Why Territory Design Breaks at $15M

In Issue 002, I introduced four layers of a revenue operating system. In Issue 003, I went deep on signal infrastructure, the detection layer that feeds the system.

Today I want to show what happens when the system produces a decision. Specifically, a territory decision, because territory design is where most revenue organizations first realize their architecture is broken.

Something happens to EdTech companies between $10M and $15M ARR. The territory model that got them there stops working. The VP Sales designed the territories in a spreadsheet three years ago using account count and geography. It was good enough when they had five reps. Now they have twelve, and three things have gone wrong.

First, the territories are unbalanced. The top rep has a book of 60 accounts with $4M in potential. The newest rep has 25 accounts with $800K. Both carry the same quota. One will overperform. The other never had a chance. Leadership will call this an execution problem. It is a design problem.

Second, territory design is disconnected from capacity planning. The territory was carved, the quota was assigned, and nobody validated whether the territory can actually support that number. When the rep misses, the territory gets restructured at the next annual planning cycle. That is a 12-month feedback loop on a problem that should be detected in real time.

Third, the inputs are wrong. Territories are designed around account count and geography. They should be designed around account potential, expansion whitespace, renewal risk, and procurement complexity. A territory with 40 accounts where 15 are in active procurement evaluation and 8 have critical renewal risk is a fundamentally different territory than one with 40 accounts where the book is stable. The spreadsheet does not know the difference.

In a revenue operating system, the scoring engine (Layer II) feeds the decision engine (Layer III). The decision engine does not just flag accounts at risk. It models the consequences at the territory level.

If three accounts in a territory are scored as critical renewal risk, the decision engine calculates the ARR at risk, the save cost, and the projected territory yield if those accounts churn. That territory just went from “on track” to “structurally underperforming” before a single deal was lost.

The first shift is designing territories around scored account potential instead of volume. An account scoring 85 on expansion readiness with $400K in whitespace is not the same as an account scoring 30 with no expansion path. The territory model needs to know the difference.

The second shift is connecting territory design to quota validation. Every quota should be tested against the territory it sits on. If the scored potential of the territory cannot support the number at historical conversion rates, the quota is wrong, the territory needs to be restructured, or both.

The third shift is moving from annual to continuous. Territory health should be a computed metric that updates as account scores change, reps turn over, and market conditions shift. The annual planning cycle is a symptom of not having the infrastructure to do it in real time.

I wrote the full framework on territory design and coverage — the cost of imbalanced territories, the math behind potential-weighted balancing, and the structural changes that move teams past the $15M wall.

Next week: the board deck problem, and why your forecast confidence number is the most important metric nobody is computing.