How do you design a territory plan for a 10-person sales team in 2027?

The go-to-market motion in one picture

Before you draw a single boundary, you need to see the whole motion the territory sits inside. A territory plan is not a map exercise — it is the physical layout of your demand. Every account that enters the funnel has to land in exactly one rep's book, and every rep has to have enough of those accounts to hit a number that the company has already committed to the board. If those two facts are not reconciled, the territory plan will fail no matter how elegant the boundaries look on a slide.

The motion for a 10-person team usually breaks into four stages: define the total addressable universe, score and segment it, allocate it into books, then run the coverage cadence that keeps the books honest through the year. Most teams do the first two well and then guess at the third. The guess is where the damage happens, because allocation errors compound — a rep who starts the year with 18% less opportunity than the median does not fall 18% short, they fall further behind as ramp, morale, and pipeline coverage all decay together.

Here is the flow from raw market to a working book of business.

Read the loop carefully. The two feedback edges matter more than the forward path. The first loop — balance check back to capacity model — is the one you run in November and December before the year starts, and it usually takes three or four iterations before the books stop looking lopsided. The second loop, quarterly coverage review back to balance check, is deliberately narrow. You are not re-cutting territories every quarter; you are checking whether reality has diverged far enough from the model to justify a targeted fix, like moving a handful of accounts or reassigning a book after attrition.

The upstream dependency people underestimate is data quality. If your CRM has duplicate accounts, missing employee counts, or parent-child hierarchies that were never mapped, the scoring step produces garbage and the allocation inherits it. For a 10-person team, plan on two to four weeks of account hygiene before the planning work itself. That means deduplication, appending firmographics from a data provider, and resolving corporate hierarchies so that a global parent and its five subsidiaries land with one rep instead of scattering across three books and triggering a credit dispute in month two.

The downstream dependency is marketing. If demand generation runs campaigns by industry and the territory design is geographic, every campaign creates an uneven lead spray — one rep drowning, three starving. The cheapest fix is aligning the segmentation axis: if marketing runs vertical plays, cut territories by vertical. This is one of the strongest arguments for named-account or industry-based design over pure geography in a modern software or services business, where a rep's physical location rarely constrains who they can sell to.

Who owns what across the revenue org

Territory design fails politically more often than analytically. The math is tractable; the ownership is not. For a 10-person team the org is small enough that roles blur, which is exactly why you should write the ownership down before the first draft circulates.

Sales leadership owns the design decision. The VP or director of sales makes the final call on the segmentation axis and signs off on the final books. This is not a delegated decision, because it directly determines who is likely to make quota. If a rep believes the boundaries were drawn by a spreadsheet nobody owns, every miss becomes a territory complaint instead of a performance conversation.

RevOps owns the model, the data, and the arbitration. This is the group that builds the capacity model, runs the balance analysis, maintains the account scoring, and — critically — adjudicates disputes during the year. RevOps should hold the master account-to-owner mapping as the single source of truth, synced into the CRM, with a change log. On a 10-person team RevOps may be one person, possibly part-time; the function still needs to exist by name so that "who decides" is never ambiguous.

Finance owns the quota envelope. The sum of the ten individual quotas has to exceed the company target by a coverage factor — commonly in the 10% to 20% range for a stable team, higher if you expect attrition or have unramped reps. Finance sets that envelope. Sales leadership distributes it across the books. If those two steps are done by the same person without a check, you get quotas that add up to exactly the company number, which guarantees a miss the moment one rep underperforms.

Marketing owns lead routing conformance. Once territories are published, the routing rules in the marketing automation platform and CRM must match the territory model exactly. A mismatch between the territory table and the lead-assignment rules is the single most common source of "I never got that lead" escalations. Audit the routing rules against the territory table the same week the plan goes live, then again 30 days later.

Customer success and partners own the overlap rules. In a 10-person sales org there is almost always a CS or account management function that touches the same accounts, plus possibly a channel. Rules of engagement need to say explicitly who owns expansion revenue in an existing account, who owns a partner-sourced deal in a rep's named territory, and what happens when a partner brings a deal in an account that a rep has been working for six months. Write the answer down. "We will figure it out case by case" is a promise of quarterly conflict.

The reps own their book plan. Each rep should produce a one-page plan per territory within two weeks of assignment: top 20 target accounts, why those, the coverage pattern, and the pipeline math that gets them to quota. This is the step that converts an allocation into a strategy, and it also surfaces allocation errors fast — if three reps independently say their top-20 list is thin, the design has a problem the spreadsheet did not show.

One structural note on adjacent motions. If the 10-person team includes SDRs or is supported by a separate SDR pod, the SDR territories should map cleanly onto the AE territories, ideally one-to-one or two-to-one. Splitting an AE's accounts across three SDRs produces incoherent outbound and duplicate touches into the same buying committee. The same logic applies to solutions engineers and any overlay specialist: align their coverage to territory boundaries, not to a separate map, or you build a matrix that nobody can staff.

