Section 15 · Deal Autopsy & Sales Coaching
15.1Explain why deals are lost
Designed for Phase 2- No.
- 15.1
- DMA's reference
- Decision Book p.29
- Phase
- Phase 2
- Autonomy
- L1 · Recommend
- DMA's decision
- Yes
What DMA asked for · Decision Book p.29
Classify why deals are lost: price, trust, weak proof, bad timing, slow response, wrong fit, competitor, poor follow-up, no urgency.
Why it matters to DMA: It underpins both reactivation and coaching, but needs clean lost-reason data first, so the book schedules it for Phase 2.
In our words, from DMA's Product Decision Book v3, page 29.
How we do it
- Scheduled for Phase 2, after the lost-reason clean-up (17.4). The first half already runs: the audit codes every lost deal from its notes.
- The design widens the code list to the book's nine reasons and adds signals the CRM holds but notes miss: response times, gaps with no follow-up, how long a deal sat in each stage.
- Every classification keeps its evidence (a quoted note or a measured gap), so a manager can check it.
Already in place
- The audit already classifies every lost deal into a lost-reason code (price, timing, existing agency, no budget, competitor, poor fit, other), quoting the note it came from, or "not enough evidence" when the notes don't say.
- Won and lost outcomes are mirrored from Nutshell, and every AI decision can be linked to the outcome it led to.
What the later phase adds
- The book's nine loss reasons as the code list.
- Response-time and follow-up-gap signals from the activity history.
The flow
It plays on its own while it's on screen; hover or use the controls to pause or step through.
A lost deal
A $4,500-a-month PPC deal was marked lost in May.
- Nutshell
- AI
- Rule in code
- Stored data
Where the data goes
The same six stages on every page. Nothing reaches Nutshell except through the write gate.
- 1Nutshell
Source
Won and lost deals, notes and calls.
- 2Sync
Copies Nutshell changes into the bridge database
Mirrors outcomes as they close.
- 3Bridge database
A copy of the CRM data, plus what the AI works out
Reason codes with their evidence; the patterns found.
- 4AI
Claude models, only through DMA's own gateway
Reads notes and summaries; code counts the patterns.
- 5Write gate
The only way back into Nutshell: checked, approved when needed, sent once
Not involved: this is analysis, nothing is written to Nutshell.
- 6Nutshell
The system of record
Unchanged.
The tables behind it
Drawn from the POC's database catalogue: structure only, no data.
PK primary keyFK reference the database enforces (solid line)ID reference kept by id (dashed line)
| Table | What it holds | Columns |
|---|---|---|
| crm.leads | Deals mirrored from Nutshell, with outcome, value and close date. | 26 |
| crm.outcomes | Nutshell's outcome list (won, lost and their reasons). | 13 |
| ai.extracted_facts | Facts from the notes with their evidence quotes. | 15 |
How the tables connect (1)
| Column | Points to | Kept by |
|---|---|---|
| ai.extracted_facts.prompt_id | agent.prompts | The application (by id) |
Worked example
Synthetic demo data: every name, business and number is made up.
Designed output for three lost deals (synthetic example).
Lost deals, explained
| Deal | Reason | Evidence |
|---|---|---|
| Harbor & Finch Law — PPC | Price; slow response | "went with a cheaper freelancer"; first follow-up after 9 days |
| Peak Comfort HVAC — SEO | Timing | "revisit after the busy season" (note, July 2025) |
| Lakeview Orthodontics — web | Weak proof | "wanted an orthodontics case study" |
In the running POC
There is no screen for this one yet: it is designed for Phase 2, as DMA's book decides. The parts listed under “Already in place” run in the POC today.
Status
Designed for Phase 2. DMA's book schedules this for Phase 2. The design is ready and builds on parts that already run in the POC.