How to design territory carve-up after a 50% headcount expansion in 2027

The scenario: 40 reps becoming 60 by Q2

Picture a mid-market SaaS org that closed 2026 with 40 quota-carrying AEs and a board-approved plan to reach 60 by the end of Q2 2027. That is not a hiring problem — the recruiting team can fill 20 seats. It is a territory problem, and it lands on the RevOps Director's desk roughly six weeks before the first cohort's start date.

The instinct is to treat it as arithmetic. Forty territories, twenty more mouths, so slice each existing book by a third and hand out the remainders. That instinct is wrong for a specific and expensive reason: the existing 40 books were never balanced to begin with. They accreted. Rep A inherited a book from someone who left in 2024. Rep B got a geographic carve that made sense when the company sold to a single vertical. Rep C quietly hoarded three accounts that generate 40% of their number and would never appear in a fair redistribution. Slicing an unbalanced map into more pieces produces a more unbalanced map with more pieces.

The second instinct — full clean-sheet redesign, every account back in the pool, optimizer picks winners — is wrong for a different reason. It maximizes theoretical balance and destroys relationship equity. Tenured reps who lose their best accounts start interviewing. You will spend the savings from a perfectly balanced map on replacing the people who were hitting their number.

The workable path sits between those poles. Freeze first so the map stops moving while you model it. Re-score so you are clustering on current potential rather than 2024 assumptions. Cluster on weighted opportunity, not account count. Protect a small, explicitly-named set of crown-jewel relationships for tenured reps. Then push the expansion into whitespace, where new hires are not competing with an incumbent's relationship history and where their activity actually creates net-new revenue rather than reshuffling existing coverage.

One framing that helps executives commit: the carve-up is not a fairness exercise. It is a capacity-utilization exercise. You just bought 50% more selling capacity. The map determines whether that capacity points at addressable demand or at accounts someone already worked six months ago.

How the sequence actually runs

The mechanism has five stages and they are strictly ordered. Skipping or reordering any one produces a map that gets relitigated in week six.

Freeze. The day offers go out, the existing map is locked in the CRM. No moves, no splits, no "just this one account" exceptions. This is unpopular and non-negotiable — every mid-flight adjustment invalidates the dataset you are about to model against, and the modeling takes two weeks. Publish the freeze as a written policy with the CRO's name on it, and give the Deal Desk an exception path that requires the CRO's signature rather than a VP's. In practice you will get three to eight exception requests in the first week; approving more than one or two signals the freeze is soft and the requests multiply.

Score. Pull a 24-month account history from the warehouse — or from CRM native if no warehouse exists — and re-score every account against a current ICP. Six columns are mandatory before any clustering begins: account ID, weighted TAM, fit score, trailing two-year revenue, open pipeline, and last-touch date. Teams that skip weighted TAM produce pods that look balanced on account count and vary by 3x on actual potential, which is precisely the failure the whole exercise exists to prevent.

Cluster. Run the optimizer to produce candidate maps constrained on geography, named-account legacy ownership, vertical specialization, language coverage for international coverage, and a tenure floor for complex segments. Generate five to ten candidates, present three to the CRO. Presenting one candidate reads as a fait accompli and invites the CRO to reject the whole model; presenting ten reads as indecision.

Assign. New hires land on roughly 70% net-new whitespace and 30% inherited install base with an active renewal or expansion event in the next nine months. That 30% is deliberate — a pure-whitespace book gives a ramping rep no near-term wins and no reference customers, and their first two quarters look like failure even when the activity is right.

Ramp-gate. Quota steps up over roughly six months rather than landing at full load on day one. A common shape is 25% of full quota in the first quarter of tenure, 50% in the second, 75% in the third, 100% thereafter. The exact curve matters less than the fact that it exists and is written into the comp plan rather than living as a verbal understanding between a rep and their manager.

The ordering constraint worth internalizing: design the pods before you assign the people. Teams that start from "who deserves what" and back into boundaries end up with a map that encodes politics. Teams that build balanced pods first and then match reps to pods have a defensible answer when someone asks why their book shrank — the pod is balanced, here is the weighted TAM, here are the constraints it satisfied.

