Operating model · Mantle
The shared model underneath decisions, automation and AI.
Most enterprise technology problems start below the interface. Systems disagree about customers, outlets, dealers, orders, rules, metrics or ownership. A Mantle is the governed model that resolves those definitions and gives every downstream system the same operating truth.
01
Specification
Not a metaphor. The specification your systems run against.
A Mantle names the real objects a business runs on and how they relate. It is readable by executives, precise enough for engineers, and concrete enough to build applications, analytics, automation and AI on top of it.
Entities
The nouns: dealer, outlet, candidate, order, asset, territory. Named once, defined once, referenced everywhere — not redefined slightly differently in five systems.
Relationships
How the nouns connect: a dealer operates an outlet, an outlet raises an order, an order is placed by a customer. Get this wrong and every report downstream disagrees with every other one.
Rules
What counts as valid, what triggers a decision, what a metric actually means — written down once instead of buried separately in five people’s heads and three codebases.
Documentation
The reasoning behind the model, in plain language, so a new engineer on the client’s own team can understand it in an afternoon rather than reverse-engineering it from code.
02
How it’s built
We define the operating model before we build the visible system.
This is the step many projects skip. They build dashboards, apps or AI workflows first, then discover the business definitions underneath are unstable. We start with the model so the visible system has something trustworthy to stand on. See the method for how that works.
03
Real Mantles
Examples of operating models we have built.
Next