How do you determine the right number of sales reps for a new outbound territory in 2027?

What changes by company stage

The single biggest mistake in outbound territory sizing is applying a model built for one company stage to a company at a different stage. A Series A startup opening its first geographic territory and a $200M ARR company opening its fourteenth are solving different problems with the same vocabulary, and the arithmetic that works for one produces absurd answers for the other.

At the pre-product-market-fit and early-seed stage — typically under $2M ARR — you should not be sizing a territory at all. You are still testing whether outbound works as a motion. The right answer at this stage is one or two founder-adjacent sellers running against a hand-built list, and the number you are trying to determine is not "how many reps" but "does a rep who is not the founder close anything." Territory boundaries at this stage are noise. Sizing formulas assume a known conversion rate, a known average contract value, and a known sales cycle length; a pre-PMF company has none of these with enough confidence to divide by. Applying a capacity model to guesses produces a headcount plan with three decimal places of false precision.

From roughly $2M to $15M ARR, you have one repeatable motion and are opening territory two, three, or four. Here the constraint is almost never total addressable market — it is management bandwidth and enablement capacity. A single frontline manager can meaningfully coach six to eight outbound reps; below four reps a manager is underloaded and usually still carrying a bag, above eight the coaching becomes triage. So a new territory at this stage is sized in units of "pod": two to four reps plus a fraction of a manager, plus whatever SDR support the motion requires. The model gives you a number; the pod structure rounds it.

From $15M to $75M ARR, you have enough historical data that the capacity model becomes genuinely predictive, and the constraint shifts to account supply. This is the stage where companies discover they have been carving territories that look balanced on a map and are wildly unbalanced in buyable accounts. A territory with 4,000 companies in it might contain 400 that match the ICP tightly enough to work outbound, and if your reps need 250 workable accounts each to hit quota, that territory supports fewer than two reps regardless of how big the map looks.

Above $75M ARR, the question changes again. You are usually not asking "how many reps" but "what is the marginal revenue of rep number N+1 in this territory, and is it above the fully-loaded cost of that rep plus the cannibalization of existing reps' pipeline." At this stage territory sizing becomes a portfolio allocation problem, and the honest answer often involves shrinking a territory that has been over-covered for two years rather than adding to a new one.

The practical implication: before you run any formula, name your stage out loud and name the constraint that stage actually faces. Running a $50M-stage capacity model at $3M ARR is the single most common way companies over-hire into a territory and then spend a year unwinding it.

Stage-by-stage playbook

The sequence below is deliberately ordered so that the cheap, reversible checks happen before the expensive, irreversible ones. Every step gates the next.

Step one — validate the account supply. Before any quota math, count the accounts. Pull the territory boundary you are proposing, apply your ICP filters at the strictness you actually use in practice, and count what survives. Not "companies in the region" — companies that match employee-count band, industry, technographic signal, and whatever else your closed-won cohort actually shares. If a broad filter gives you 6,000 accounts and a real ICP filter gives you 380, your territory supports the 380 number.

Step two — establish accounts-per-rep from your own history. Take your last four quarters of closed-won deals. For each rep, count the distinct accounts they touched meaningfully — sequenced, called, engaged — not the accounts sitting in their CRM ownership field. Most outbound teams find the honest number is 150 to 400 active accounts per rep per year depending on ACV and cycle length. High-ACV enterprise motions land near the low end; SMB velocity motions near the high end. Divide your surviving account count by this number and you have a supply-side rep ceiling.

Step three — run the quota-capacity model. Take the revenue you need from the territory, divide by fully-ramped rep quota, then inflate for ramp and attrition. This gives you a demand-side rep count. It will usually be higher than the supply-side ceiling from step two, and when the two disagree, the supply side wins. You cannot conjure accounts.

Step four — apply the pod floor. Never launch a new outbound territory with one rep. One rep gives you a sample size of one on every question you will need to answer — is the messaging wrong, is the territory wrong, or is the rep wrong? With two or three reps you can distinguish rep variance from territory reality within two quarters. The cost of the extra rep is far lower than the cost of a year of ambiguous data.

Step five — stage the hiring against ramp. Do not hire the full number on day one. Hire the pod floor, let them ramp, and gate the next hires on a specific signal — typically the first two reps each generating a defined number of qualified opportunities per month, sustained for a full quarter.

Numbers that matter at each stage

Every input below should be pulled from your own data before you borrow a benchmark. Benchmarks are for sanity-checking a number you already computed, never for generating one.

Fully-ramped quota per rep. This is the denominator of the whole model, and it should be the quota reps actually attain, not the quota you assign. If you assign $800K and median attainment is 72%, your planning denominator is closer to $576K. Planning against assigned quota is how a territory ends up 30% under plan while every individual rep looks "on track for a good year."

Ramp time. Outbound ramp is longer than people budget for, because outbound reps must build pipeline before they can convert it. The rep is not productive when they close their first deal — they are productive when they close deals at a steady-state rate. Practically, ramp is roughly one sales cycle for pipeline generation plus one full sales cycle for conversion. A 90-day cycle implies six months to steady state. Budget the first two quarters of a new rep at a fraction of full quota — a common structure is 25% in quarter one, 50-60% in quarter two, 85-100% by quarter three — and carry that discount in the capacity model rather than pretending it away.