Metrics, targets, and realistic ranges

The whole design exercise reduces to a handful of numbers you can actually check. Here is what to compute and roughly where healthy teams land — treat these as starting reference points to calibrate against your own historical data, not as universal constants.

Balance index. For each book, compute total addressable potential — usually modeled revenue opportunity, not just account count. Divide the highest book by the lowest. A spread above roughly 1.3x means your top rep is running with materially more raw opportunity than your bottom rep, and quota attainment will reflect the territory more than the talent. Getting under 1.2x is achievable for a 10-person team with reasonable data. Getting to 1.0x is not, and chasing it produces geographically incoherent books that cost more in travel and context-switching than the balance is worth.

Accounts per rep. This is a capacity question, and it varies enormously by motion. A high-touch enterprise motion with six-figure deals and long cycles supports far fewer accounts per rep than a transactional mid-market motion. The honest way to set this is to work backward from touch frequency: decide how many meaningful touches per account per quarter the motion requires, divide by the selling hours a rep actually has, and let the arithmetic tell you the ceiling. Reps do not get 40 selling hours a week — after internal meetings, CRM work, proposal building, and travel, the real number is often closer to half that. Build the model on the real number or the plan overstates coverage from day one.

Pipeline coverage ratio. Each territory should be able to generate enough pipeline to cover quota at your historical win rate with margin. If you close roughly one in four qualified opportunities, you need meaningfully more than 4x quota in pipeline to survive normal slippage — most teams target somewhere in the 3x to 5x range depending on stage definitions and win rate. The territory-relevant question is whether the accounts in a given book can plausibly produce that pipeline. If a rep's entire book, fully penetrated, yields less than their quota times the coverage ratio, the territory is mathematically unwinnable and no amount of activity fixes it.

Quota-to-territory-potential ratio. Divide each rep's quota by the modeled annual opportunity in their book. If a rep must win an implausibly large share of everything available in their territory to hit plan, the assignment is broken. This single ratio catches more bad territories than any other check, because it directly ties the number the rep is measured on to the market they were handed.

Attainment spread as the after-the-fact audit. At the end of the year, plot attainment across the ten reps. If the distribution correlates strongly with territory potential rather than with observable rep behavior — activity, meetings, multi-threading — the design carried more weight than performance. That is your evidence for the next cut.

Time and travel load. If the motion involves field visits, model drive time or flight time explicitly. Two reps with identical account counts and identical modeled revenue can have wildly different effective capacity if one covers a dense metro and the other covers four states. The usual correction is to give the sparse-geography rep a lower account count or a lower quota, and to be transparent about why.

Churn and installed-base weighting. Existing customers are not free revenue, but they are cheaper revenue. If books are cut without accounting for installed base, a rep who inherits 40 existing accounts and a rep who inherits four are not running the same job even at identical quota. Either weight installed-base accounts when computing potential, or split quota into new-logo and expansion components with separate targets. The second approach is cleaner for a small team, because it makes the mix explicit and prevents a rep from farming renewals all year while new-logo goals quietly die.

Ramp adjustment. Any rep under roughly six months of tenure should carry a ramped quota, and their territory should be sized for where they will be at month nine, not month one. A common failure is handing a new hire a full-size enterprise book because the map needed covering, then watching the best accounts in it go untouched for two quarters. If you have unramped reps, consider parking the top accounts from their book with a senior rep temporarily, with an explicit written handback date.

Where the motion breaks down

Territory plans rarely fail dramatically. They erode. Here is where, in rough order of frequency.

Stale data underneath a beautiful model. The account list was appended with firmographics 14 months ago. Companies have grown, been acquired, moved headquarters, or gone out of business. The territory model is a photograph of a market that no longer exists. Refresh firmographics before every annual cut, and spot-check 20 to 30 accounts manually against public sources — if more than a small fraction are wrong, fix the data before you cut anything.

Rules of engagement that exist verbally. The most expensive disputes are the ones with no written answer: a lead comes in from a division of an account owned by another rep; a partner registers a deal in a named territory; an inbound request arrives from a company that just crossed the segment threshold mid-quarter. Every one of these has a right answer, and the answer is cheap to write down in advance and expensive to negotiate under pressure with commission on the line. Publish a short rules-of-engagement document, name the arbiter, and set a resolution SLA measured in days.

Mid-year re-cuts. Re-cutting territories mid-year destroys pipeline continuity, breaks trust, and creates a commission mess for deals in flight. Do it only for genuine forcing events — a rep departs, a major acquisition changes the market, a segment collapses. When you must, protect in-flight opportunities: a common approach is to let the original rep retain credit on anything already at a late stage, with a defined transition window. Decide the rule before the situation, not during it.

