Win/Loss Analysis Agent

The short version

The Win/Loss Analysis agent runs bi-weekly Tuesday at 10 AM and reads every won and lost deal from the last two weeks. It pulls from Salesforce (deal size, close reason, sales notes), call transcripts, and win/loss interviews, then extracts seven categories of insight: win patterns, loss patterns, feature insights, segment insights, competitive positioning, pricing insights, and three recommendations. The point is to stop building features nobody asked for and start building the ones that move deals. The output of one run: "We win enterprise on API reliability and lose mid-market on price. Two product fixes would convert 30% of mid-market losses."

You won a big deal last week. You lost one too. Your sales team knows why, but they're busy with the next deal. The insights get lost.

Meanwhile, product doesn't know: are you losing deals because of a product gap? A pricing problem? Bad UX? Or just competitive pressure? You're building features nobody asked for while missing the ones that would actually close deals.

The Win/Loss Analysis agent extracts insights from your deal outcomes. Bi-weekly, it reads: recent wins (Salesforce), recent losses (with call notes or post-mortems), and customer interviews. It synthesizes: what products/features are you winning on? What gaps are costing you deals? What are competitors doing that you're not?

How It Works

The agent pulls three data sources and analyzes patterns:

Win analysis: Recent closed/won deals from Salesforce. For each: deal size, customer segment, deal length, and sales notes. The agent looks for patterns. "All our enterprise wins mention API reliability. API reliability is a differentiator."

Loss analysis: Recent closed/lost deals. The reason recorded (if any), sales notes, and calls notes if you have them. "We lost to competitor X because they had feature Y that we don't." "Customer selected competitor because of lower pricing."

Interview data: If you do win/loss interviews, pull those transcripts. "Why did you choose us?" and "Why didn't you choose them?" interviews give you the real story.

The output: clear product insights you can act on.

Data Sources and Setup

Prerequisites: You'll need:

• Salesforce: Win/loss records, deal size, customer segment, close reason, sales notes
• Call recordings / transcripts: Sales calls with lost prospects (if available)
• Win/loss interview data: Structured interviews with customers and churned customers
• Competitive context: Rough knowledge of what competitors offer
• CRM contact history: To understand decision-making (who was involved? how long did it take?)

Schedule: Bi-weekly Tuesday at 10 AM. Analyzes 2 weeks of deal outcomes.

The Claude Prompt

You are analyzing won and lost deals to extract product insights.

Here are the deals we won in the past 2 weeks:
[WON DEALS:
- Deal size and customer segment
- Sales cycle length
- Key decision-maker roles
- Reason selected us (from Salesforce or sales notes)
- Key features that came up in conversations
- Objections we overcame]

Here are the deals we lost in the past 2 weeks:
[LOST DEALS:
- Deal size and customer segment
- Sales cycle length
- Reason we lost (if recorded)
- Competitor who won (if known)
- Key objections we couldn't overcome
- Notes from sales team]

Here's win/loss interview data (if available):
[INTERVIEWS:
- Customer quotes on why they chose us
- Reasons they almost didn't
- Competitor comparisons
- Feature gaps they mentioned
- Anything surprising they said]

Here's our competitive context:
[COMPETITORS: rough comparison of features, pricing, positioning]

Please analyze and report:

1. **Win Patterns**
   - What's common across our wins? (segment? deal size? time to close?)
   - Which features are we winning on most?
   - Which customer segments are easiest to close?
   - What objections do we overcome successfully?

2. **Loss Patterns** (PRIORITY)
   - What's common across our losses?
   - Are we losing to a specific competitor? If so, why?
   - What's the #1 reason we lose deals?
   - Are we losing to price? To features? To something else?
   - Which segments are hardest to close?

3. **Feature Insights**
   - Which features do winning customers care most about?
   - Which feature gaps are costing us deals?
   - Are there features we're building that don't move deals?
   - Which features should we emphasize in sales?

4. **Segment Insights**
   - Do winning segments have different needs than losing segments?
   - Are we positioned best in some segments and weak in others?
   - Should we focus on segments where we win more?

5. **Competitive Positioning**
   - Which competitors are we losing to most?
   - What are they doing that we're not?
   - What's our advantage vs. them?
   - Where are we vulnerable?

6. **Pricing Insights**
   - Are we losing deals to price?
   - Does price sensitivity vary by segment?
   - Are we winning on price? Against whom?

7. **Recommendations**
   - Top 3 product improvements that would help us win more deals?
   - Which segments should we invest in most?
   - What should we fix urgently? What can wait?

Format as a clear analysis with patterns and actionable insights. I should be able to use this for roadmap planning.

What You Get

Instead of deal outcomes being invisible to product:

• Validated roadmap: You know which features would actually close deals
• Churn insight: Why did we lose this deal? Same reason we churn? Data-backed.
• Competitive positioning: Clear picture of what you're winning on and where you're vulnerable
• Segment insight: Some segments are easy to close (win rate 60%), others hard (win rate 20%) - why?
• Pricing understanding: Are you losing to price or to product? The answer changes your strategy.

Real outcomes:

• You stop building features based on the loudest customer and start building features that move deals
• Sales has customer data to refine positioning ("We win when we emphasize API reliability")
• You catch competitive threats early ("Competitor X is winning enterprise by adding X feature")

For the full agent fleet and scheduling details, see Your AI Agent Fleet.