Part 2 · 2 chapters · ~12 min
Scoping the Smallest Valuable Slice
Vertical slices versus layers, the smallest valuable and learnable release, walking skeletons for risk, cutting scope by audience, case, polish or automation, feature flags to ship partial work, and saying clearly what was deferred.
5
Slices that teach
The smallest valuable slice is the smallest release that solves the problem for some real users and produces evidence about the rest.
VERTICAL SLICES, NOT LAYERS
deliver thin end-to-end value instead of finishing layers one at a time
swipe the figure sideways, or tap expand for full screen
1/5
layers
Building the full database layer, then the full API, then the full UI means nothing reaches users for months, and nobody learns whether it works.
layer by layer: nothing ships for monthsno learning until the end
6
Ways to cut scope
| cut by | example |
|---|---|
| audience | staff first, then one segment, then everyone |
| case | one loan product, the common path; edge cases handled manually at first |
| automation | a concierge version (operations does it manually behind a simple form) before automating |
| polish | a plain but correct UI first; animation later (never cut correctness or accessibility) |
| platform | web first, app later (or the reverse, where users are) |