Attrition. Outbound sales roles carry meaningfully higher voluntary and involuntary turnover than most functions. If you plan for exactly the number of reps you need, you will be under-staffed within two quarters because the model assumed everyone stays. The standard fix is a coverage multiplier: plan headcount at 110-125% of the raw capacity number depending on your observed attrition. This is also why the pod floor matters — losing your only rep in a one-rep territory zeroes it out.

Accounts per rep. The number that most often breaks territory plans. It varies enormously by motion, but the discipline is constant: derive it from your own reps' actual working behavior, not from a CRM ownership count. A rep who "owns" 900 accounts and meaningfully touches 220 has an accounts-per-rep number of 220.

Sales cycle length. Determines how long you must fund a territory before you can judge it. A territory with a 120-day cycle cannot be evaluated at 90 days, and evaluating it anyway is how good territories get killed. Set the evaluation gate at cycle length plus ramp, not at a calendar quarter boundary.

Conversion rates by stage. You need at minimum: contact-to-meeting, meeting-to-opportunity, and opportunity-to-close. New territories almost always convert worse than mature ones for the first two to three quarters — no brand recognition, no referenceable local customers, no rep familiarity. A common planning adjustment is to discount new-territory conversion by 20-30% relative to your mature-territory baseline for the first year, then reassess.

Fully-loaded rep cost. Base plus variable at target, plus benefits, plus tooling, plus the fraction of management and enablement cost that rep consumes. This is often 1.3 to 1.5 times OTE. You need this number to answer the marginal-rep question, and it is the number most commonly understated in a territory business case.

Worked example. Suppose you need $3M from a new territory in its first full year. Fully-ramped attained quota is $600K. Raw demand is five reps. Apply a new-territory conversion discount of 25% and the effective requirement rises to roughly 6.7, call it seven rep-years of capacity. But hires ramp — if you hire five reps in month one, each delivers roughly 60-70% of a rep-year in year one, giving you about 3.3 rep-years, not five. Meanwhile the supply side: the territory has 1,400 ICP-qualified accounts and your accounts-per-rep is 300, giving a ceiling of about 4.7 reps. The honest conclusion is that this territory cannot deliver $3M in year one under any staffing plan, and the correct output of the exercise is not a headcount — it is a renegotiated revenue expectation of roughly $1.8M to $2M in year one with a path to $3M in year two. Producing that renegotiation, rather than a number that satisfies the plan, is the actual job.

Where territory sizing goes wrong

Certain failure modes recur across companies and stages with enough regularity that they are worth checking for explicitly.

Sizing to the plan instead of to the market. Someone decides the territory must produce $4M because the board deck says so, works backward to a rep count, and hires it. The market does not consult the board deck. When the plan number and the supply-side ceiling disagree, the ceiling is a fact and the plan is a hope. Escalating the disagreement early is uncomfortable and correct; discovering it in month nine is expensive and career-limiting for everyone involved.

Counting accounts, not buyable accounts. A territory that "has 8,000 companies" tells you almost nothing. The question is how many are in the ICP, currently reachable, not already customers, not in an active competitor contract with 30 months remaining, and large enough to clear your minimum deal size. Subtracting existing customers and active opportunities alone often removes 15-25% of a mature-adjacent territory.

Ignoring the SDR-to-AE ratio. If your outbound motion depends on SDRs, sizing AEs without sizing the SDR layer produces AEs with no pipeline. Common structures range from 1:1 to 1:3 SDR-to-AE depending on how much prospecting the AE does themselves. Decide the ratio explicitly, staff it, and include the SDR cost in the territory business case — otherwise the territory looks profitable on paper and starves in practice.

Over-hiring on a good first quarter. A new territory's first quarter is frequently inflated by low-hanging fruit: inbound leads that were already there, warm relationships a rep brought with them, a deal that was going to close anyway. Extrapolating a headcount plan from quarter one is how territories go from two reps to eight and back to three. Wait for two consecutive quarters of steady-state performance before scaling.

Under-hiring out of caution and calling it discipline. The mirror failure. A single rep in a large territory generates the appearance of thin results, the territory gets labeled weak, and it is abandoned while genuinely under-covered. Under-coverage and low potential produce identical dashboards. The pod floor exists specifically to break this ambiguity.

Treating territory lines as permanent. Boundaries drawn at launch are hypotheses. Plan an explicit review at two quarters and again at four, with named criteria — coverage ratio, opportunity creation per rep, win rate versus baseline. A territory redraw is a normal operational event, not an admission of failure, and companies that treat it as normal redraw earlier and cheaper.

Forgetting the manager. Reps in a new territory need more coaching, not less — the messaging is unproven, the objections are unfamiliar, and the reps have no local peers to learn from. If the new territory's reps report to a manager who is already at eight direct reports in a different region, the new territory is effectively unmanaged. Include management capacity in the sizing decision or accept a materially longer ramp.

