Early Access. The Decision Context Graph Platform

The reasoning layer underneath your data.

Cartographer reads your warehouse and builds the Decision Context Graph: entities, relationships, and causality, automatically. Every layer above it, from drilldown to deep analysis to journey intelligence, queries that same grounded foundation. Same question, same answer, every time.

Co-Pilot AIJourney IntelligenceDeep AnalysisKPI DrilldownLiving MapSemantic Layer

ARCHITECTURE

The Architecture of the Decision Context Graph

Layers 1+2 build your context graph. Layers 3-6 unlock its value. Click any layer to learn more.

Layer 1 / CONTEXT GRAPH FOUNDATION

Hours, not months. Your context graph, automatic.

Traditional semantic-layer projects take months: interviewing stakeholders, writing LookML, documenting relationships by hand. Cartographer reads your actual query logs and produces the same structure, in hours, without anyone authoring it.

The pipeline learns from behavior, not just schema. After analyzing thousands of historical queries, it knows which KPIs your team actually cares about, which dimensions they slice by, and which relationships matter for which questions. No manual definitions. No interviews. No guesswork.

Cartographer Engine

Ready to scan

QUERY LOGS
0queries
6 AI AGENTS
Entities
Measures
Relations
Time
Segments
Insights
SEMANTIC LAYER
Waiting...

6 AI agents4 hours vs 4 monthsAutomatic semantic discovery

Entities

Tables classified by their role: objects, events, mappings

Measures

KPIs inferred from aggregation patterns in real queries

Relationships

JOIN paths mapped from actual usage, not just schema

Time patterns

Temporal granularity and cadence detected automatically

Segments

Dimensions surfaced from how the data is sliced in practice

Causal hints

Drill paths and significance signals synthesized across the graph

Multi-agent semantic discovery. Real query logs. Context graph in hours, not months.

Layer 2 / CONTEXT GRAPH VISUALIZATION

Your Context Graph, Visualized and Navigable

Together with Layer 1, this completes your Decision Context Graph, an interactive map that becomes your organization's single source of truth for decisions.

Start from your North Star Metric, the one number that matters most, and see exactly how every metric connects, who owns it, and what drives it. Replace twenty disconnected dashboards with one navigable context graph.

Living Map

Ready to visualize

Revenue

Customers

AOV

Frequency

New

Returning

Price

Quantity

One living viewReal-time healthGateway to all intelligence

North Star Decomposition

Every organization has one ultimate measure of success. The Living Map starts there and decomposes it into contributing drivers.

MECE-Validated Structure

Our AI validates that your KPI structure is Mutually Exclusive and Collectively Exhaustive, no overlaps, no gaps.

KPI Ownership

Every metric on the map has an owner, a person or team accountable for that number. No more finger-pointing.

Real-Time Health

Color-coded indicators show performance at a glance. Red, yellow, green, know instantly where to focus.

Navigate 200+ interconnected KPIs in a single view. Your complete Decision Context Graph.

Layer 3 / DISCOVERY INTELLIGENCE

Type a KPI. Get a complete data story.

Type any metric and the system generates a four-act story around it: the headline number, the drivers behind it, the metrics that move with it, and what to do about it. No analyst queue. No dashboard build. No waiting for someone to slice it the way you needed.

KPI Drilldown analyzes every dimension at once, surfaces patterns you would not have thought to ask about, and tells you which one to investigate first. Built on the context graph, so every claim is traceable back to the underlying data.

Type any KPI...

Click play to see the magic

ACT 1
ACT 2
ACT 3
ACT 4

Type any KPI, get a complete story

MindPalace transforms a single metric into a 4-act data narrative with AI-powered insights.

4 acts30 secondsComplete data story with AI insights

01

THE HERO

What's the number?

Current value, trend, AI summary

02

THE DRIVERS

What's behind it?

Dimension breakdowns, contributions

03

THE CONTEXT

What moves with it?

Correlations, relationships

04

THE INSIGHTS

So what?

Opportunities, alerts, actions

30 seconds. Any KPI. A full data story with statistical drivers.

Layer 4 / ANALYTICAL INTELLIGENCE

Statistical rigor. Human judgment. Board-ready output.

Deep Analysis is a four-step investigation flow. You see value at every stage, and you decide what gets analyzed at each step. No black boxes. No wall of charts to make sense of afterwards.

The AI proposes hypotheses based on the actual patterns in your data. You select the ones worth testing, edit them to match your business, or add your own. The system runs the statistics and returns an executive-ready answer with p-values, effect sizes, and quantified business impact.

4 steps. ~30 seconds total. Executive-ready output.

4 steps. ~30 seconds total. Board-ready output.

Step 1: Configure
Instant

Analyzing

Customer Lifetime Value (CLV)

-8% this quarter

What would you like to understand?

