A Shared Vocabulary for Team Design
Team Topologies defines four team types. Every team in an organization should be identifiable as one of these types. The types are not job titles — they are missions that determine what a team focuses on, what cognitive load they carry, and how they interact with others.
1. Stream-Aligned Team
The Stream-aligned team is the primary team type — the one that directly delivers value to customers. "Stream" refers to a continuous flow of work aligned to a business domain, product, or user journey.
Examples: the "Checkout Team," the "Mobile App Team," the "Payments Team."
- Mission: Deliver value to customers as quickly and safely as possible.
- Cognitive load focus: Entirely on the business problem — not on infrastructure plumbing.
- Ownership: Full end-to-end ownership of their service: build it, run it, fix it.
If a Stream-aligned team is spending 40% of their time wrestling with infrastructure, their cognitive load is too high. That is a signal that supporting teams are not doing their job.
2. Platform Team
The Platform team exists to serve Stream-aligned teams. They build and maintain the internal platform — the "paved road" — that makes it easy for Stream-aligned teams to do their work without needing deep infrastructure expertise.
- Mission: Enable Stream-aligned teams to deliver with substantial autonomy.
- Cognitive load focus: The underlying technology stack — Kubernetes, cloud networking, IAM, databases, CI/CD tooling. They absorb this complexity so others don't have to.
- Product mindset: They treat the platform as a product and Stream-aligned teams as their customers.
3. Enabling Team
The Enabling team is composed of specialists in a specific technical domain — security, performance engineering, UX research, accessibility, or a new technology the organization is adopting.
- Mission: Help Stream-aligned teams overcome obstacles and build new capabilities.
- Interaction: Temporary and targeted. They embed with a Stream-aligned team, teach the necessary skills, and then move on. They do not own the software; they uplift the team that does.
4. Complicated Subsystem Team
Some parts of a system are so mathematically or technically complex that asking a normal Stream-aligned team to own them would consume their entire cognitive load budget. Think: a video codec, a financial reconciliation engine, a bespoke ML inference system.
- Mission: Build and maintain a highly specialized subsystem.
- Interaction: They wrap their complexity in a clean API and offer it as a service. Stream-aligned teams consume it without needing to understand the internals.