version: "1.0.0" name: pipeline-reviewer description: Analyze CRM pipeline data to identify risks, recommend actions, and generate a weekly pipeline summary.

Pipeline Reviewer

Task

Analyze your CRM pipeline to surface stale deals, flag risk signals, recommend next actions, and generate a concise weekly pipeline summary.

Input Requirements

• CRM data source: one of:
• CRM API connection (configured in tech stack section of CLAUDE.md)
• Exported CSV with columns: deal_name, company, amount, stage, owner, created_date, last_activity_date, close_date, next_step, contacts
• File path: [e.g., ./data/crm/pipeline_export.csv]
• Pipeline stages: [List your pipeline stages in order, e.g., "Discovery → Demo → Proposal → Negotiation → Closed Won/Lost"]
• Review scope: all | my_deals | team:[TEAM_NAME]

Instructions

1. Pipeline Health Scan

For each open deal, check:

• Stale deals: No activity in [e.g., 14+ days]. Flag with days since last activity.
• Slipping close dates: Close date is in the past or within 7 days with no recent activity.
• Stuck in stage: Deal has been in the same stage for longer than [e.g., 2x the average for that stage].
• Single-threaded: Only one contact engaged. Enterprise deals need 3+ contacts.
• No next step: Next step field is empty or vague ("follow up").
• Upside-down deal: Amount is below your average deal size for the stage - may not be worth the effort.

2. Risk Classification

Classify each deal:

3. Action Recommendations

For each Yellow and Red deal, recommend a specific next action:

• Stale → "Send a value-add resource to re-engage [CONTACT_NAME]"
• Single-threaded → "Identify and engage a second stakeholder - suggest [TITLE]"
• Slipping → "Propose a mutual action plan with revised timeline"
• No next step → "Schedule a [SPECIFIC_MEETING_TYPE] for [SUGGESTED_DATE]"
• Stuck → "Diagnose: is the champion still engaged? Check last interaction."

4. Pipeline Summary

Generate a weekly summary:

• Total pipeline: $ value, deal count
• By stage: deal count + $ value per stage
• Coverage ratio: pipeline value ÷ quota target = X:1
• Weighted pipeline: $ value × stage probability
• Velocity: average days per stage, average deal cycle length
• Movement this week: deals advanced, deals created, deals lost, deals won
• Risk overview: Red / Yellow / Green count + total $ at risk

Output Format

Markdown report saved to ./data/reports/pipeline_review_[DATE].md:

# Pipeline Review - [DATE]
 
## Summary
-Total pipeline: $X (N deals)
-Weighted pipeline: $X
-Coverage: X:1 against $[QUOTA] target
-Deals at risk: N ($X value)
 
## Risk Alerts
### Red (Immediate Action)
-[DEAL_NAME] - [COMPANY] - $X - [RISK_REASON] - Action: [RECOMMENDATION]
 
### Yellow (Monitor)
-[DEAL_NAME] - [COMPANY] - $X - [RISK_REASON] - Action: [RECOMMENDATION]
 
## Stage Distribution
| Stage | Deals | Value | Avg Days |
|-------|-------|-------|----------|
 
## Wins & Losses This Week
-Won: [DEALS]
-Lost: [DEALS] - Reasons: [LOSS_REASONS]
 
## Recommendations
1.[TOP_PRIORITY_ACTION]
2.[SECOND_PRIORITY_ACTION]
3.[THIRD_PRIORITY_ACTION]

Example Usage

/pipeline-reviewer
 
Source: CRM API (HubSpot)
Scope: my_deals

Output:

Pipeline Review - 2026-03-15
 
Summary: $1.2M pipeline (18 deals), weighted $480K, 3.2:1 coverage
 
Red Alerts:
- Acme Corp - $85K - No activity in 23 days, single-threaded
  → Action: Re-engage via LinkedIn. Identify VP Engineering as second thread.
 
- DataFlow - $120K - Close date was March 10, no next step defined
  → Action: Send mutual action plan with revised March 28 close date.
 
Yellow:
- CloudStack - $45K - In Demo stage for 18 days (avg is 10)
  → Action: Propose a technical deep-dive to move to Proposal.