Optimizing purely for revenue potential and ignoring effort. A book with enormous theoretical potential concentrated in six accounts that each require an 18-month enterprise cycle is not equivalent to a book with the same modeled potential spread across 60 mid-market accounts. Same number on the spreadsheet, completely different jobs. Weight for cycle length and deal complexity, or at minimum acknowledge the difference when setting quotas.

Whitespace invisibility. Most territory models count accounts and estimate revenue but never model penetration. In an installed base, the highest-yield opportunity is often expansion inside accounts you already own, and that opportunity is invisible if your model only counts logos. Add a whitespace layer: product gaps, seat gaps, and department gaps per existing account. For teams with a meaningful installed base, this frequently changes the allocation more than any new-logo scoring does.

Routing rules that drift from the territory table. Someone adds a new lead-source rule in the automation platform in March. It routes by state. The territory model is by industry. Six weeks later a rep escalates. Automate the check: a scheduled report that flags any account whose CRM owner does not match the territory table's expected owner, reviewed weekly by RevOps. It is a small job that prevents a recurring category of conflict entirely.

Hoarding. A rep sits on 300 accounts and works 25 of them. The other 275 are dead inventory that no one else can touch. Institute a coverage floor with real consequences: accounts with no meaningful activity in a defined window return to a shared pool for reassignment. Announce the rule at the start of the year, apply it consistently, and make the reassignment visible. The point is not punishment; it is preventing a slow leak of addressable market into nobody's hands.

Comp plan misalignment. The territory says "go win new logos in manufacturing." The comp plan pays the same rate on a renewal. Reps optimize for the comp plan, always. If the design and the plan disagree, the design loses. Review them together, in the same meeting, before either is finalized.

Ignoring the SDR and marketing feedback loop. Territories are usually designed once and then treated as fixed input. But if outbound response rates in one vertical are a fraction of another's, that is real information about territory difficulty, and it should feed the next cut. Capture it: track meetings-booked-per-100-touches by territory, and use the spread as a difficulty coefficient.

How to sequence the build

For a 10-person team, the full design cycle is realistically six to ten weeks of elapsed calendar time, though the concentrated work is far less. The sequence matters because several steps have hard dependencies — you cannot model capacity before you have clean data, and you cannot set quotas before you know what each book contains.

A few notes on the individual stages.

Data hygiene first, always. Deduplicate, append firmographics, resolve corporate hierarchies, and mark dead accounts. Every hour here saves several later. The specific check that catches the most problems: pull every account with a null employee count or null industry and either fix it or exclude it explicitly. Silent nulls end up in someone's book as invisible filler.

Choose the segmentation axis deliberately. The main options are geography, industry vertical, company size or segment, named accounts, and product line. Geography is simple and travel-efficient but often produces unequal books and clashes with vertical marketing. Vertical alignment builds real expertise and matches campaign structure, but concentrates risk — if one industry has a bad year, one rep eats it entirely. Named accounts give the most control and the cleanest fit for enterprise motions but require the most maintenance. Segment-based splits are clean for teams running distinct motions at different deal sizes. For a 10-person team in a software or services business, a hybrid is common: two or three reps on named enterprise accounts, the rest split by vertical or region across the mid-market.

Score and tier before allocating. A simple, defensible model beats a sophisticated opaque one. Fit signals — size, industry, tech stack, growth — plus engagement or intent signals, rolled into a score, then cut into A/B/C tiers. The tiers matter more than the exact score, because the allocation step balances tier mix, not raw score sums. A book with 12 A-accounts and one with three A-accounts are not equivalent even if total modeled revenue matches.

Balance on three axes simultaneously. Modeled revenue potential, account count, and effort load. Optimizing one at a time produces a book that is balanced on revenue and absurd on travel. Run all three and accept a compromise on each.

Involve the reps before you finalize, not after. Circulate draft books to the team with a defined window for structured feedback — specifically: which accounts do you have relationships in that the data does not show, and which accounts in your book do you believe are dead. Reps hold relationship knowledge that no data provider captures. Two rounds of this, tightly time-boxed, materially improves the final cut and buys enormous political goodwill. What it must not become is an open negotiation; be explicit that leadership makes the final call.

Load it into systems the same week you publish it. Territory tables in the CRM, lead routing rules, reporting filters, dashboards. A plan that lives in a spreadsheet is a plan that will be quietly ignored. Then verify: run a report of every account's owner against the territory table and confirm zero mismatches before you announce.

