Business model
Model the business as it operates
Objects, the links between them, and the logic and actions attached to them, in one model.
Access to Snowflake, Salesforce, SAP, APIs, and documents is not enough. AI also needs to know what the data means, which logic to trust, and which actions are safe.
Inside the ontology
Data & logic
Data, logic, and actions in one object
Three components make an ontology object operational. Each one contributes a layer to the same object, and every agent reads all three.
Data — source datasets ground the object. Logic — forecasts and rules inform the object. Actions — governed operations and audit.
Ontology object · SKU
SKU-4417
Product · Outerwear · Northeast region
Example data
Reorder due in ~2.6 wks
Reorder due in ~1.3 wks
01 Data → properties
| Property | Value | Source | Updated |
|---|---|---|---|
| on_hand | 418372 | ERP | 2s ago |
| unit_price | $18.40 | POS | 14s ago |
| lead_time | 6 days | EDI ASN | 3m ago |
| location | NE-104 | WMS | 1m ago |
02 Logic → Computed
| Computed | Value | Kind | Reads |
|---|---|---|---|
| demand_forecast | 38 / wk41 / wk | FORECAST | sales history |
| reorder_point | 320 | RULE | lead_time · demand |
| stockout_risk | 0.410.72 | ML | on_hand · forecast |
on_hand · baseline projection
crosses reorder_point in ~2.6 wks
220
440
660
reorder_point 320
418
−6 wks
today
+8 wks
03 Actions → registered
trigger_reorder
preconditions ✓
≥ $10K → buyer
write-back → ERP
audit · append-only
in sync ERP change → datasets → forecast → object
Live sync
When your business changes, the ontology changes with it
When source systems change, backing datasets update, dependent logic recomputes, and the state of the business changes with them.
ERP · Inventory
WMS · Receipts
POS · Sales
f_inv_lc_dy
f_sls_it_lc_dy
Inventory
Stockout Rule
Store
Product
Replenishment Agent
Buyer Dashboard
Weekly Report
Surfaces
One operational model. Every AI surface.
Every interface operates against the same governed business context, analytical logic, and available actions.
AI Analytics
AI Analytics
Ask questions about your business.
Get answers from the governed model.
Autonomous Agents
Autonomous Agents
Agents reason over the same objects and take
governed action on them.
Your own AI
Your own AI
API and MCP access bring the same context
and actions into what you build.
One operational model
0%
Model for every system, team, and agent
Every system, one model
ERPs, CRMs, warehouses, phone systems, and documents map to the same objects. A customer is the same customer everywhere.
Conversations become data
Calls, chats, emails, and PDFs become structured facts, attached to the right object.
Logic you can trust
Definitions, metrics, and business rules are written once. Every person and agent uses them the same way.
Bring your own model
Use your preferred models or let us optimize across providers in real time.
Built with your team, one process at a time.
We build the ontology with the people who run the work. It starts with one process and grows as you add more.
Map the business.
We work with your team to name what matters: customers, products, orders, suppliers, and how they relate.
Connect it to live systems.
Each object links to the systems that hold its data, and stays in sync as they change.
Encode how decisions are made.
Metrics, policies, and approval rules go into the model, so AI follows them every time.
Frequently asked questions
Bring transactions, store activity and product sales into the same operational context as the rest of your business.
Bring transactions, store activity and product sales into the same operational context as the rest of your business.
Bring transactions, store activity and product sales into the same operational context as the rest of your business.
Bring transactions, store activity and product sales into the same operational context as the rest of your business.
Bring transactions, store activity and product sales into the same operational context as the rest of your business.
Turn your business context into action
Connect your data, define how the business works, and give AI the context to analyze, decide, and act with confidence.