The Hidden Value in Your Territories

Territory & Quota Planning

With Hassan Mahmood  

Hassan Mahmood on what AI should really be looking for in territory design.

Most territory conversations start with the accounts you already have. The more valuable conversation starts with the accounts you don't.

When teams talk about applying AI to territory planning, the default instinct is to optimize what already exists — rebalance the book, smooth out the workload, redraw a few lines on the map. That's useful. But it treats a territory as a fixed pool of known accounts, and that framing quietly caps how much value the exercise can return.

The more interesting opportunity, as Hassan puts it in the clip above, is to point AI at the non-obvious factors — the things a manual, geography-first process tends to miss entirely.

A territory isn't what it's made of. It's what it should be made of.

Here's the reframe worth sitting with: a territory is not simply the set of accounts it currently contains. It's the set of accounts it ought to contain.

That distinction sounds small, but it changes the whole job. If a territory is just its current accounts, planning becomes an allocation problem — divide the known book fairly and move on. If a territory is the accounts it should hold, planning becomes a discovery problem: which accounts belong here that aren't in the book yet?

A territory is not just a set of accounts that it is comprised of. It can be the set of accounts it ought to be comprised of.

For most growing organizations, the answer to that question is greenfield — accounts with real potential that have never been scored, segmented, or assigned to anyone. They sit outside the existing footprint precisely because the planning process was built around accounts the team already knew about.

Greenfield is where the missed potential hides

Greenfield accounts are hard to plan for manually, and for an understandable reason: there's no engagement history to lean on. No past pipeline, no prior owner, no closed-won signal to anchor a judgment. So they get scored thinly, if at all — and a thinly scored account is an account that's easy to leave out.

This is exactly the kind of work AI is well suited to. Instead of relying on the handful of signals a human can hold in their head, a model can score greenfield accounts against a far richer picture:

• Firmographic and growth signals — size, segment, trajectory, and how closely an account resembles your best existing customers.
• Fit and propensity — likelihood to convert based on patterns across accounts you've already won, not just the ones a rep happens to know.
• Intent and external indicators — signals that an account is moving in your direction before it ever raises a hand.
• Whitespace within the footprint — adjacent accounts that fit the profile of a territory but were never folded into it.

Score greenfield thoroughly, and those accounts stop being invisible. They become candidates — and candidates can be placed.

Why this matters now

Heading into FY27 planning, the temptation is to treat AI as a faster version of the same annual exercise. The bigger unlock is to let it change what the exercise is for. A territory built only from known accounts will always reflect yesterday's footprint. A territory built from the accounts it ought to contain reflects where the business could actually grow.

That's the hidden value. It was never about drawing cleaner lines around the accounts you already had. It's about seeing the ones you didn't.