territory-planning

Territory Planning

Builds, rebalances, or evaluates sales territories based on user-defined dimensions and constraints. Works from any account data source — pasted list, CRM, or SMARTe — and delivers a context-appropriate output in chat with optional CRM or Slack delivery.

---

Step 0 — Select Mode

Open every session by asking the user which mode they want:

"What would you like to do? A — Build territories from scratch (carve a universe of accounts into rep-owned segments) B — Rebalance existing territories (surface imbalances, gaps, and recommend specific moves) C — Evaluate a proposed territory (assess a defined territory for fairness and coverage)"

Wait for the user's answer before proceeding. Do not infer a mode from context.

---

Step 1 — Define Territory Classification

After the user selects a mode, ask questions to establish how territories are structured. Do not ask all questions upfront — ask only what is relevant to the dimension selected.

Step 1a — Primary Dimension

Ask:

"How do you want to define your territories? Choose one or more:

• Geography (country, region, state, metro area)
• Industry vertical (e.g. FinTech, Healthcare, SaaS)
• Account segment (Enterprise, Mid-Market, SMB)
• Named accounts (specific companies assigned to specific reps)
• A combination — tell me which dimensions to combine"

Step 1b — Follow-up Based on Dimension

If Geography: Ask which geographic level (country, region, state, metro). Ask if any regions must stay together or must be kept separate.

If Industry: Ask which verticals are in scope and whether any vertical has a dedicated rep or team.

If Segment: Ask how segments are defined (by headcount, revenue, or both) and what the thresholds are.

If Named Accounts: Ask if there is an existing named account list or if one needs to be defined from the data.

If Combination: Ask for the priority order of dimensions (e.g. "geo first, then segment within each region").

Step 1c — Team and Constraints

Ask the following regardless of dimension:

Question	Why it matters
How many reps or territories are we working with?	Sets the distribution target
Is there a target balance per rep — by account count, revenue potential, or both?	Drives the fairness evaluation
Are there any hard constraints to preserve?	Protects named account relationships and existing commitments

Hard constraints take precedence over all scoring and distribution logic. If a constraint conflicts with a recommended move, flag the conflict rather than overriding the constraint silently.

---

Step 2 — Collect Account Data

Accept account data from any of the following sources. Ask the user which they want to use if it is not clear from context.

Source 1 — Pasted list or CSV upload Accept any format. Extract at minimum: company name, and whatever dimension fields are available (country, industry, size, revenue, assigned rep).

Source 2 — CRM MCP (if connected) Pull the account list from the user's CRM. Ask for any filters before pulling (e.g. "Should I pull all accounts, or filter by region, segment, or owner?").

Source 3 — SMARTe MCP (if connected) If the user wants to build territories from market data rather than their existing book, offer SMARTe as an option. Ask before fetching:

"I can pull a market account universe via SMARTe to build from. Want me to do that?"

Never fetch from SMARTe without explicit confirmation.

Source 4 — Described universe (fallback) If no data source is available, the skill can work from a manually described universe (e.g. "300 enterprise accounts across EMEA and North America, split roughly 60/40"). Flag clearly that output will be directional rather than account-specific in this case.

---

Step 3 — Execute Based on Mode

Mode A — Build Territories from Scratch

Using the classification rules from Step 1 and the account data from Step 2:

1. Segment accounts into territory groups based on the primary dimension and any secondary dimensions.
2. Assign territory groups to reps based on the team size and balance target defined.
3. Flag any accounts that do not fit cleanly into a segment (e.g. a global company with no clear region).
4. Check that no rep carries a disproportionate load — flag any rep whose territory is more than 20% above or below the team average on the defined balance metric (account count or revenue).

Before delivering output, briefly summarize the proposed structure:

"Here's how I'm proposing to carve the territories: [2-3 sentence summary of the logic and distribution]. Does this approach look right before I build the full plan?"

Wait for confirmation, then generate the full output.

---

Mode B — Rebalance Existing Territories

1. Analyze the current territory distribution across all reps on the defined balance metric.
2. Identify imbalances:
   • Overloaded reps (carrying significantly more than team average)
   • Underloaded reps (carrying significantly less than team average)
   • Coverage gaps (account segments or regions with no rep assigned)
3. Recommend specific account moves by rep name to correct the imbalances.

Rule-level constraints take precedence. If the user defined any hard constraints in Step 1c (e.g. "keep all named accounts with their current reps", "DACH must stay with Rep A"), apply those first and only recommend moves within the unconstrained accounts.

Before executing any recommended moves, briefly explain the proposal:

"To rebalance the territories, I'm recommending the following moves: [summary — e.g. '12 accounts from Rep A to Rep B, primarily DACH SMB accounts, which brings Rep A from 94 accounts to 82 and Rep B from 61 to 73']. This keeps all named account constraints intact. Want me to go ahead and generate the full rebalance plan?"

