimation: 
  It’s easy to become overly optimistic with the filters, potentially inflating the estimate.

The top-down approach provides high-level market context and is best for initial market size estimates. However, investors may question such estimates unless you can prove they accurately reflect your target market.

Bottom-Up Approach

The Bottom-Up approach builds estimates from the ground up using internal data and smaller-scale research.

How it Works:

1. Identify Customer Segments: 
   Define your specific target customer profiles.
2. Count Potential Customers: 
   Estimate the number of potential customers in each segment you can realistically reach, given your go-to-market capabilities.
3. Estimate Revenue per Customer: 
   Determine the average revenue you expect from each customer segment (e.g., ARPU (Average Revenue Per User) for B2C, ACV (Average Contract Value) for B2B, etc.). This should be based on your pricing model and expected usage/purchase frequency.
4. Multiply and Aggregate: 
   Multiply the number of customers by the average revenue per customer for each segment, then sum across segments.
5. Result: 
   This gives you a market size estimate, typically best suited for calculating a realistic SOM and validating SAM.

Pros:

• More Realistic & Defensible: 
  Based on your specific product, pricing, target customers, and sales potential.
• Actionable: 
  Directly tied to your sales strategy and go-to-market plan, helping set concrete targets.
• Good for Niche Markets: 
  Works well when broad industry data isn't available or relevant.

Cons:

• Data Intensive & Time Consuming: 
  Requires significant effort to identify segments, count customers, and validate revenue assumptions.
• Risk of Underestimation: 
  Easy to miss potential customer segments or underestimate market penetration potential.

The bottom-up approach is best for calculating a credible SOM and validating SAM based on your specific product/service, business model, and go-to-market strategy.

It’s better not to rely on just one approach but to use both:

• Top-Down provides the theoretical maximum for market size.
• Bottom-Up offers a more grounded estimate based on what you can realistically target and achieve.

If the top-down and bottom-up estimates align reasonably well, your market size calculations gain credibility.

Calculating TAM

TAM can be measured either by the total number of potential customers or by the total revenue that could be earned from them.

TAM_customers = Total Number of Potential Customers

When calculating TAM based on revenue (which is generally preferable), you're looking at the potential total market size in monetary terms.

TAM_revenue = TAM_customers × ARPC (Average Revenue Per Customer)

Thus, to calculate TAM, you need to estimate two things: the total number of potential customers and the average revenue you could earn from each.

Step 1. Identify the Total Number of Potential Customers

Imagine you were a monopoly with 100% adoption: no limits, no competition, just you, the market, and groundbreaking success. Who would your customers be in this case?

If you're in B2B, consider all companies globally or within relevant geographical segments that experience the problem your product or service is designed to solve. For example, if you're offering a SaaS project management tool, your total potential B2B customers could include all companies globally that manage projects and require collaboration between teams.

You can search for such data in business directories and government databases (often free) or use platforms like:

• Crunchbase
• LinkedIn Sales Navigator
• ZoomInfo

Keep in mind that some data might be missing. In such cases, you can estimate the total by working with the available data. For example, if you know how many companies there are in a specific region, you can estimate the global number. If the U.S. had 50K companies in your industry and accounted for 30% of the global market, then the total worldwide would be 50K ÷ 0.3 ≈ 167K companies.

If you're in B2C, your potential customers would include all individuals globally or within relevant geographical segments who have the need or desire for your product or service. For example, if you're offering a fitness app, your TAM could include all health-conscious individuals, people interested in fitness, or those looking to improve their lifestyle.

Statista is a great source for market data, but key insights might be locked behind a paywall. You can also look for such data in government databases and public sources (though data on niche preferences may be limited or unavailable):

• Global: UN Data, World Bank, OECD Stats
• Europe: Eurostat
• USA: U.S. Census Bureau
• Country-wise: national statistics agencies

If no direct data is available, you can still gauge interest in your idea by analyzing relevant trends and keyword search volumes:

• Use tools like Google Trends, Exploding Topics, Pinterest Trends to spot what's trending.
• Use Google Ads Keyword Planner, Meta Audience Insights, Pinterest Ads Manager to estimate search volumes.

💡Tip: If your target market can be segmented by size, industry, geography, or other factors, estimate the total number of potential customers for each segment separately. This will give you a more accurate and actionable view of your TAM.