Numbers that make or break the model

Quota math is where most 50% expansions quietly fail, because the carve is done against the quota number instead of the pipeline number required to hit it.

The working formula is straightforward: quota = rep capacity × ramp factor, and the territory must contain enough addressable pipeline to support that quota at your actual coverage ratio. If your historical win rate implies you need 3.5x pipeline coverage, then a rep carrying a $1M number needs $3.5M of qualified pipeline sitting inside their boundaries. Carving to the quota number rather than the pipeline number is the single most common comp mistake in expansion scenarios — it produces territories that are technically "worth" the quota in TAM terms but cannot generate enough at-bats to convert it.

Balance tolerance is the second number worth arguing about. Pods balanced within a tight band on weighted TAM — single-digit percentage variance — behave materially better than pods balanced on account count. The mechanism is intuitive: account count treats a 40-employee prospect and a 4,000-employee prospect as equivalent units, and no seller does. If your optimizer cannot enforce a TAM band, enforce it manually by sorting pods on total weighted TAM and hand-swapping accounts at the edges until the spread closes.

Attainment is the number the CRO actually cares about. Industry medians for quota attainment in B2B SaaS have sat in the 40-50% range across segments for several years — meaning roughly half your team misses in a normal year. A 50% headcount jump without a territory rebuild pushes that down, because the new reps are competing for the same demand the existing reps were already working. The rebuild's entire economic justification is recovering those points.

Ramp economics deserve their own line. A fully-loaded AE costs somewhere between their OTE and roughly 1.5x that once you count benefits, tooling, enablement, and management overhead. Twenty new AEs at, say, $200K fully loaded is $4M of annual cost that produces meaningfully less than full-quota output for two to three quarters. That gap — the ramp deficit — is why the 70/30 whitespace/install-base split matters. The 30% install-base slice is what generates first-quarter wins that keep a ramping rep employed long enough to become productive.

Data quality has a price too. Third-party revenue estimates from major enrichment providers routinely disagree by wide margins for the same account. The practical answer is not to find the true number; it is to pick one source and lock it for the entire planning cycle. A consistently wrong TAM estimate still produces a balanced map. Two inconsistent sources produce a map nobody trusts.

Score decay is the last quiet number. A fit score from six months ago is less predictive than an intent signal from last month. Re-score on roughly a monthly cadence during the rebuild and quarterly afterward. Most optimizer platforms support time-decay weighting and most teams leave it at default, which silently degrades cluster quality over the year.

Trade-offs, and the models you could run instead

There is no single correct carve. There are four or five defensible ones, and the right choice depends on what your product actually sells against.

Geographic carve is the oldest model and still correct for anything with a field-sales motion or drive-time component. It is trivially explainable — reps understand a map — and it minimizes travel cost. It fails when demand density is uneven, which in SaaS it almost always is. A Bay Area territory and a Mountain West territory are not comparable books no matter how you draw the lines.

Vertical carve concentrates domain expertise and usually improves win rates in complex sales, because the rep who has closed eleven credit unions is materially better in the twelfth conversation. The trade-off is fragility: if your strongest vertical hits a budget freeze, one or two reps absorb the entire shock while everyone else is fine. Vertical carves also make new-hire onboarding slower, since domain fluency takes a quarter or two to build.

Named-account carve works at enterprise scale where the account list is short and each logo is worth pursuing for years. It produces the cleanest accountability and the worst flexibility — reassigning a named account is a political event, not an operational one.

Hybrid pods — a geographic or vertical base with a named-account overlay for the top logos — is where most 50%-expansion orgs land, because it lets you protect crown jewels while still redistributing the long tail.

The round-robin / no-territory model deserves mention because it keeps resurfacing. Leads route to whoever is available, no boundaries at all. It genuinely maximizes speed-to-lead and it eliminates carve-up disputes entirely. It also eliminates account ownership, which means nobody builds a multi-year relationship and expansion revenue suffers. It fits high-velocity, low-ACV motions and almost nothing else.

