Table of contents
TAM, SAM and SOM
Founders rarely fail because the market is "too small." They fail because they build and hire against a market story that isn't real, then discover too late that acquisition costs, sales cycles, or buyer budgets don't support the plan.
TAM (total addressable market) is the maximum annual revenue your company could generate if you captured 100% of the customers who could reasonably use your product, at the prices you assume, within a defined market category.
TAM is not your forecast. It's a boundary line that forces clarity about: who the buyer is, what you charge, and what "counts" as your market.
TAM is the outer ring of three nested market sizes, and this article covers all three:
- SAM (serviceable addressable market) is the part of TAM you can actually serve today, given your product, geography, integrations and compliance.
- SOM (serviceable obtainable market) is the part of SAM you can realistically win in a defined time window, given your channels, sales capacity, win rate and competition.
| Question it answers | Limited by | Best used for | |
|---|---|---|---|
| TAM | How big could this get? | Market definition and price | Category choice, investor context |
| SAM | How much can we serve today? | Product, geography, compliance, integrations | ICP focus, roadmap priorities, ARR ceilings |
| SOM | How much can we win in the next few years? | Reach, win rate, sales capacity, competition | ARR targets, hiring and budget plans |
Put simply: TAM is about possibility, SAM about feasibility and SOM about probability.
What TAM reveals (and what it doesn't)
TAM is useful because it answers a handful of high-stakes questions quickly:
- Is this market large enough to justify the product and go-to-market motion?
- Are we building a niche business (great) or accidentally trapped in one (not great)?
- Do our pricing and packaging choices leave money on the table?
- Are we defining the market in a way that matches how buyers actually purchase?
What TAM doesn't tell you:
- How fast you can grow (that depends on distribution and execution)
- Whether you will win (competition and differentiation matter)
- Whether you can access the market (that's what SAM and SOM answer, covered below)
The Founder's perspective
Use TAM to prevent category mistakes: building an enterprise sales team for a market that only supports SMB pricing, or pricing like SMB when the buyer is a regulated enterprise with real budget.
How to calculate TAM without fooling yourself
There are three common approaches. In SaaS, the most defensible is usually bottom-up.
Bottom-up TAM (most practical)
Count the customers in your defined market, then multiply by realistic annual revenue per customer.
If you sell subscriptions, "annual revenue per customer" should reflect your expected pricing and packaging, often approximated by ARPA (Average Revenue Per Account) or ASP (Average Selling Price) – but using market-appropriate values, not your best-case enterprise deals.
Example (B2B SaaS, tight ICP):
- Target buyers: US e-commerce brands with $5M–$100M GMV
- Count of eligible brands: 48,000
- Expected annual contract value for that segment: $6,000/year
So TAM ≈ $288M per year (as a revenue pool), under these assumptions.
Founders mess this up by letting one variable silently drift:
- "Potential customers" becomes "all businesses with a website"
- "Annual revenue per customer" becomes "our highest tier price"
Top-down TAM (useful, but risky)
Start with an industry number (research report), then assume a share.
This is fast, but it often breaks because:
- Reports mix software + services
- Category boundaries are vague
- You inherit someone else's assumptions about adoption and pricing
Top-down is best used as a sanity check, not your primary model.
Value-theory TAM (pricing-led)
Estimate the economic value you create and what portion you can charge for.
This helps when:
- You're creating a new category
- Pricing is usage-based or outcome-based (see Usage-Based Pricing)
- Your product replaces headcount or reduces risk
But value-theory only works if you can defend willingness-to-pay with real evidence (pilots, win/loss notes, procurement feedback).

SAM (serviceable addressable market)
SAM (serviceable addressable market) is the part of TAM you can realistically serve with your current product and go-to-market. TAM ignores most constraints; SAM is where they become real: who you sell to, where you can sell and support, which integrations and compliance requirements you meet, and what that segment can pay.
