How intent flows down and learning flows up

In the previous articles, we introduced the Intent Hierarchy and explored how intent emerges across organizational, domain, and project levels. These concepts describe how intent is structured and how new intent appears. But they do not yet explain how organizations learn and improve over time.

Intent-Driven Development is not just a way of writing specifications or governing delivery. It is a model for how organizations learn. In an intent-driven organization, intent flows down the hierarchy through inheritance, while learning flows back up through measurement, feedback, and promotion. Over time, this allows the organization to evolve its standards, architecture, and ways of working based on real delivery experience rather than top-down prediction.

Traditional organizations are hierarchies of authority.
Intent-driven organizations are feedback systems.

Intent flows down the hierarchy

In the Intent Hierarchy, intent exists at multiple levels: organization, domain, and project. Higher-level intent provides direction, constraints, standards, and governance, while lower-level intent provides specificity and implementation detail.

Organizational intent might include regulatory requirements, security standards, architectural principles, or platform constraints. Domain intent may define domain models, integration patterns, or shared services. Project intent defines the specific goals, constraints, and outcomes for a particular initiative.

This intent flows down the hierarchy through inheritance. Projects do not start from nothing; they inherit organizational and domain intent automatically. This ensures alignment, compliance, and consistency without requiring every team to rediscover the same constraints and standards repeatedly.

This is the downward flow of intent: direction, constraints, and standards flowing from the organization into execution.

Execution and measurement

Execution happens at the project level. This is where systems are built, services are deployed, features are delivered, and outcomes are produced. Increasingly, execution may be performed not just by humans, but by automation, platforms, and AI agents.

But execution alone is not enough. Execution must be measured. Without measurement, the organization cannot learn. Measurement provides feedback on whether intent is being achieved and whether the chosen approaches are effective.

Measurement might include reliability metrics, performance metrics, cost metrics, delivery speed, user outcomes, incident frequency, or business KPIs. These measurements provide evidence about what works and what does not.

Execution produces outcomes.
Measurement produces learning.

Learning flows up the hierarchy

Once outcomes are measured, the organization can learn. This learning flows upward through the hierarchy.

Teams discover better ways of structuring services, better testing strategies, better deployment patterns, better observability approaches, or better ways of organizing work. These discoveries start at the project level, because that is where execution and reality meet.

When these approaches prove successful and repeatable, they can be reviewed and promoted into domain intent so that other projects can inherit them. If the same patterns prove valuable across multiple domains, they may eventually be promoted into organizational intent and become standards.

This is sometimes called intent promotion, but more broadly it is organizational learning. The organization observes what works, validates it through evidence, and then incorporates that learning into higher-level intent so that future projects benefit from it automatically.

Intent flows down the hierarchy, but learning flows up the hierarchy.

Together, these two flows create a feedback system that allows the organization to evolve over time.

Intent is both imposed and discovered

Not all intent originates inside the organization.

Some intent is imposed from the outside world. Regulatory requirements, legal obligations, security standards, accessibility requirements, and industry compliance frameworks do not emerge from delivery teams. They are imposed on the organization by the environment in which it operates.

This leads to an important principle:

Intent is both imposed and discovered.

Imposed intent typically enters at the organizational level and flows down through inheritance. Discovered intent typically begins at the project level and flows up through learning and promotion.

This means intent moves through the organization in multiple directions:

• Imposed intent flows down from the external environment
• Organizational and domain intent flow down through inheritance
• Learning flows up through measurement and promotion
• Execution happens at the project level

This multi-directional flow is what allows the organization to remain compliant, aligned, and continuously improving at the same time.

The changing role of architecture

This model also changes the role of architecture and architects.

In traditional models, architecture is often defined centrally. Standards are written, guardrails are created, and projects are expected to comply. Over time, this can lead to a disconnect between architecture and delivery, where guidance is created by people far removed from the realities of building and operating systems.

In an intent-driven organization, architecture evolves from learning. Successful patterns discovered in projects are promoted upward and become domain or organizational intent. Over time, architecture becomes the accumulated result of what the organization has learned from delivery.

This leads to a simple but powerful reframing:

Architecture is promoted intent.

Architecture is no longer primarily a set of diagrams and standards created in isolation, an ivory tower. It becomes the mechanism by which organizations capture successful patterns, scale them across teams, and ensure alignment without preventing experimentation.

Architecture shifts from prediction to promotion.
From control to governance.
From ivory tower to learning system.

Organizations that learn

When intent flows down and learning flows up, the organization behaves like a learning system rather than a static hierarchy.

Teams experiment within projects.
Execution produces outcomes.
Measurement produces evidence.
Learning flows upward.
Successful patterns are promoted.
Future projects inherit improved intent.
External constraints continue to flow downward.

Over time, the organization’s intent improves and evolves.

Instead of standards being written once and slowly becoming outdated, organizational intent continuously evolves based on real delivery experience and external constraints.

The hierarchy does not just distribute intent, it learns.

The bigger idea

This points to a broader shift in how organizations operate in the age of automation and AI.

Organizations used to manage people.
Then they managed processes.
Then they managed software.
Next they will manage agents.
But what they really need to manage is intent.

When execution is increasingly automated, the most important human responsibility is not performing execution, but defining, governing, and evolving intent. Human intent.

Intent-Driven Development provides a structure for doing this. The hierarchy organizes intent, inheritance distributes intent, execution delivers outcomes, measurement provides feedback, learning flows upward, and promotion evolves organizational intent over time.

An intent-driven organization is not just a hierarchy.
It is a feedback system for evolving intent.

Intent flows down.
Learning flows up.
Architecture is promoted intent.
Govern Intent. Delegate Execution.

Next Article ->

Intent-Driven Development – Humans have Roles, Agents have Functions

For decades, organisations were structured around the work humans had to do manually. But if agents can write code, run tests, deploy systems, and operate platforms, then organisations may no longer be structured around execution at all. Humans define intent. Agents enact it.