Market sizing and territory sizing get treated as the same exercise. They are not. A TAM figure describes an opportunity, and a territory describes what one person can work. The path between them runs through three filters, and skipping any of them produces territories that look well designed and are functionally uncovered.
What is the difference between TAM and territory size?
TAM is a dollar figure for a market. Territory size is an account count bounded by rep touch capacity. Converting one to the other requires narrowing the universe three times.The sequence:
| Layer | What it contains | What the filter removes |
|---|---|---|
| Total addressable market | Every company that could theoretically buy | Nothing, this is the baseline |
| Serviceable market | Companies matching your segment, geography, and tech requirements | Removes companies outside your segment, geography, and technical fit |
| Qualified account universe | Serviceable accounts passing your ICP filters | Removes accounts failing ICP filters |
| Territory | Qualified accounts a rep can touch at required frequency | Removes accounts beyond rep touch capacity |
The last step is where most models break. Teams filter down to a qualified universe of 4,000 accounts, divide by 20 reps, and assign 200 accounts per territory. If the motion requires four meaningful touches per account per year and a rep can produce 500 meaningful interactions a year, the real ceiling is 125 accounts. The other 75 are on a list and receive nothing.
How do you build the qualified account universe?
Filter the serviceable market on the attributes that historically predict a closed-won outcome, using your own data. Firmographic guesswork produces lists that inflate territory potential.Pull three to five years of closed-won accounts and identify which attributes separate them from closed-lost and from never-engaged. Common separators in B2B SaaS: employee band, presence of a specific adjacent system, department headcount for the buying function, and funding stage or growth rate.
Then apply those filters to the serviceable market. The output is an account list where every entry has a defensible reason to be there, and each carries an expected value rather than a flat weight.
Expected value per account = addressable contract value x segment win rate x expected engagement rate
That third term is the one teams omit. An account you can qualify but cannot reach contributes zero, so engagement rate belongs in the model.
How many accounts can one rep actually cover?
Divide the rep's meaningful interactions per period by the touches per account the motion requires. Both numbers are measurable, and neither is a matter of opinion.Meaningful interactions means customer-facing contacts that advance an account: discovery calls, demos, executive meetings, substantive multi-threaded email exchanges. It excludes one-way outreach that gets no response.
A worked example for a mid-market motion:
- Selling hours per month after internal obligations: 96 - Average hours consumed per meaningful interaction including preparation and follow-up: 1.5 - Meaningful interactions per month: 64 - Interactions per quarter: 192 - Touches an account needs per quarter to progress: 6 - Accounts a rep can genuinely work per quarter: 32
That 32 is the active working set, not the territory. The territory can be larger, because accounts rotate in and out of active work across the year. A reasonable territory size is the active working set multiplied by the number of cycles per year the account base turns over, which for most mid-market motions lands between three and five.
How do you check whether the territory is oversized?
Look at activity coverage across the account list, not at the potential total. An account with no touches in two quarters is not in a territory in any operational sense.Rank each territory's accounts by expected value and pull activity data for the bottom half. If the bottom half shows no meaningful activity across two quarters, that portion of the territory is uncovered and its potential should be removed from the quota model.
This connects to a broader pipeline hygiene problem. ORM sees 10 percent or more of a typical pipeline untouched for twelve months, using a definition of meaningful activity that means a change in stage, close date, or amount. Territories carry the same disease. Accounts sit on lists, contribute to the potential score that drove the quota, and receive nothing.
Two fixes. Cut the tail from the territory and route it to a pooled or partner motion, or reduce the quota to match the portion of the book that actually gets worked. Doing neither means the quota was set against revenue nobody is pursuing.
How should territory size vary by segment?
Inversely with deal complexity, because complexity consumes touch capacity per account. A single company-wide account-per-rep number produces overloaded enterprise reps and underloaded commercial ones.Calculate every number in this relationship from your own interaction data and stage-progression rates, because touch requirements differ by product and by how much of the buying process is self-serve. What holds across segments is the direction, not a set of values.
| Segment | Touches needed per account per quarter | Accounts per active working set | Reasonable territory size |
|---|---|---|---|
| Enterprise | Highest | Smallest | Smallest, named accounts |
| Mid-market | Moderate | Moderate | Moderate |
| Commercial | Lowest | Largest | Largest |
How does territory sizing connect to the forecast?
Territory size determines how much pipeline a rep can generate, which sets the ceiling on what the territory can forecast. Sizing errors surface as coverage gaps in specific books long before they surface in the company number.Convert each territory's size into an expected pipeline contribution: qualified accounts x engagement rate x opportunity creation rate x average deal size. Compare that against the pipeline the territory's quota requires. ORM's benchmark for pipeline coverage is 3x to 5x with most customers near 3.5x, so a $1M territory quota at 3.5x needs $3.5M of pipeline generated from that account list.
If the territory cannot produce it, the options are a bigger book, more marketing support into that book, or a lower quota. Assigning the quota anyway does not create the pipeline.
Track sales velocity by territory once the design is live. It shows whether the sizing assumptions held, since a territory sized correctly should show stable cycle times and stable deal sizes rather than a slow degradation as the rep spreads thinner.
Frequently Asked Questions
How many accounts should a single territory contain?
The number a rep can touch at the frequency the sales motion requires. Work backward from meaningful interactions per quarter divided by required touches per account. That number is usually far smaller than the addressable account list.
Can you size territories directly from a TAM figure?
No. TAM is a dollar figure covering companies you cannot reach and cannot serve. Convert it to a serviceable account list first by filtering on segment fit, geography, and product qualification, then size territories from that list.
What conversion rate should the sizing model use?
Your own closed-won rate by segment over at least eight quarters. Industry averages produce territories that look balanced on paper and are unbalanced in practice, because conversion varies more between companies than between segments.
Should territory size change as a rep gains tenure?
Coverage capacity grows with tenure since experienced reps qualify faster and waste less time. Adding accounts to a tenured rep's book is defensible. Removing accounts from a new rep's book to fund it is where most redesigns go wrong.
How do you avoid oversized territories?
Check whether the low-priority tail of the account list is receiving any touches at all. If the bottom half of a territory has no activity in two quarters, that half is uncovered, and the territory is oversized regardless of what the potential model says.
See how ORM turns these insights into action
ORM builds custom revenue forecast models for B2B SaaS companies. Not dashboards. Prescriptive analytics that tell you what to do next.
Schedule a Demo