Confidential · © 2026 Space-O Technologies Inc. · Prepared exclusively for DMA. Please do not copy, save or share.

Section 15 · Deal Autopsy & Sales Coaching

15.5Win-pattern mining (mirror of loss analysis)

Designed for Phase 2
No.
15.5
DMA's reference
Decision Book p.30
Phase
Phase 2
Autonomy
L1 · Recommend
DMA's decision
Yes

What DMA asked for · Decision Book p.30

Classify why deals are won: the right proof, timing, a strong champion, price accepted, a fast response.

Why it matters to DMA: It tells DMA what to repeat, not only what to avoid.

In our words, from DMA's Product Decision Book v3, page 30.

How we do it

  • Scheduled for Phase 2, the mirror of 15.1 with the same clean-data dependency.
  • The same reading and timing rules run on won deals: which proof was shown, how fast DMA answered, who the champion was.
  • Win and loss patterns sit side by side, so the difference is visible (for example, won deals got a first reply within a day).

Already in place

  • Won and lost outcomes are mirrored from Nutshell, and every AI decision can be linked to the outcome it led to.
  • The audit already reads won deals and former clients to classify the relationship and its evidence.

What the later phase adds

  • Win-reason codes with evidence.
  • A side-by-side view of win and loss patterns.

The flow

It plays on its own while it's on screen; hover or use the controls to pause or step through.

Won dealnotes, activitiesRead + timedsame rules as 15.1Side by sidewins vs losses
Step 1 of 3

A won deal

A three-location dental group signed in 2023.

  • Nutshell
  • AI
  • Screen

Where the data goes

The same six stages on every page. Nothing reaches Nutshell except through the write gate.

  1. 1Nutshell

    Source

    Won and lost deals, notes and calls.

  2. 2Sync

    Copies Nutshell changes into the bridge database

    Mirrors outcomes as they close.

  3. 3Bridge database

    A copy of the CRM data, plus what the AI works out

    Reason codes with their evidence; the patterns found.

  4. 4AI

    Claude models, only through DMA's own gateway

    Reads notes and summaries; code counts the patterns.

  5. 5Write gate

    The only way back into Nutshell: checked, approved when needed, sent once

    Not involved: this is analysis, nothing is written to Nutshell.

  6. 6Nutshell

    The system of record

    Unchanged.

The tables behind it

Drawn from the POC's database catalogue: structure only, no data.

crm.leadsPKiduuidnametextstatustextvalue_amountnumericoutcome_idtextclosed_attimestamp+ 20 more columnscrm.outcomesPKiduuidnametextkindtext+ 10 more columnsai.extracted_factsPKiduuidfact_keytextconfidencenumericevidence_quotetextIDprompt_iduuid+ 10 more columns

PK primary keyFK reference the database enforces (solid line)ID reference kept by id (dashed line)

TableWhat it holdsColumns
crm.leadsDeals mirrored from Nutshell, with outcome, value and close date.26
crm.outcomesNutshell's outcome list (won, lost and their reasons).13
ai.extracted_factsFacts from the notes with their evidence quotes.15
How the tables connect (1)
ColumnPoints toKept by
ai.extracted_facts.prompt_idagent.promptsThe application (by id)

Worked example

Synthetic demo data: every name, business and number is made up.

A designed comparison (synthetic example).

Won versus lost

SignalWon dealsLost deals
First reply within a day82%35%
Vertical case study shown71%22%

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

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.

15.5 Win-pattern mining (mirror of loss analysis) · DMA AI proof