Collaboration & Process › Agile & Delivery Process

Vertical Slice

A feature built through every layer, end to end.

A vertical slice is a small piece of a feature that works through the relevant layers, from the user or caller to the data or service behavior and back. It provides an end-to-end outcome instead of completing one technical layer in isolation.

For a new alerting feature, a slice might detect one condition, deliver one notification, and let one user acknowledge it. Later slices can add alert types, routing, and configuration. This exposes integration risks early: an API shape, permission rule, or data-quality assumption can be tested before many layers depend on it.

A slice is not necessarily a prototype or a shortcut around quality. It should meet the team’s security, correctness, and operability expectations for its intended release. Some enabling work does not directly show up in the user interface; keep it tied to a clear first outcome rather than building a broad platform in advance.

Backend, frontend, and data engineers should agree what “usable” means across their interfaces and test the path as a whole. Slices can be harder to divide when teams own isolated layers, so use them to reveal dependencies and coordinate work rather than to demand artificial parallelism. See iterative delivery and MVP.

Check the workflow with a recent piece of work rather than relying on an abstract rule. If the practice adds a queue or hides blocked work, adjust it; if it helps the team finish and learn, keep it. Backend developers can flag service constraints, frontend developers can check user-facing behavior, and data engineers can surface pipeline or data-quality dependencies.

After trying it, review several completed items and adjust the practice if it created a queue, hid a blocker, or failed to produce a useful outcome.