Step 2. Estimate the Average Revenue per Customer

In market sizing, ARPC (Average Revenue per Customer) typically refers to the average annual revenue you expect to earn from a single customer.

The exact metric depends on your business model:

• B2C → annual ARPU (Average Revenue per User)
• B2B → ACV (Annual Contract Value)

If you've been in the market for some time, you likely already have this data available. If you're feeling adventurous (and have a solid execution plan), you can also factor in potential revenue growth based on your historical performance when making your calculations.

Proxy ARPC

If you don't have any historical sales data yet, you'll need to use a proxy for ARPC. This could be based on industry benchmarks, competitor data, or estimates from similar products or services in the market.

For example, you could use Average Spending per Customer if you know what customers currently spend on alternatives or substitutes to solve the problem your product/service addresses.

Alternatively, you could (with caution) estimate ARPC based on Willingness-to-Pay (WTP) if you’ve conducted a survey to gather such data. However, keep in mind that survey-based WTP may overstate actual purchase behavior due to hypothetical bias (i.e., people may claim they’ll buy, but their actual purchasing behavior may differ).

Another option is competitor data. If you can't find any benchmarks for ARPU in your industry but a major competitor reported $400M in annual revenue from 2M users, you could estimate ARPU as $400M ÷ 2M = $200.

Step 3. Calculate TAM

To calculate your TAM, multiply the number of potential customers by the average revenue per customer for each segment, then sum across all segments.

This will give you a pretty handsome number.

However, keep in mind that TAM is an aspirational figure. It doesn't account for the messy realities of competition, demand distribution, your own limitations, or regulatory constraints, among other factors.

What's more, this estimate can quickly drift into pure fantasy.

And that's where inflated TAM comes into play.

Inflated TAM

"My product is for humans, there are 8 billion people on Earth, so that’s my TAM." - an example of an overly aspirational founder.

While exaggerated, this is a common way to inflate TAM.

Industry-Wide Assessment

Let's say you're selling VPNs.

You come across a Deloitte article stating, "The global market for security products is growing rapidly and is expected to reach US$200 billion by 2028, reflecting its critical role in securing digital initiatives."

So, you decide your TAM is $200 billion.

I hope this assumption raised some questions.

This approach overlooks critical factors, starting with product focus. The security market covers a wide range of products (antivirus, firewalls, etc.), not just VPNs.

All-Users Assessment

Imagine you're developing an app.

Statista says, "The number of smartphone mobile network subscriptions worldwide reached almost seven billion in 2023, and is forecast to exceed 7.7 billion by 2028."

You plan to sell your app for $1, so you estimate your TAM as 7.7 × $1 = $7.7 billion.

Unless your app is absolutely essential to every single smartphone owner on the planet, this estimate is wildly unrealistic.

On the other hand, The Startup Owner's Manual by Steve Blank and Bob Dorf suggests you shouldn't limit TAM at all, arguing that the TAM for a mobile app could include all smartphone owners worldwide.

Optimism vs. Realism

Some might argue that a bigger TAM is more impressive to investors, and being optimistic is always a good thing, right?

Others might say this reflects unrealistic expectations and a weak understanding of your true target audience.

Realistic TAM

It's hard to say what a realistic TAM is because it depends on your product and market. But here are some things to think about.

Focus on Your Product and Target Audience

Let’s continue with the VPN example.

If you’re developing a VPN for personal use (B2C) rather than businesses, it makes sense to focus on data specific to the consumer segment.

For instance, Verified Market Research states, "The consumer VPN market was valued at USD 51.1 billion in 2024 and is projected to reach USD 73.8 billion by 2031, growing at a CAGR of 4.7% from 2024 to 2031."

This is a more realistic estimate than using the $200 billion figure for the entire security market, as it accounts for the specific consumer VPN segment. Of course, without fully understanding Verified Market Research’s methodology, it's impossible to say whether this estimate is inflated or underestimated.

Look at Industry Numbers

Look at real numbers from your industry (or adjacent ones).

For example, if you’re evaluating the potential of a new mobile app, research the most popular ones. TikTok, for instance, has 1.12 billion monthly active users. This number can serve as a benchmark for understanding the adoption scale in the mobile app industry.

Of course, you might believe your app will be the best and most popular, but data shows the actual volumes you’ll be (optimistically) dealing with. This helps you set more realistic expectations and ambitions.