Decision framework

When every input is in front of you, the decision itself compresses into a small number of gates. The framework below is written to be run in a single planning session with the data already assembled, and it is intentionally biased toward "not yet" — the cost of an unnecessary quarter of delay is almost always lower than the cost of unwinding an over-hired territory.

Run the gates in order and stop at the first failure. Each gate has a specific remedy, and the remedy is almost never "hire anyway and hope."

The coverage-ratio gate asks whether the territory has enough qualified accounts to support at least the pod floor. If not, the remedy is structural: merge the territory with an adjacent one, widen the ICP deliberately and re-test, or defer the territory entirely. The economics gate asks whether the expected first-year contribution from the marginal rep exceeds their fully-loaded cost within your payback tolerance — for most companies that tolerance sits somewhere between 12 and 24 months for a new territory, longer than for a mature one because the first cohort is paying for market development the second cohort will inherit. The management gate asks whether a manager has real capacity for these reps. The enablement gate asks whether territory-specific messaging, references, and objection handling exist yet — if the reps will be inventing the pitch themselves, expect ramp to run 50% longer and staff accordingly.

Only after all four gates pass do you commit the headcount, and even then you commit it in tranches with a named signal gating each release.

Related questions

How long should you wait before judging a new territory?

Ramp time plus one full sales cycle, at minimum. For a 90-day cycle with six-month ramp, that is roughly nine months before the data means anything. Judging earlier measures ramp, not territory quality.

Should the first reps in a new territory carry a reduced quota?

Yes. New-territory reps face longer cycles, no local references, and unproven messaging. A ramped quota schedule stepping from roughly 25% to full over three quarters is standard, and the shortfall belongs in the plan, not in the rep's comp risk.

Do you need SDRs to open an outbound territory?

Not necessarily, but you must decide explicitly. Full-cycle AEs can open a territory at lower ratios of output per head; SDR-supported models generate more meetings but cost more per territory. Pick one and staff it fully rather than half-staffing both.

What if the account supply says one rep but the plan needs four?

The supply number wins. Four reps in a one-rep territory will cannibalize each other's accounts, burn the list in two quarters, and produce four under-performing reps instead of one strong one. Renegotiate the plan or expand the boundary.

How do you split a territory once it outgrows its reps?

Split on account count and pipeline value, not geography alone, and split before reps hit capacity rather than after. Protect in-flight opportunities with a transition period so no rep loses a deal they sourced.

FAQ

How many reps should a brand-new outbound territory start with?

Two to three in almost every case. One rep gives you no way to separate rep performance from territory quality, and losing that rep zeroes the territory. Four or more before any validated signal risks burning the account list and over-committing spend to an untested boundary. The pod of two to three is the smallest configuration that produces interpretable data.

Which number matters more, quota capacity or account supply?

Account supply. Quota capacity tells you how many reps you would need to hit a revenue target; account supply tells you how many reps the territory can actually feed. When they disagree, capacity is a wish and supply is a constraint. Hiring to the capacity number in a supply-constrained territory produces reps competing for the same accounts.

How do you determine accounts-per-rep for your own team?

Look at the last four quarters. For each rep, count distinct accounts they meaningfully engaged — not CRM ownership counts, which are inflated by dormant records. Take the median across reps who hit quota, since you want the working number of a successful rep, not the average across performers and strugglers.

What coverage multiplier should you apply for attrition?

Plan headcount at 110-125% of raw capacity, calibrated to your own observed outbound attrition rather than a generic figure. If you lost three of twelve outbound reps last year, your multiplier lives at the higher end. The multiplier compensates for the gap between a funded seat and a productive person occupying it.

Should a new territory's revenue target be lower than a mature one's?

Yes, and materially so for the first year. New territories carry no brand recognition, no local references, unproven messaging, and reps in ramp. Discounting first-year conversion by 20-30% against your mature-territory baseline is a reasonable starting assumption, refined once you have two quarters of real data from the territory itself.

When should you stop adding reps to a territory?

When the marginal rep's expected contribution stops clearing their fully-loaded cost within your payback tolerance, or when accounts-per-rep drops below your working threshold — whichever comes first. The second usually arrives before the first, and it shows up as reps reporting that they are running out of accounts worth sequencing.

Sources

• https://hbr.org/2012/07/how-to-really-motivate-salespeople
• https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights
• https://www.bain.com/insights/topics/sales-and-channel-effectiveness/
• https://www.salesforce.com/resources/research-reports/state-of-sales/
• https://www.gartner.com/en/sales
• https://www.bls.gov/ooh/sales/sales-managers.htm
• https://www.saastr.com/category/sales/
• https://openviewpartners.com/blog/

Related on PULSE

• [How do you calculate the right SDR-to-closed-won ratio for a new market entry in 2027?](/knowledge/gp552)
• [Top 10 metrics to track in the first 30 days of a new outbound motion in 2027](/knowledge/gp533)
• [How do you design a territory plan for a 10-person sales team in 2027?](/knowledge/gp556)