The adjacent trade-off most teams underweight: overlay coverage. If you are adding 50% more AEs, you are implicitly changing the ratio of AEs to SDRs, solutions engineers, and customer success managers. A 40-to-60 AE expansion with a static SDR team means each AE gets a third less prospecting support, and the whitespace-heavy new-hire books are exactly the ones that need it most. Model the supporting-function ratios in the same cycle, or the map will be technically balanced and practically unworkable.

Similarly downstream: post-sale ownership. If new AEs are landing net-new logos in previously uncovered whitespace, those logos need a CSM assignment that did not exist in the old model. Carving the AE map without carving the CS book pushes the imbalance one quarter into the future rather than solving it.

Where these rebuilds go wrong

Relitigating the map in week six. This is the most common failure and it is always a sign-off problem, not a modeling problem. Five roles need to sign: the CRO on policy, RevOps on data and modeling, the line VPs on rep-level assignments, the comp lead on quota calibration, and the Deal Desk lead on mid-quarter split rules. Skip any one and that person's objections arrive after go-live, when changing the map costs ten times more.

Stripping tenured reps to the studs. A naive equal-redistribution takes a tenured rep's best accounts and hands them to a new hire who cannot work them. The rep updates their LinkedIn within the quarter, and you have traded a proven producer for a marginal improvement in map symmetry. The fix is explicit and cheap: name a small number of protected accounts per tenured rep — a top-10 rule is a common shape — and put everything else in the pool. The retention math is not close. Replacing a productive enterprise AE costs a large multiple of whatever balance you gained.

No relationship-transition protocol on inherited accounts. When an account moves from rep A to rep B, the default outcome is silence — the buyer emails their old contact, gets no reply or a cold handoff, and the renewal wobbles. Build a 90-day transition SLA into the plan: two warm introductions, one joint call or QBR with both reps present, and a written context handoff covering champion, blockers, contract dates, and open commitments. The outgoing rep is not fully released from the account until those are done.

Publishing comp after go-live. Reps who receive a new territory before they receive the plan that pays them for it assume the worst, and they are frequently right. Plans should be live in the comp tool several business days before territories go live, with in-app countersignature so there is an audit trail. Email PDFs are not an audit trail. When a Q3 dispute reaches the CRO, "I never agreed to this quota" needs a documented answer.

Open-ended dispute windows. A 48-hour window with a named owner closes disputes. An open window means reps litigate boundaries all quarter instead of selling. Route disputes through the Deal Desk with RevOps and the line VP on every call, and escalate anything unresolved at 48 hours to the CRO for a written decision. No verbal exceptions — verbal exceptions become precedent, and precedent becomes the next map's baseline.

Treating go-live as the finish line. It is the start of a 30/60/90 cadence. At day 30, run a coverage check: has every account been touched at least once, and is every new hire hitting their ramp-activity target? Misses route to the line VP inside a day. At day 60, gate on pipeline creation — a ramping AE should have built pipeline meaningfully above their ramp-period quota by then. Well below target means a remediation plan with enablement; well above target may mean the territory is under-carved and should be flagged for the next optimizer run. At day 90, the CRO, RevOps, VPs, and comp lead sit for a two-hour map review with exactly three questions on the agenda: where is the map wrong, where is the quota wrong, where is the comp wrong. Adjustments commit within five business days of that meeting or they do not happen at all.

Forgetting that the map is a living object. Re-run the optimizer quarterly against fresh scores. Territories drift as accounts grow, churn, get acquired, and change segment. An annual carve with no quarterly correction is a map that is accurate on day one and progressively wrong for the next eleven months.

Related questions

Should we carve territories before or after the new hires start?

Before. The map should be final and comp countersigned five business days before the first cohort's start date. Onboarding a rep into an undefined territory wastes their first two weeks and signals disorganization at the exact moment you need their confidence.

How do we handle a rep who loses a deal that was mid-cycle when the map changed?

Split-credit rules, published with the map. The standard shape is full credit to the originating rep for opportunities past a defined stage at freeze date, shared credit for earlier-stage deals, and Deal Desk arbitration for anything ambiguous.