Set the review cadence, not the re-cut cadence. Quarterly reviews that look at attainment spread, pipeline coverage per book, and any accumulating disputes. Most quarters, the outcome is "no change." That is the correct outcome. The annual re-cut is where structural changes happen. If you find yourself making structural changes every quarter, the underlying problem is usually the data or the segmentation axis, not the boundaries.

Related questions

How often should territories be re-cut?

Once a year, aligned to the fiscal year, with a quarterly review that usually results in no change. Mid-year re-cuts should be reserved for forcing events like rep departure, acquisition, or a segment that materially collapsed — and always with a written rule protecting in-flight deals.

Should territories be geographic or by industry?

Depends on the motion. Geography wins when travel is a real constraint and deal sizes are similar. Industry wins when expertise compounds, marketing runs vertical campaigns, and reps sell remotely. Most 10-person teams end up hybrid: named enterprise accounts plus vertical or regional mid-market books.

How do you handle a rep leaving mid-year?

Redistribute their book temporarily rather than re-cutting everyone. Protect late-stage in-flight deals with credit rules defined in advance. Assign accounts to the reps with the closest adjacency and the most capacity, and set an explicit date to reassess when the replacement ramps.

What data do you actually need to build a territory model?

Clean account records with resolved hierarchies, firmographics (size, industry, location), installed-base and product data for existing customers, historical closed-won data by segment, and win rate by segment. Intent or engagement data helps prioritization but is not required for a first defensible cut.

How does territory design interact with the comp plan?

Directly. Quota is applied to a territory, so an imbalanced book distorts attainment and payout. Design the territories first, then set quotas against measured territory potential, then confirm the comp plan's accelerators and mix reward the behavior the territory was designed to produce.

FAQ

How many accounts should each rep have in a 10-person team?

There is no universal number — it falls out of the motion. Work backward: define the touch frequency the sales cycle requires per account per quarter, estimate realistic selling hours per week after internal overhead, and divide. Enterprise motions land in the dozens; transactional mid-market motions land in the hundreds. The wrong approach is dividing the total account universe by ten and calling it capacity, because that guarantees dead inventory in every book.

What is the biggest mistake in territory design?

Balancing on one dimension. Teams optimize for revenue potential, declare the books equal, and then discover one rep has six accounts requiring 18-month cycles while another has 60 that close in 45 days. Balance revenue potential, account count, and effort load together — and accept a compromise on each rather than perfection on one.

Do you need territory management software for a 10-person team?

Usually not for the design itself. A well-built spreadsheet plus your CRM's territory or account-owner fields handles ten books. Dedicated software earns its cost at larger scale, with complex overlays and multiple hierarchies. What you do need regardless is a single authoritative account-to-owner mapping with a change log, and automated routing rules that provably match it.

How do you keep reps from hoarding accounts?

Set a written coverage floor: accounts with no meaningful activity in a defined window return to a shared pool for reassignment. Announce it before the year starts, apply it consistently, and make reassignments visible. Pair it with a lower account count per rep — hoarding is frequently a symptom of books that were too large to cover in the first place.

Should new hires get the same size territory as tenured reps?

No. Size their book for where they will be around month nine, not month one, and pair it with a ramped quota. If the top accounts in a new hire's assigned region are too valuable to leave untouched during ramp, temporarily park them with a senior rep under an explicit written handback date rather than letting them go dark for two quarters.

How do you measure whether a territory plan worked?

Look at the attainment spread across the ten reps at year end and test whether it correlates with territory potential or with rep behavior. Strong correlation with potential means the design drove outcomes more than performance did. Also track disputes raised, accounts with zero touches, and pipeline coverage per book — those three catch problems long before December.

Sources

• https://hbr.org/2012/03/getting-beyond-show-me-the-mon (Harvard Business Review — sales compensation and territory incentives)
• https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights (McKinsey — Growth, Marketing & Sales insights)
• https://www.bain.com/insights/topics/sales-and-channel-effectiveness/ (Bain & Company — sales and channel effectiveness)
• https://www.salesforce.com/sales/territory-management/ (Salesforce — territory management overview)
• https://www.gartner.com/en/sales (Gartner — sales research and practice)
• https://hbr.org/2015/07/how-to-set-sales-quotas-that-work (Harvard Business Review — setting workable sales quotas)
• https://www.bcg.com/capabilities/marketing-sales/overview (BCG — marketing and sales capability)
• https://www.forrester.com/research/ (Forrester — B2B revenue research)

Related on PULSE

• [Developer-First Launch Playbook: API-First Product Adoption Without a Sales Team](/knowledge/gp0487)
• [How do you determine the right number of sales reps for a new outbound territory in 2027?](/knowledge/gp542)
• [What is the ideal SDR-to-AE ratio for a physical therapy software company in 2027?](/knowledge/gp569)
• [What is the go-to-market playbook for full-service restaurants in 2027?](/knowledge/gp0579)