Pipeline Review

AI-assisted pipeline analysis with deal health scoring, forecast modeling, and risk identification across your funnel. This skill turns pipeline reviews from subjective gut-feel exercises into data-driven inspection sessions.

When to Use This Skill

Use this skill when the user:

• Is preparing for a weekly or monthly pipeline review meeting
• Wants to identify deals that are stalled, at risk, or likely to slip
• Needs to build a forecast — weighted, best-case, or commit — from current pipeline data
• Wants to analyze pipeline coverage ratios and identify gaps by segment
• Needs inspection questions for reviewing specific deals with their team

What This Skill Does

Request pipeline data from the user. Accept it in any format: a list of deals with stages and amounts, a CRM export, or even a verbal description. At minimum, you need deal names, stages, expected amounts, and close dates. Activity history and engagement data improve the analysis but are not required.

Perform the analysis in three layers:

Layer 1 — Deal-Level Health Scoring For each deal, assign a health score (Green / Yellow / Red) based on:

• Stage duration vs. historical average for that stage
• Days since last meaningful activity (email, call, meeting)
• Number of stakeholders engaged
• Whether the close date has been pushed
• Qualification completeness (if data is available)

Flag any deal as Red if it has been in the same stage for more than 2x the typical duration, if there has been no activity in 14+ days, or if the close date has been pushed more than once.

Layer 2 — Pipeline-Level Analysis

• Total pipeline value by stage
• Coverage ratio: pipeline value vs. quota (target is 3x for early stage, 2x for mid-stage)
• Stage conversion rates if historical data is available
• Concentration risk: flag if more than 30% of pipeline value sits in 1-2 deals

Layer 3 — Forecast Modeling Generate three forecast scenarios:

• Weighted: Each deal's value multiplied by its stage probability
• Best case: Weighted forecast plus deals with strong engagement signals
• Commit: Only deals where the rep has high confidence and qualification is strong

Anti-Pattern: "Forecast by Feelings" A pipeline review that ends with "I feel good about hitting our number" is not a forecast. Every commit-level deal needs specific evidence: confirmed decision criteria met, economic buyer engaged, verbal commitment received, and procurement process initiated. If a deal cannot cite these, it belongs in best-case, not commit.

Anti-Pattern: "Ignoring Stalled Deals" Deals that have not moved stage or had activity in 14+ days do not get better with time. They get worse. When identifying stalled deals, generate a specific re-engagement action for each — not "follow up," but a concrete next step: "Send a value-add email referencing their Q1 earnings mention of cost reduction, then propose a 15-minute check-in with a specific agenda."

When generating inspection questions for pipeline review meetings, tailor them to each deal's specific situation. Avoid generic questions like "How's this deal going?" Instead: "You mentioned the CFO was reviewing the business case two weeks ago — what was their feedback, and has the procurement team been looped in yet?"

Example Prompts

• "Analyze my pipeline and tell me which deals are at risk of slipping this quarter"
• "Generate a forecast for Q2 based on my current pipeline and historical conversion rates"
• "Which of my deals have been stuck in the same stage for more than 14 days?"
• "Prepare inspection questions for my pipeline review meeting tomorrow"