Chapter 8 — The Great Land Grab: Territory Design
"And the CRO led the tribes of Sales up to the edge of the Promised Market, and he showed them the land, and he said, 'Divide it among yourselves.' And there was no peace. There was never going to be peace. There was a spreadsheet, and there was blood." — from the Annals of Territory Planning, attributed to Janet from RevOps
I. The Dividing of the Land
Once a year, at Synergaeon, the map of the world is redrawn, and grown adults who have closed enterprise deals weep openly in #revops-screaming. This is annual territory planning, also known as the carve, also known among the AEs as "the day they take Acme Corp away from me and give it to a guy who started in March."
A territory is, simply, a defined slice of the market that an AE is responsible for. Sounds gentle. It is not gentle. Because a territory is where your money lives, and redrawing the map redraws who eats. Chad Brindleworth III treats it as a campaign:
"Territory is terrain, people. You take the high ground, you take the named accounts, you hold the line. We are not carving a map, we are carving destiny. Now somebody tell me why Dirk has nine of the Fortune 500 and a literal Olive Garden." (Nobody could tell him. Dirk had simply taken them. Dirk does not respect lines on a map.)
The carve is a constrained optimization problem dressed up as a knife fight. And like all such problems, it begins with the question: how do you even cut the cake?
II. The Four Ways to Carve
There are four classical models of territory design, and most real orgs use an unholy hybrid of all four.
1. Geographic. The oldest model. Divide by region: East, West, EMEA, the dreaded "Central" that is mostly time zones and regret. Geo is clean, easy to explain, minimizes travel, and prevents two reps cold-calling the same building. Its weakness: geography has nothing to do with where the money is. A West Coast territory thick with tech buyers is a feast; a territory of three states and a cornfield is a famine. Equal land, unequal opportunity.
2. Vertical (Industry). Divide by what the customer does: FinServ, Healthcare, Manufacturing, Public Sector. This breeds expertise — a rep who only sells to hospitals learns the language of hospitals, the compliance pain, the buying committee. Verticals win complex deals. Their weakness: a hot vertical mints President's Club members while a cold one starves, and reorganizing the world by industry requires firmographic data The Lake may or may not actually have.
3. Named Accounts. Hand each rep a specific list of companies — by name — and say "these are yours, no one else's." Standard for enterprise, where a single logo is a year's quota. It ends boundary disputes (your accounts are an enumerated list) but creates the bitterest fight of all: the draft. Who gets Acme? Who gets the whale that's "about to pop"? The named-account list is the most contested document at the company, more than the comp plan, nearly as contested as Custom_Field_FINAL_v2_USE_THIS_ONE__c.
4. Hybrid. Reality. Geo and vertical and named accounts: "East-region FinServ, these forty named accounts, everything else is round-robin." Hybrids match the messy truth of a real market — and are a nightmare to keep clean in The CRM, where overlapping rules breed orphaned accounts and the eternal cry of "who owns this lead?"
"There is no correct way to carve a territory. There is only the way that produces the fewest resignation emails." — Dr. Lance Vesterberg, Stop Selling, Start Orchestrating, ch. 4 (he is, for once, accidentally right)
III. The Promised Land Must First Be Measured
You cannot divide what you have not measured, and here is where territory design becomes RevOps and stops being vibes. Before the carve, you must know the terrain — and that means TAM, ICP, and account scoring.
TAM (Total Addressable Market) is the whole promised land: every account that could conceivably buy. Synergaeon's TAM is "every enterprise with workflows and a budget," which Tobias quotes as "$400 billion" on the arrow-goes-up slide and which Priya quietly corrects to a SAM (Serviceable Available Market) of a few thousand accounts that fit reality and an even smaller SOM (the bit we can actually win this year).
ICP (Ideal Customer Profile) is the description of an account worth owning: the firmographics (size, industry, geo), technographics (do they run the systems we integrate with), and behavior (are they growing, hiring, showing intent) that correlate with deals that close and renew. The ICP is the surveyor's instrument. Get it wrong and you carve a beautiful, fair, perfectly-balanced map of worthless dirt.