A bottom-up SAM looks like the TAM build, except every filter is a hard "we cannot sell without this" constraint:
If you think in monthly terms, average annual revenue per account is roughly ARPA × 12, using realized prices after discounts rather than list price.
Example (workflow tool for agencies):
- Agencies in your target geographies: 120,000
- With 10 to 200 employees: 38,000
- Using a billing tool you integrate with: 24,000
- With the problem you solve (multi-client approvals): 15,000
- Realistic annual revenue per account after discounts: $6,000
A $90M SAM tells you things TAM never will. A $30M ARR goal would mean winning a third of every serviceable account, which is a category-domination plan. To get there without heroic share, you need to grow SAM (a new segment, geography or use case) or raise what each account pays.
SAM moves when your capabilities move: adding a language or region, supporting the dominant system of record in your niche, getting SOC 2 or SSO, or repackaging so you can serve smaller or larger customers profitably. It shrinks when you narrow your ICP or raise your minimum price, which is often a good trade if retention improves.
Keep two versions: SAM today (what you can sell now) and SAM after roadmap (what specific features or certifications unlock, with dates). Mixing them is the most common way SAM gets inflated. SAM is also the denominator in the required-share check further down.
SOM (serviceable obtainable market)
SOM (serviceable obtainable market) is the portion of SAM you can realistically capture within a defined time window, given your channels, sales capacity, win rate, pricing and competition. It always needs a timeframe: "150 mid-market logistics customers in 36 months" is a SOM, "there are 50,000 logistics companies" is not.
The quick version is a share of SAM:
That share is easy to write and hard to defend, so build SOM bottom-up from your funnel instead:
Example: 2,000 target accounts over 36 months. You can reliably get 60% into a real buying conversation or trial, 35% of those are a true fit, you win 20% of those, and ARPA is $800 per month.
The winnable slice is about 84 customers, or roughly $806k ARR. Now the question is concrete: does that justify your burn and hiring plan, and if not, which input can you change?
In sales-led motions, check capacity too. Wins per year can't exceed the number of deals your team can actively run (limited by sales cycle length) multiplied by your win rate. Many plans size demand but not throughput.
SOM grows when you narrow the ICP and win faster, find a repeatable channel, shorten the sales cycle, add proof (case studies, security, integrations) or make expansion reliable. It shrinks when competition intensifies, churn rises in the segment or procurement slows deals down. A simple sensitivity check on the two inputs you can move fastest, usually win rate and ARPA, shows which lever closes the gap to your ARR target.
What drives TAM (the levers)
TAM is a model made of assumptions. Your "number" is only as good as the inputs. In SaaS, TAM usually moves because of changes in market definition, price, or eligibility.
1) ICP and eligibility filters
Your TAM grows or shrinks when you change who qualifies:
- Company size (employees, revenue, GMV)
- Industry (vertical focus)
- Tech stack (Shopify vs custom, Salesforce vs HubSpot)
- Geography and language
- Compliance requirements (HIPAA, SOC 2, GDPR)
A common pattern: founders start broad, then narrow after learning who retains well (see Cohort Analysis and Retention). That usually reduces TAM but improves business quality.
2) Pricing and packaging
Your pricing model directly scales TAM if customer counts stay constant.
"Attach rate" matters when not every eligible customer will buy (for example, only companies with a certain workflow pain).
Pricing changes that affect TAM:
- Moving from a $49 plan to $99 (if demand holds)
- Adding an enterprise tier that raises ASP (Average Selling Price) in a segment
- Switching to per-seat pricing (see Per-Seat Pricing) that scales with customer size
Be careful: raising "price" in the TAM model without evidence is the easiest way to create fantasy markets. Use willingness-to-pay signals and Price Elasticity thinking, even if it's directional.
3) Product scope (use cases)
Adding a real second use case can expand TAM more than a dozen features. Examples:
- You move from "billing alerts" to "revenue workflow automation"
- You add a second buyer persona (finance + revops)
- You support an adjacent vertical with similar needs
This is also where TAM and positioning can get sloppy. If the buyer wouldn't search for you or budget for you under that broader category, your TAM expansion is probably premature.