You might argue that you’ll create a whole new industry, one where no benchmark fits, and that you’ll set the trends and define the numbers.

But consider this: OpenAI reached 400 million weekly active users as of March 2025. ChatGPT was released in November 2022, so in less than 3 years. What are the odds you’ll surpass their growth? (Unless you have unlimited funding and a flawless plan, in which case, you probably wouldn’t be reading this article 😉).

So yes, I'm a big proponent of seeking out benchmarks and realistically assessing your assumptions.

Nothing beats a benchmark built from your own historical data, but if you're just getting started, industry averages can offer a helpful starting point.

Balancing Ambition with Reality

Limiting your TAM early on reduces the risk of unrealistic expectations, making your estimate more reliable.

On the other hand, limiting TAM may underestimate your potential. If your product or business model proves revolutionary and drives rapid growth, limits based on current data may not capture future expansion.

The key is to have a logical rationale for your TAM estimate.

If you can reasonably explain why everyone on the planet will buy your product, awesome!  Just make sure it makes sense.

Calculating SAM

SAM, like TAM, can be measured either by the total number of potential customers or by the total revenue that could be earned from them.

SAM_customers = TAM_customers × Market Accessibility Coefficient

Where:

• Market Accessibility Coefficient = the percentage of the total market you can realistically target and serve. It takes into account factors like how well your product or service fits the target customers, geographic constraints, legal restrictions, and other barriers to market entry.

To estimate this, you can use analytical reports, industry studies, or government databases that offer insights into how your market is divided based on these factors.

For example:

• Let's say your TAM is 100 million customers.
• Your product is focused on small businesses, and you've found data that 40% of TAM are small businesses. You can sell it only in three countries, which make up 50% of this segment.
• SAM = 100 million × 0.4 × 0.5 = 20 million customers

If you have multiple customer segments, your market accessibility coefficient may differ across them. In that case, it’s helpful to calculate SAM for each segment individually and then sum them.

Alternatively, if you use platforms like LinkedIn Sales Navigator to research your target audience, you can directly filter your target audience based on your specific criteria and see how many potential customers remain. This can give you a straightforward SAM without needing to apply calculations or coefficients.

So, leverage the data you have and avoid overcomplicating things when an easier solution is available.

For monetary value, SAM can be calculated as:

SAM_revenue = TAM_revenue × Market Accessibility Coefficient

or

SAM_revenue = SAM_customers × ARPC

Continuing from the example above:

• Your SAM in customer count = 20 million
• Let's assume your average revenue per customer = $100 per year
• Then SAM = 20 million × 100 = $2 billion annually

Calculating SOM

Like SAM and TAM, we can measure SOM either by the total number of potential customers or by the total revenue that could be earned from them.

SOM can also be calculated using both top-down and bottom-up approaches, depending on the data you have available.

Bottom-Up:

SOM_customers = Customers you realistically expect to capture

SOM_revenue = SOM_customers × ARPC

Top-Down:

SOM_customers = SAM_customers × Market Share Coefficient

SOM_revenue = SAM_revenue × Market Share Coefficient

Where:

• Market Share Coefficient = the percentage of the market you expect to capture, based on your marketing and sales strategy, budget, and other factors. This is often an educated estimate, informed by industry benchmarks, historical data, or competitor performance.

Estimating realistic market share is tough.

The best way to make your projections solid is to base them on pilot launch data, conversion testing, historical performance, or industry benchmarks. Avoid relying on pure guesswork.

For example, would you believe me if I said I’ll capture 5% of the chatbot market? Just trust me, I have a big plan! Probably not.

But what if I showed you hard data from a statistically valid primary study and a pilot program proving that 1% of the target population is willing to switch to my product with ABC features for $X price in the next 3-6 months?

Or better - what if I told you that last year my market share was 1%, here are my operational metrics, NPS scores, and this year I expanded my team with experts in critical areas and increased my marketing budget by 30%? Based on this, I project my market share to grow to 3% this year by expanding into new regions, improving customer retention, and increasing marketing efficiency. With the right funding next year, I plan to scale these efforts further and reach 5% market share.

That’s the difference between guessing a market share coefficient and backing it with data.