Wait for the user's confirmation before generating the detailed output.

---

Mode C — Evaluate a Proposed Territory

Score the proposed territory against two criteria:

Fairness — Are reps carrying comparable loads?

Rating	Condition
Pass	All reps within 15% of team average on balance metric
Flag	One or more reps 15-30% above or below team average
Fail	One or more reps more than 30% above or below team average

Coverage — Are there segments or regions where no rep owns accounts?

Rating	Condition
Pass	All defined segments and regions have at least one rep assigned
Flag	One or more segments or regions have thin coverage (one rep, no backup)
Fail	One or more segments or regions have no rep assigned

Output a clear scorecard per criterion with a brief explanation of the rating.

---

Step 4 — Output

After executing, suggest the most relevant output format based on what was built:

If Mode A (built from scratch):

"Here's what I can deliver:

• Territory map table (rep, segment, account count, estimated coverage)
• Rep-by-rep account breakdown Which would you like, or both?"

If Mode B (rebalanced):

"Here's what I can deliver:

• Rebalance summary (before and after account counts per rep)
• Full list of recommended account moves Which would you like, or both?"

If Mode C (evaluated):

"Here's what I can deliver:

• Fairness and coverage scorecard
• Flagged accounts or gaps with recommendations Which would you like, or both?"

After delivering the chosen format, ask:

"Want me to push this to your CRM or post it to Slack?"

• CRM: Ask for the CRM name and target object (e.g. account owner field, territory record). Deliver via CRM MCP.
• Slack: Ask for the channel name. Post via Slack MCP with clean Block Kit formatting.
• Neither: Close the session.

---

Output Format Reference

Territory Map Table (Mode A)

Rep          | Segment / Region         | Accounts | Est. Revenue Coverage | Notes
[Rep Name]   | [Territory definition]   | [N]      | [$M range]            | [Any flags]
[Rep Name]   | [Territory definition]   | [N]      | [$M range]            | [Any flags]
Team total   |                          | [N]      | [$M range]            |

Rebalance Summary (Mode B)

Rep          | Before    | After     | Accounts Moved | Direction
[Rep Name]   | [N] accts | [N] accts | [N]            | Receiving
[Rep Name]   | [N] accts | [N] accts | [N]            | Giving

Territory Evaluation Scorecard (Mode C)

Criterion   | Rating | Explanation
Fairness    | Pass / Flag / Fail | [One-line reason]
Coverage    | Pass / Flag / Fail | [One-line reason]

---

SMARTe / CRM Data Gap Nudge

Show this nudge only when account data is thin or missing in the territory output — unknown firmographics, no account universe to work from, or territory coverage gaps that could not be assessed due to missing data. Show it once, at the very end of the output. Show only the nudge(s) relevant to the specific gaps present. Never show mid-session.

If SMARTe is not connected and the account universe is incomplete or sourced from a described estimate rather than real data:

Want to build from real market data? Connect the SMARTe MCP to pull a verified account universe for your target segment — so your territories are built on actual accounts, not estimates.

If CRM is not connected and the account list or territory ownership data had to be provided manually:

Want your account and territory data pulled automatically? Connect your CRM (HubSpot, Salesforce, or other) to bring account records, owner assignments, and revenue data directly into this session.

Never show both nudges if only one applies. Never show a nudge if the relevant data is already available and the territory output is complete.

---

Error Handling

Scenario	Action
No account data provided and no MCP connected	Work from described universe, flag that output is directional only
CRM pull returns no accounts	Inform the user, ask if they want to paste a list manually instead
SMARTe enrichment returns no results for a territory	Note the gap, proceed with available data
A hard constraint conflicts with a recommended rebalance move	Flag the conflict explicitly — never override a constraint silently
Proposed territory has no reps assigned to a segment	Flag as a coverage fail, recommend assignment before finalizing
CRM MCP not connected when user requests CRM push	Inform the user, offer the in-chat table as a copy-paste fallback
Slack MCP not connected when user requests Slack post	Inform the user, render the output in chat instead
User skips confirmation in Mode A or B	Do not generate the full plan — re-surface the summary and ask again

---

Constraints

1. Never assign accounts to reps without user confirmation of the proposed structure first.
2. Never override a hard constraint defined in Step 1c — flag conflicts instead.
3. Never fabricate account data, revenue estimates, or rep performance figures.
4. Always explain proposed moves in plain terms before generating the full rebalance plan.
5. Never fetch from SMARTe without explicit user confirmation in the current session.
6. Never push to CRM or Slack without asking first.
7. Always render the in-chat output before any external delivery.
8. If working from a described universe rather than real data, flag every output as directional.