Win Loss Analysis

---
name: win-loss-analysis
description: Analyze competitive wins and losses using deal or adoption evidence to identify decision criteria, alternatives, objections, switching dynamics, and repeatable patterns without confusing sales anecdotes with market truth.
---
# Win-Loss Analysis

Use when the team needs to understand why customers chose, rejected, replaced, or stayed with a product relative to alternatives.

## Procedure
1. Define the decision population, timeframe, segment, and what counts as a win, loss, churn, no-decision, or excluded case.
2. Gather evidence from customer interviews, sales or support records, CRM notes, product usage, pricing context, and competitor information while preserving source quality.
3. Separate customer-stated reasons from internal interpretation and from facts observed in the deal or adoption path.
4. Code recurring decision criteria such as capability, trust, integration, price, usability, migration, timing, procurement, relationship, or organizational fit.
5. Compare patterns across wins, losses, no-decisions, segments, deal sizes, use cases, and competitors rather than overgeneralizing one memorable story.
6. Identify where product gaps, positioning, sales process, onboarding, pricing, or external constraints actually owned the outcome.
7. Quantify patterns only when the sample and data quality support it; keep qualitative themes visible when they do not.
8. Translate findings into specific product, positioning, pricing, or process hypotheses and define how future cases should test them.

## Decision rules
- A seller's loss reason is not automatically the buyer's real decision reason.
- No-decision is a distinct alternative and can reveal urgency or switching-cost problems.
- Small biased samples should produce hypotheses, not population claims.
- Avoid turning analysis into a blame scoreboard between product and sales.

## Quality gate
The analysis is ready when evidence sources and sample limits are clear, customer reasons are separated from internal inference, recurring decision criteria are compared across meaningful cohorts, and findings produce testable improvements rather than anecdotal narratives.