Every proposal teaches the system — client psychographics, deal outcomes, and strategy patterns stored and retrieved via vector search

Client Intelligence — Memory That Compounds

Every proposal teaches the system. Client psychographics, deal outcomes, and strategy patterns are stored and retrieved via vector search.

How It Works

The intelligence flywheel follows a continuous cycle: generate a proposal, extract psychographics (tone, themes, strategy), store with a vector embedding, track the deal outcome (won/lost/pending), and when the next proposal comes in for the same or a similar client, vector search retrieves the full context so the strategist can personalize the approach.

Data Model

Psychographics (JSONB)

Stored per client, per org. Updated after every proposal.

Field Example
preferred_tone "professional but approachable"
key_themes ["scalability", "cost savings"]
risks_noted ["budget conscious"]
pricing_strategy_used "value-based"

Deal History (JSONB Array)

Append-only log of every proposal created for this client.

Field Example
proposal_id uuid
date "2026-03-15"
positioning "technical partner"
key_themes ["automation"]
outcome "won"

Vector Search

How vector search works: When a new proposal is created, the brain calls match_client_intelligence() with the requirements text embedded via OpenAI's text-embedding-3-small (1536 dimensions). PostgreSQL's pgvector extension finds the most similar past clients using cosine similarity, returning their psychographics and deal history. The strategist uses this to personalize the approach.

Intelligence Over Time

Day 1

Cold start. No history. Generic but on-brand proposal using org brain knowledge.

Day 30

5 proposals generated. Vector search finds similar clients. Strategy starts personalizing.

Day 90

Returning client with full context. Knows their tone, themes, what won before. Near-automatic.

Ready to create AI-powered proposals?

Start Free