WHERE is it coming from?Aggregate
WHY is it performing this way?
RECOMMENDED

Find factors causing the -8% decline

HOW has it been changing?Trend

Segments to analyze

CityLoyalty TierEducation

Choose Your Question

No statistical jargon. Just three simple questions, WHERE, WHY, or HOW. MindPalace recommends the best one based on your KPI's current trend.

You stay in control

The AI proposes hypotheses from real patterns in your data. You decide which ones to run, edit them to fit your business, or add ones the model missed. Results stream in as each hypothesis completes, so you can act on the first finding without waiting for the rest.

Layer 5 / JOURNEY INTELLIGENCE

See the Journey. Find the Leak. Know Why.

Most funnel tools require you to define every stage by hand, then guess at the drop-offs. MindPalace discovers the journeys from the warehouse itself, ranks the biggest leak, and tells you why entities are falling out at that step.

Every business has flows: customers progress from signup to purchase to loyalty. Orders move from placed to shipped to delivered. Leads advance from qualified to closed. Flow Analysis discovers these journeys from your Cartographer metadata, identifies the biggest leak, and auto-drills into why. Cohort Analysis then asks: which groups of customers are succeeding through this flow, and which are failing?

Auto-discovered funnels. Biggest leak identified. Root cause explained.

Discover
~3 sec
AI analyzed your Cartographer metadata

6 entities, 14 segments, 8 temporal columns scanned

Customer Lifecycle92% confidence

Signup → First Purchase → Repeat → Loyalty

customersorders4 stages
Order Fulfillment78% confidence

Placed → Paid → Shipped → Delivered

Lead Pipeline71% confidence

Created → Qualified → Contacted → Won

AI Discovers Your Flows

Cartographer reads your warehouse metadata, entities, status columns, temporal fields, relationships, and suggests business process flows automatically. No manual funnel setup.

Flow Discovery

AI reads your Cartographer metadata and suggests business process flows, no manual funnel setup.

Leak Detection

Automatically identifies the biggest drop-off and classifies severity: critical, warning, or healthy.

Auto Root Cause

One click drills into the biggest leak using KPI Drilldown, get the 'why' without building anything.

Cohort Retention

Group entities by signup month, tier, or behavior, see retention heatmaps across time.

Cohort Comparison

Compare metrics across groups: which customer segment has the highest lifetime value?

Flow x Cohort

Cross-cut: run the same flow for different cohorts. See which group fails at which stage.

Auto-discovered funnels. Biggest leak identified. Root cause explained, under 15 seconds.

Layer 6 / STRATEGIC INTELLIGENCEComing Soon

Your AI Partner for Business Strategy

The foundation is built. The context graph is grounded. The map is alive. The next step: let you talk to it.

The Co-Pilot will bring conversational intelligence to your Decision Context Graph. Ask complex questions in natural language. Get contextual, data-backed answers grounded in everything Cartographer has learned about your business.

Built for human and AI agents alike. The context graph you build today will power the autonomous business operations of tomorrow.

AIMindPalace Co-PilotWhy did revenue drop last week?AIAnalysis complete:1. Returning customers down 2.1%2. Root cause: Email campaign delay→ Recommend: Trigger win-back flowDeep DiveShare ReportSet Alert

Planned Capabilities

"Why did revenue drop last week?" → Contextual, causal explanation

"Is our KPI structure MECE? What's missing?" → Structural validation

"What would happen if we raised prices 10%?" → Scenario simulation

MECE optimization suggestions

Automated strategic briefings

UNDER THE HOOD

Built for the data you actually run on.

The architecture is designed for the warehouses, security posture, and statistical workloads of real enterprise environments.

Warehouse Support

Snowflake
BigQuery
Redshift
Azure SQL
PostgreSQL

Statistical Engine

Python-powered analysis with scipy, statsmodels, pandas

P-values, confidence intervals, effect sizes, rigorous methods

Security

Enterprise-grade security
Role-based access control
Audit logging

Architecture

React + TypeScript frontend

FastAPI Python backend

Supabase for auth/metadata

Claude AI for intelligence

THE DIFFERENCE

A different shape of system.

Faster matters. The bigger shift is what the system is actually doing on the other side of the question.

Capability
Traditional BI
MindPalace
Time to first insight
2-4 weeks
30 seconds
Finding 'why' behind changes
Manual SQL queries
Automatic across all dimensions
Hypothesis testing
You guess what to analyze
AI suggests from patterns
Statistical rigor
Eyeballing charts
P-values, effect sizes, confidence
Business impact quantification
Not included
Dollar impact per finding
New dashboard request
3 weeks + data team time
Click any KPI, instant

Most stacks return numbers. The Decision Context Graph returns the reasoning behind them: which driver moved, by how much, and why it matters, every time the same question is asked.

See the Full Platform in Action

Ready to go deeper? Our team will walk you through every layer of MindPalace and show you how it maps to your specific data challenges.