Now, for the sake of continuity, let’s go on with the calculation from the previous example:

• Your SAM was 20 million customers.
• You plan to capture 5% of the market in the first year (ambitious but realistic if backed with data).
• Then SOM = 20 million × 0.05 = 1 million customers
• Average annual revenue per customer = $100 per year
• Thus SOM,$ = 1 million × 100 = $100 million annually

Sources

Forget just asking AI (it's great for brainstorming but not always spot-on with the facts). To accurately calculate your market size, you’ll need to do some real digging.

Here's a breakdown of the best places to look.

Government & Public Databases (Free)

Raw data on all sorts of demographic & economic indicators:

• Global: UN Data, World Bank, OECD Stats
• Europe: Eurostat
• USA: U.S. Census Bureau, Small Business Administration (SBA)
• National Business Registries & National Statistics Agencies.

Reports from Research Companies (Free & Paid)

These provide market sizes, trends, and sometimes even customer spending insights:

• Gartner, IDC, Forrester, McKinsey, BCG, Deloitte, IBISWorld, Euromonitor, Nielsen (look for free report previews).
• Statista - a mix of free and paid data covering a broad range of industries.

💡 Use Google Advanced Search to specify your search queries.

Company Aggregators (Free & Paid)

• LinkedIn Sales Navigator: Use advanced filters to find companies by industry, location, size, and job titles.
• Crunchbase: Great for tracking startups.
• Apollo and ZoomInfo: For detailed company data.

Other Sources

• Reddit, Twitter, LinkedIn Groups: See what real people say about your industry.
• Surveys (Survey Monkey, Qualtrics, Typeform): Ask potential customers directly.
• Review Aggregators (G2, Capterra, TrustPilot, etc.): See who your competitors are and what customers are saying about them.
• Company Reports (SEC 10-K filings, investor calls): If your competitors are public companies, you can find a lot of information in their financial reports.

This list gives you a solid starting point.

The key is to be resourceful and combine different sources to get the most accurate picture possible.

Additional Tips

Estimating TAM, SAM, and SOM isn’t easy. It involves tons of data, assumptions, and hypotheses.

Here's how to stay in control:

1. Document your assumptions.
   Track the logic behind your segmentation and coefficients to ensure transparency and consistency.
2. Cross-check data sources.
   Comparing insights from multiple sources helps reduce errors and build a more accurate picture.
3. Validate with real data.
   Use pilot launches or surveys to test conversions and refine your SOM based on actual market response.
4. Account for uncertainty.
   Build scenario analyses (optimistic, pessimistic, base case) to create flexible estimates that reflect real-world risks.

And please, don’t just say, "Statista estimates the ABC market at $500 billion, and my product will capture 10% in year one because my ChatGPT wrapper is the best ever. Trust me!"

Using AI for TAM, SAM, SOM Analysis

An unpopular opinion:

Generative AI won't give you accurate numbers.

To craft a perfectly accurate prompt for estimating TAM, SAM, and SOM, you need:

1. A strong understanding of your market
2. A well-defined business model for your product
3. A significant amount of time dedicated to writing and refining the prompt itself

Even with all this effort, you’ll still end up with numbers you can’t be 100% certain about unless you manually verify them later. And that, again, takes time.

On the other hand, Generative AI with deep research capabilities can optimize the search for relevant data. It can help you aggregate and analyze large datasets quickly.

Here's a prompt template to get you started. Feel free to use, improve, or share it as you see fit.

Don't blindly trust the numbers AI chatbots provide, but leverage the methodologies and resources they suggest.

Final Thoughts

Estimating TAM, SAM, and SOM is often a messy, iterative process.

Perfect numbers are elusive, but well-reasoned, data-backed estimates are invaluable. Don't get lost chasing unrealistic perfection. Focus on building a logical case, documenting your assumptions, and using these insights to navigate your path forward.

Hopefully, this guide helps you do just that.

P.S. I'd like to hear your thoughts! How do you approach market sizing calculations? Top-down, bottom-up? What works for you? What doesn't?

Disclaimers:

This blog is all about sharing insights and ideas. While I do my best to provide accurate information, I’m not offering legal or financial advice. If you need expert guidance, please reach out to a qualified professional.

I am not affiliated with any of the tools mentioned in this post, and none of the links are referral links. My recommendations are based solely on their relevance, usefulness, and my personal experience using them.