Closes the loop from closed deals back to positioning, competitive, and CS. Win/loss without continuous synthesis stays ad hoc; the signal is in the deals, the gap is in the loop. This playbook standardizes the loop.
Source synthesis: Bob Moesta (JTBD switch interview — JTBD interviews surface the customer's actual language and the switch trigger); Dunford (sales detects positioning failure first — The sales team detects positioning failure months before the dashboard does); the structural insight that "no decision" is the real competitor (40–60% of B2B buyers say "no decision" — your real competitor is the status quo).
When to use
- Continuous monthly cadence on closed-lost.
- Quarterly synthesis pass.
- After any pricing change or competitive entrant.
- Before a positioning or battlecard refresh.
The loop
Closed-lost deal → call-intel deep dive (if AE call recorded)
→ closed-lost interview (5–7 question script)
→ theme extraction
→ positioning-drift signal (if patterns diverge from ICP / pillar)
→ battlecard insight (if competitive loss)
→ objection-library update
Closed-won deal → pattern match against best-deal cohort
→ narrative reinforcement signal
→ case-study candidate flag (if allowed-claims fit)
Closed-lost interview question bank
Keep to 7 max, one primary per theme:
1. Timing. What was happening in your business when you started looking?
2. Trigger. What made you start evaluating solutions at that moment?
3. Alternatives. Who else did you look at? What was the short list?
4. Decision drivers. What were the 2–3 things that mattered most?
5. Vendor comparison. What did the vendor you chose do better than us?
6. Our gap. Where did we fall short — price, capability, team fit, confidence?
7. Would reopen? What would we need to show you to get back in the conversation?
Theme extraction (monthly)
Cluster closed-lost reasons into six gap types:
- Positioning gap. Prospects thought we were X; we are Y. Feeds positioning.
- Capability gap. Genuinely lacks feature. Feeds product.
- Competitive gap. Competitor positioned a capability we have but we didn't communicate. Feeds battlecards.
- Price gap. Genuine value-price mismatch. Feeds pricing.
- Trust/brand gap. Confidence issue. Feeds content + social.
- Team-fit gap. Hiring/timing. Not addressable via PMM.
Closed-won pattern match
Pattern-match every won deal against the best-deal cohort. Surface narrative reinforcement signals. Flag case-study candidates that fit allowed-claims rules.
Outputs
1. themes.md — top 5 lost-reason themes with supporting transcripts.
2. won-patterns.md — top 3 won-reason patterns with supporting transcripts.
3. Per-theme signals emitted to positioning, battlecard, objection library.
4. Case-study candidate list.
Non-goals
- Customer surveys. Too many caveats. Interviews only.
- Calling every lost deal. Capacity-bound — sample smartly, weight by ARR.
- Public case studies without written customer approval.
Quality gates
- Every closed-lost deal in scope has a written disposition reason; interviews on a weighted sample.
- Theme extraction monthly, not ad hoc.
- Drift between themes and current positioning surfaced explicitly.
- Battlecard updates traced to specific lost deals.
- No invented quotes — every quote in themes file traces to a recorded interview or call.
Common failure modes
- CRM disposition codes treated as the answer. Codes tell you who; interviews tell you why.
- One-shot win/loss audits that never recur. Themes only matter if they update positioning.
- Surveying instead of interviewing. Surveys average out the signal.
- Sampling skewed to closed-lost only. Closed-won patterns are equally informative.
- "No-decision" deals excluded from analysis. The real competitor is often inertia.
- Themes generated but not propagated — battlecards and positioning not updated.
- Public case studies pulled from unapproved transcripts.