Architecture · Enterprise Intelligence

Enterprise
intelligence
architecture.

Building an intelligence-driven business — across the enterprise, the product, and the customer.

Eight mental models · Three surfaces · One governed bus Before you think about technology

Part one · The mental models

Eight ways to think about intelligence — before you think about technology.

01Governance as the objective functionEvery optimization needs an objective. Leadership's first job is to define the summit — so AI accelerates one shared goal instead of amplifying conflicting ones.
02The layered enterpriseDesign from the highest level of abstraction down, so every tactical decision traces back to the objective. Intelligence lives at the core.
03Enterprises are information systemsA company is information flowing between people making decisions. Every department is a black box: input → transformation → output.
04Functions before departmentsDepartments are human conventions. Functions create value — and functions are reusable, exposed by one team and consumed by many.
05Intelligence is a layer, not an appDon't ask where to install AI. Ask where intelligence should exist — in products, workflows, decisions, planning, and customer interactions.
06Judgment is the scarce resourceExecution is becoming cheap — code, content, analysis, automation. Judgment is becoming decisive: which market, which feature, which customer.
07AI should always compoundIsolated pilots carry nothing forward. Knowledge, patterns, governance, and infrastructure should compound — each deployment makes the next easier.
08Organizations transform by becoming legibleThe centerpiece. Before you automate a business you must understand it; before you understand it you must model it.

Mental model 08 · The centerpiece

Organizations are not transformed by AI. They are transformed by becoming legible.

First, become legible

Before you can automate a business, you must understand it. Before you understand it, you must model it — what each function does, consumes, produces, and decides, and where judgment is required.

Then, AI becomes obvious

Once the organization is legible, AI becomes a clear and logical implementation. You can see where copilots belong, where automation is safe, where prediction adds value, and where human judgment must remain.

Mental model 02, in depth

The layered enterprise.

Start at the top and work your way down. Executives design at the highest level of abstraction; intelligence lives at the core.

6Executive StrategyWhat objective are we optimizing toward?
5Business FunctionsWhat capabilities must the organization possess to achieve it?
4WorkflowsHow are those capabilities delivered today?
3Tasks & DecisionsWhat decisions are made at each step?
2Applications & InformationWhere is the information required to make each decision well?
1IntelligenceHow can AI improve those decisions?

The map · Surfaces of intelligence

Two surfaces, one system.

Intelligence improves how the company operates, and what the company sells. The same discipline, applied inward and outward.

Surface 01

Enterprise Intelligence — how the business operates

Every function acquires a decision layer:

Finance — forecasting, margin optimization, anomaly detection · Sales — qualification, account planning, renewal prediction · Customer Success — churn prediction, health scoring · Support — routing, copilots, predictive escalation · Product — prioritization, telemetry, roadmap simulation · Engineering — copilots, review, incident investigation · Operations — scheduling, capacity, exception detection · HR · Legal · Data — search, contract review, data quality.

Surface 02

Product Intelligence — what the company sells

From telling customers what happened to telling them what to do:

Predictive maintenance — “These assets will likely need service within their next operating window.” · Utilization — “Reallocate this resource to avoid renting another.” · Efficiency — “You're spending more than comparable operations due to idle behavior.” · Safety — “This pattern correlates with elevated risk.”

No menus. No dashboards. No reports. Just the decisions that deserve attention.

The wiring · Role Programming Interface

The real-time enterprise.

One engine per role is leverage. The whole organization wired together is transformation — every department exposes functions the others call machine-to-machine, over a single governed bus.

01 · The problem

The translation tax

Today, one department needs something from another and a human relays it — losing fidelity and time at every hop.

02 · How it works

Exposed functions

Each department publishes callable functions — request_quote, validate_budget, flag_churn_risk. The org becomes an API.

03 · The result

Real-time at the edge

IT governs one consolidated data store and one contract. Departments subscribe only to approved data; every call is routed, enforced, and audited.

Sales exposes account state and expansion signals. Finance exposes budget validation and discount approval. Customer Success exposes account health. Product exposes usage telemetry. The RPI bus routes the call, enforces the contract, governs the data, and writes the audit log.

Remove the relay between every department, and the whole organization begins to move at the speed of its systems — and people move at the speed of judgment.

Make it real

The method that turns this into an architecture.

The approach →