4) Market maturity and adoption
Even if a customer "could" use your product, they may not buy because:
- The category is new; budgets don't exist yet
- Switching costs are high
- The status quo is good enough
These adoption constraints should usually be reflected by moving from TAM to SAM and SOM, rather than inflating TAM.
How founders use TAM in real decisions
TAM is most valuable when it forces tradeoffs: where to focus, how to price, and what growth motion is viable.
Decide your GTM motion
A small TAM with high ACV can still be great – if your sales motion matches.
Use TAM together with:
- Sales Cycle Length
- CAC (Customer Acquisition Cost)
- CAC Payback Period
- ARR (Annual Recurring Revenue) targets
Practical interpretation:
- If your TAM depends on thousands of small accounts, you need a scalable channel (PLG or efficient inside sales).
- If your TAM depends on a few hundred large accounts, you need enterprise discovery, procurement readiness, and strong retention economics.
The Founder's perspective
TAM should change your hiring plan. If your realistic SAM supports only a $20M business, hiring a VP Sales and building a 10-rep team "for the future" is usually a cash burn trap.
Sanity-check revenue targets with required share
A simple check prevents years of mismatch between ambition and reality.
If you need an implausible share of SAM, you have three options:
- Expand SAM (new segment, geography, use case)
- Raise achievable price (through packaging or moving upmarket)
- Reduce targets (or accept a smaller outcome)
Example
| Item | Value |
|---|---|
| Target ARR (5–7 years) | $50M |
| TAM (broad category) | $1.2B |
| SAM (your ICP + geos) | $250M |
| Required share of SAM | 20% |
A 20% share might be possible in a winner-take-most market, but in most SaaS categories it's a red flag unless you have a clear distribution advantage.
Pressure-test pricing strategy
TAM is a good forcing function for pricing because it highlights where revenue pool is coming from.
If your TAM only looks big when you assume low price and massive volume, you're betting on:
- cheap acquisition channels
- low support costs
- low churn
That's a fragile plan if your product requires high-touch onboarding or if retention is mediocre (see Logo Churn and NRR (Net Revenue Retention)).
Conversely, if your TAM depends on high price, you need:
- a credible ROI story
- proof you can sell and retain at that price point
- lower churn tolerance (because each lost logo is expensive)
Guide roadmap and expansion bets
A TAM model can tell you where expansion is worth doing:
- New vertical: adds X customers at Y price
- New geography: adds X customers but increases compliance and support cost
- Moving upmarket: fewer customers, higher ASP, longer sales cycle
This pairs well with unit economics frameworks like LTV (Customer Lifetime Value) and LTV:CAC Ratio.

When TAM breaks (common mistakes)
Most TAM errors come from one of these failure modes:
Mixing TAM with "anyone who could"
If your definition includes customers who could theoretically use it but would never budget for it, you're counting "interest" instead of "addressable."
Fix: define addressable as "has the problem, has the budget, and has a path to purchase."
Double-counting revenue pools
This happens when you add multiple segments that overlap (for example, counting the same company in "mid-market" and "Shopify Plus," or counting multi-product bundles twice).
Fix: ensure segments are mutually exclusive, then sum.
Using list price instead of realized price
Realized price includes discounts, downgrades, and packaging reality (see Discounts in SaaS). A TAM based on list price will overstate revenue pool if your market expects discounts.
Fix: model price as what you can consistently win at for that segment, not your price page.
Confusing big TAM with easy growth
A market can be enormous and still hard:
- entrenched incumbents
- high switching costs
- long procurement cycles
- fragmented buyers
Fix: treat TAM as space available, then use SAM/SOM plus pipeline and retention data to judge accessibility.
The Founder's perspective
If your TAM slide makes you feel safe, you're probably using it wrong. A good TAM model makes you uncomfortable in specific ways – by revealing distribution constraints and forcing a focused wedge.
