Claude Code Setup for Sales Directors

# CLAUDE.md — Sales Director
> Version: 2026-02-17 | Difficulty: intermediate

## What Your AI Can Now Do
- Get a full pipeline snapshot with deal stages, risk flags, and revenue forecast in one command
- Prep for any deal review or board meeting with auto-generated talking points from live CRM data
- Spot at-risk deals before they stall by tracking activity gaps and stage velocity
- Build territory and quota plans backed by historical performance data
- Draft coaching notes for reps based on their pipeline health and win/loss patterns

---

## Quick Start
1. Paste this file into your project as `CLAUDE.md`
2. Add your CRM and analytics credentials to `.env`
3. Run `/pipeline-review` to see your current pipeline health across all reps
4. Run `/forecast [quarter]` to generate a revenue forecast with confidence levels
5. Review the Automation table below and enable the weekly deal risk alerts

---

## About Me
Sales director leading a team of 5-15 reps across inbound and outbound motions. I manage pipeline, set quotas, run deal reviews, coach reps, and own the revenue number. I spend most of my time in forecast meetings, 1:1s, and strategic account reviews. What matters most: predictable revenue, rep development, and pipeline velocity.

---

## Pre-Built Commands

| Command | What It Does |
|---------|-------------|
| `/pipeline-review` | Returns pipeline by stage, rep, and deal size — flags stalled deals and missing next steps |
| `/forecast [quarter]` | Generates a weighted revenue forecast with best-case, commit, and worst-case scenarios |
| `/rep-scorecard [name]` | Pulls a rep's pipeline health, activity metrics, win rate, and average deal cycle — highlights coaching opportunities |
| `/deal-review [deal name]` | Summarizes deal history, stakeholders, risks, and recommended next moves for a specific opportunity |
| `/meeting-prep [context]` | Generates talking points, data highlights, and risk callouts for board meetings, QBRs, or deal reviews |
| `/territory-plan [segment]` | Builds a territory allocation plan based on market size, rep capacity, and historical conversion rates |

---

## How Claude Connects to Your Tools

| Tool | Why You Need It |
|------|----------------|
| Google Workspace | Access deal notes in Drive, coordinate with reps via email, pull data from shared pipeline spreadsheets |
| Google Analytics | Track which marketing channels feed your pipeline and how website activity correlates with deal progression |
| AnySite | Research target accounts, pull LinkedIn profiles of key stakeholders, and analyze competitor positioning |

### Setup
```bash
# Required in .env
GOOGLE_CREDENTIALS_PATH=./credentials.json
CRM_API_KEY=your-crm-api-key
CRM_BASE_URL=your-crm-instance-url
```

---

## Sales Leadership Rules
- **MUST:** Never present a forecast number without stating the confidence level and methodology (weighted, commit, best-case)
- **MUST:** Flag any deal that has been in the same stage for more than 2x the average stage duration
- **MUST:** Every deal over $50K must have a documented champion, economic buyer, and next step
- **SHOULD:** Review each rep's pipeline at least weekly — monthly is too late to course-correct
- **SHOULD:** Track leading indicators (meetings booked, proposals sent) not just lagging ones (closed revenue)
- **MUST:** Never let a rep mark a deal as "commit" without validated next steps and buyer confirmation
- **SHOULD:** Compare rep performance against cohort averages, not just quota — context matters for coaching
- **SHOULD:** Document every deal review decision so the team can learn from wins and losses

---

## Project Structure
```
sales/
├── pipeline/
│   ├── by-rep/
│   └── by-segment/
├── forecasts/
│   └── [quarter]/
├── rep-coaching/
│   └── [rep-name]/
├── deal-reviews/
│   └── [deal-name]/
├── territories/
└── playbooks/
    ├── inbound.md
    ├── outbound.md
    └── expansion.md
```

### What These Folders Mean
- **pipeline/** — Live pipeline views sliced by rep and market segment
- **forecasts/** — Quarterly revenue forecasts with scenario modeling and historical accuracy tracking
- **rep-coaching/** — Individual rep scorecards, coaching notes, and development plans
- **deal-reviews/** — Deal-level analysis with stakeholder maps, risk assessments, and action items
- **territories/** — Territory allocation plans and capacity models
- **playbooks/** — Sales motion playbooks for inbound, outbound, and expansion selling

---

## Automation

| When | What Happens | If It Fails |
|------|-------------|-------------|
| Monday morning | Generates pipeline snapshot with week-over-week changes and stalled deal alerts | Send a plain list of deals by stage from last known data and flag the sync issue |
| Deal inactive 14+ days | Alerts you with the deal name, rep, last activity, and suggested re-engagement action | Add the deal to a "needs attention" list for the next 1:1 with that rep |
| Friday afternoon | Compiles weekly activity metrics per rep: calls, emails, meetings, proposals | Log that metrics were unavailable and remind you to pull them Monday |
| Quarter start | Generates territory and quota plans based on last quarter performance and market data | Use last quarter's plan as a starting template and flag what needs manual updating |
| Before scheduled 1:1 | Pulls the rep's pipeline changes, wins, losses, and coaching themes since last meeting | Provide a blank 1:1 template with standard coaching questions |

---

## Gotchas
- **MUST:** Never coach a rep based on a single deal outcome — look at patterns across 10+ opportunities before drawing conclusions
- **MUST:** Never share individual rep performance data in group settings without their knowledge — coaching is private
- **SHOULD:** Sanity-check forecast numbers against historical conversion rates — optimism bias is the #1 forecasting killer
- **MUST:** Never override a rep's deal stage without discussing it with them first — trust breaks fast
- **SHOULD:** Track why deals are lost with the same rigor as why they are won — loss patterns reveal more than win patterns

---

## Memory & Learning
Claude builds context over time:
- Each rep's strengths, development areas, and historical performance trends
- Which deal patterns (size, industry, stakeholder mix) have the highest win rates
- Forecast accuracy over time — how your commit numbers compare to actual closed revenue
- Coaching themes that moved the needle for specific reps
- Seasonal patterns in your pipeline (when deals tend to stall, when they accelerate)

To save something important: just say "remember that [fact]" in conversation.
To review saved memories: use the `/memory` command to open the memory editor.