What if we can't hire the full 50%?

Carve for the plan, but build the pods so unfilled ones can be temporarily covered by a pooled or overlay motion rather than absorbed into adjacent books. Absorbed territories are nearly impossible to reclaim once a tenured rep has worked them for a quarter.

Does this change if the expansion is SDRs rather than AEs?

The sequence holds, but the balance metric changes from weighted TAM to addressable contact volume and account density. SDR territories should also be nested inside AE territories, not drawn independently, or the two teams work different maps.

How often should the map be reviewed after go-live?

Formally at day 90, then quarterly. Quarterly re-scoring with an optimizer re-run catches drift before it compounds. Anything more frequent creates churn; anything less means you are working a map built against year-old assumptions.

FAQ

How long should a territory rebuild after a large expansion take?

Roughly four weeks of concentrated work: one week to freeze and re-score, one to cluster and model candidates, one to assign reps and lock the mix, and one for quota calibration, comp publication, and the dispute window. Compressing it below three weeks usually means skipping the re-scoring step, which is the step that makes the whole exercise worth doing.

Who owns the redesign?

The CRO owns policy and final approval; the RevOps director owns modeling and data; line VPs own rep-level assignment calls; the comp lead owns quota calibration; the Deal Desk lead owns split rules and disputes. Five distinct sign-offs, all captured in writing before go-live.

Should new hires get whitespace or established accounts?

Both, weighted toward whitespace. A roughly 70/30 split of net-new to inherited install base gives ramping reps room to build without competing against an incumbent's relationships, while the install-base slice provides near-term winnable deals during the ramp period.

Why balance on TAM instead of account count?

Because account count treats every logo as an identical unit and no seller does. Two pods with 150 accounts each can differ by multiples in actual revenue potential. Balancing on weighted TAM produces territories that are comparably winnable, which is the only kind of fairness reps recognize.

What happens to tenured reps' best accounts?

Protect a named, limited set — a top-10 rule is common — and pool everything else. Full redistribution of a tenured rep's book is the fastest way to turn a retention problem into a hiring problem, and the balance gained rarely justifies the producer lost.

How do we prevent disputes from dragging on?

A hard 48-hour window with a named owner. Disputes route to the Deal Desk with RevOps and the line VP present; anything unresolved at 48 hours escalates to the CRO for a written decision. Verbal exceptions are prohibited because they become precedent.

Sources

• Bridge Group, SaaS AE Metrics & Compensation Report — https://blog.bridgegroupinc.com/saas-inside-sales-metrics
• Bridge Group, Sales Development Metrics & Compensation Report — https://blog.bridgegroupinc.com/sales-development-metrics
• RepVue, Sales Salary and Quota Attainment Data — https://www.repvue.com/blog/sales-salary-guide
• Fullcast, Guide to Territory Planning for Revenue Growth — https://www.fullcast.com/content/territory-planning/
• Anaplan, Sales Territory and Quota Planning — https://www.anaplan.com/solutions/sales-planning/
• Xactly, Sales Territory Planning Resources — https://www.xactlycorp.com/
• Harvard Business Review, Getting Beyond "Show Me the Money" (sales compensation) — https://hbr.org/2012/04/getting-beyond-show-me-the-money
• McKinsey, Insights on Sales and Growth — https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights
• Salesforce, Enterprise Territory Management Documentation — https://help.salesforce.com/s/articleView?id=sf.territory_mgmt_overview.htm

Related on PULSE

• [How to set capacity plans that match Series B headcount budgets in 2027](/knowledge/ra0319)
• [Territory Design for Vertical SaaS Sales in 2027](/knowledge/ra0218)
• [Territory Design for Geographic SaaS Sales in 2027](/knowledge/ra0217)
• [Post-Merger Territory Harmonization in 2027](/knowledge/ra0498)
• [Territory Carve Review Cadence in 2027](/knowledge/ra0430)
• [Territory Realignment Playbook for SaaS Sales in 2027](/knowledge/ra0224)