How to keep TAM useful over time
You don't need to "track" TAM weekly. You do need to version it like a strategic model.
Update TAM when:
- You change pricing/packaging materially
- You add a new buyer persona or segment
- You expand geography
- You move upmarket/downmarket
- Win/loss evidence contradicts assumptions
A simple TAM review cadence
- Quarterly: sanity check assumptions (counts, price points, exclusions)
- Annually: rebuild from scratch with new learnings and refreshed data sources
The signals that justify expanding TAM
Don't expand TAM because you want a bigger story. Expand because reality supports it:
- Higher ASP achieved repeatedly in a segment (see ASP (Average Selling Price))
- Retention holds at the higher price (see GRR (Gross Revenue Retention) and NRR (Net Revenue Retention))
- Sales cycle remains manageable (see Sales Cycle Length)
- CAC and payback still work (see CAC (Customer Acquisition Cost) and CAC Payback Period)

Practical benchmarks and rules of thumb
These aren't hard rules, but they prevent common founder mistakes:
- If TAM only works at perfect pricing, it's not addressable. Assume competitive pressure and discounting exist.
- If required share of SAM is above ~10–15%, be extra skeptical. You may still win, but you need a clear distribution edge or a wedge that expands over time.
- If your SOM depends on outbound scaling, validate sales efficiency early. Watch Win Rate and Sales Efficiency alongside CAC/payback.
- If TAM looks huge but retention is weak, the market might not value the product. TAM doesn't fix churn (see Customer Churn Rate and MRR Churn Rate).
Keeping TAM, SAM and SOM consistent
The three numbers should nest: SOM sits inside SAM, which sits inside TAM. Keep them consistent by using:
- the same price assumption (realized, not list) at every level
- the same unit at every level (annual revenue or accounts, not a mix)
- exclusions that only ever narrow as you move from TAM to SAM to SOM
- an explicit time window for SOM
If these three aren't consistent, your strategy will be inconsistent too. When in doubt, start with a defensible bottom-up SAM, then expand outward to TAM and inward to SOM.
Key takeaway
TAM is not a vanity number. It's a strategic constraint: a quantified statement of who you're building for and what they can pay. Build it bottom-up, pressure-test the assumptions, and use it to make decisions about GTM, pricing, and focus, then keep it honest by versioning it as you learn. SAM and SOM turn that boundary into an operating plan: what you can serve today, and what you can win within your runway.
Frequently asked questions
TAM is everyone who could use a solution like yours, at the prices you assume. SAM is the portion you can actually serve given your product, geography, compliance, integrations and go-to-market. SOM is the portion of SAM you can realistically win in a defined time window with your budget, team and channels.
There is no universal cutoff, but you want a TAM that can support your long-term revenue ambition without requiring unrealistic market share. If you need 20 to 30 percent of your serviceable market to hit your plan, the market is likely too tight. Use TAM with SAM and SOM to sanity-check ambition.
Lead with SAM and SOM for credibility, and include TAM as context. TAM answers what is possible in the broadest definition, while SAM shows the slice you can actually serve with your product, geography, and compliance. Investors discount hand-wavy TAM, but they reward clear constraints and a realistic wedge.
For early-stage SaaS, bottom-up TAM is usually most defensible: count a specific buyer set and multiply by a realistic annual price for what you sell today. Top-down industry reports can be useful for context, but they often mix categories, include services, or assume adoption you cannot capture.
Yes, but treat it as a model update, not a monthly metric. TAM changes when your ICP expands, you add new geographies, move upmarket or downmarket, change packaging or pricing, or your product becomes applicable to new use cases. Document assumptions so the changes are explainable.
Run a market share sanity check: divide your long-term target ARR by your SAM, not TAM, and ask if the required share is plausible given incumbents, distribution, and switching costs. If you cannot describe a credible path to that share, your TAM likely includes customers who will not buy.