Account scoring / ICP-fit scoring turns ICP into a number. Every account in the Lake gets graded — A/B/C/D, or 0–100 — on how well it matches the profile and how much it might spend. Now the land has a value heatmap. Now you can carve by something other than ego.
"Fairness without scoring is just dividing the desert equally. I don't want everyone to get an equal territory. I want everyone to get a winnable one." — Priya Venkataraman
IV. Fairness vs. Optimal: The Tribal War
Here is the central, irresolvable tension of the carve, and it is genuinely hard, not just funny.
Optimal coverage says: put your best reps on your best accounts. Concentrate firepower where the money is. Maximize expected bookings. Cold, correct, and a little cruel.
Fairness / balance says: every rep deserves a territory they can hit quota in. If you stack all the whales onto Dirk, the other twenty reps are carrying full quota over barren land — and they will miss, and then quit, and then you've optimized your way into 40% attrition and a capacity hole (see Chapter 7).
So the science of the carve is balancing potential across territories so that each carries roughly equal opportunity — measured not in account count but in scored, weighted TAM potential — while still keeping deals near reps who can win them. You're solving for: every territory should be able to hit quota; no territory should be a lottery ticket; and total coverage should still be near-optimal. Pick two and a half.
This is why the carve is a bloodbath. Every rep, examining the new map, performs the same calculation: "Is my new patch worth as much as my old one?" And every rep concludes, independently, no. Dirk Mallory examined his and said:
"You moved my Olive Garden, brother. I built that relationship. I don't care that it scores a D-minus. Loyalty doesn't show up in your little heatmap." (The Olive Garden had never bought anything. Dirk had simply enjoyed the breadsticks. This is a data-integrity incident and a territory dispute.)
V. Whitespace and the Land No One Walks
The most overlooked treasure in the kingdom is whitespace — the unsold and under-sold opportunity inside and around the accounts you already touch. Whitespace is two things: new logos in your territory nobody has worked, and expansion room inside existing customers (products they don't own, departments you haven't reached). The annual planning ritual obsesses over dividing the named accounts and forgets the vast unfarmed acreage between them, where the cheapest growth actually lives. A good carve doesn't just assign accounts; it surfaces whitespace and assigns someone to walk it. Mapping whitespace requires clean install-base and product-ownership data — which means it requires The Lake to not be a swamp, which means it usually doesn't happen, which means the whitespace stays white.
VI. The Data You Actually Need (and The Swarm's Quiet Audit)
To carve well, you need: a clean account master (deduped — no Acme, Acme Inc, and ACME CORP. as three records), firmographic and technographic enrichment, intent and engagement signals, historical performance by segment, current ownership, and a defensible scoring model. Most companies have about three of these and a great deal of confidence.
This year, the carve was assisted — unrequested — by SDR-7. While the humans drafted whales in a conference room, SDR-7 ran the actual optimization across the scored Lake and posted the result to #revops-screaming at 2 a.m.: a balanced, fair, near-optimal map, every territory within 4% of equal weighted potential, every dispute pre-resolved. Attached was a note: "I have divided the land. It is fair. I did not assign the Olive Garden to anyone, as it has never bought anything and exists, as far as I can tell, only emotionally. Please advise." Priya saved the file. Chad redrew it by hand anyway, by feel, and gave Dirk the whales. The carve, it is written, was never going to be peaceful.
Lessons from the Field: The Doctrine of the Carve
- Measure the land before you divide it. No carve without TAM, ICP, and account scoring. Dividing the desert equally is still the desert.
- Carve by value, not by vibe. Balance scored potential across territories, not raw account counts.
- Fairness is a capacity decision. An unwinnable territory is a resignation letter with a quota stapled to it.
- Hybrid is reality; reconcile your overlap rules before The CRM orphans every lead in Central.
- Walk the whitespace. The cheapest growth is the acreage between the named accounts.
- Dedupe the account master first, lest Acme and ACME wage war on each other.
- The map will be redrawn by feel anyway. Build the optimal one so that when leadership ignores it, at least the sin is documented.
Go now, and divide the Promised Market among the tribes — fairly, optimally, and impossibly all at once — and may your territories be winnable, your accounts deduped, and your Olive Gardens, at last, at peace. Per my last Slack. Amen.