OTO
examples

A product through its lifecycle

The question people ask: what problem is this product solving, for whom, what did we decide, what is built and tested, and is it delivering the value we said it would?

What feeds the graph: the problem statement, interviews and journeys, requirements, architecture decisions, the API and event contracts, test reports, deployment records, and the KPI reviews. Most of it exists already, in six different tools, and none of it is connected.

The six phases, one graph

EnvisionDiscoverDesignBuildOperateRealize
Phase What enters the graph
Envision the problem statement, the opportunity, the stakeholders, the outcome and its target
Discover personas, journeys, use cases, requirements including the non-functional ones, policies
Design bounded contexts, the context map, the system design, the decisions, APIs and events, upstream and downstream systems
Build components, integrations, agents, tests and test data, the gates a release passes
Operate deployments, incidents, runbooks, what changed in production and when
Realize the KPIs and OKRs, the value realised against the target, the go-forward decision

Each phase mints its own nodes, and the traceability runs the whole way: an outcome is realised by a journey, a journey by use cases, a use case by requirements, a requirement by components, a component by tests. When the target is missed, the graph says which link is missing.

An internal tool is a product

A platform team's deployment tool, a data team's report factory, an operations team's intake portal: each has users, journeys, requirements, decisions and a value it was built to deliver. Treating them as products with the same graph is what makes the mesh answer across the company rather than only across the products it sells.

Studio

The front page shows each phase in Studio for one product of a sample insurer: the problem statement with its sources, a journey told as an arc, the architecture of the Design phase with its contexts, decisions, components and stack, the acceptance tests, and the value realised. Studio is the face; the graph underneath is an oto project per product.

How it is used

  • The product owner asks which requirements have no test, and which outcome has no journey that realises it.
  • The architect records a decision once; the spec skill reads it into the next design instead of re-deciding it.
  • The sponsor reads the Realize phase: the value against the target, rolled up from the graph, with the source of every number.

The commands and the skills

The product-delivery ontology is in preparation; until it is in the catalog, the software-architecture and organization-process ontologies merged cover Design, Build and Operate:

oto init --name "Report Factory" --ontology software-architecture,organization-process --project factory

Skills: build-knowledge-base, spec, evaluate. The report automation example is the first pack built on it.