How much goes into proving it works
The Tests slice is the share of Engineering Throughput that lands in test code: unit and integration suites, end-to-end specs, fixtures, harnesses. It used to be folded into whatever production file it accompanied. It is now measured on its own, so you can see it directly — and there is no target number. A higher Tests share is not automatically better, and a lower one is not automatically worse.
TESTS SHARE BY DAY
Each bar is one day's tests share — that day's Tests ETV divided by total ETV. The line is the 90-day trailing rolling average — the underlying trend.
Baseline
20.1%
90-day rolling avg ending 2025-04-01
Latest
30.8%
90-day rolling avg ending 2026-08-22
Change (percentage points)
+10.8 pp
from 20.1% to 30.8% across the 90-day rolling series
How to read it
Faint barsare each day's raw Tests share (that day's Tests ETV divided by total ETV). The bold line is the 90-day trailing rolling average — the trend that matters.
The headline is the change in percentage points between the first and the most recent endpoint of the 90-day rolling-average line. We leave it uncoloured on purpose: this is the only work slice alongside Docs where neither direction is a verdict. Read it against what the team was building at the time.
Why it matters
Tests share tells you where the confidence in a codebase is being built, not how good the codebase is. A rise can mean a team hardening a fragile area, or it can mean re-testing ground it already covered. Neither reads off this number alone.
It is most useful next to Fixes share. Tests rising while Fixes falls is coverage paying for itself. Both rising together usually means the tests are arriving after the bugs rather than ahead of them.
How we measure tests share change, and how to read the trend.
Show methodologyHide methodology
How we measure tests share change, and how to read the trend.
Show methodologyHide methodologyWhat counts as Tests
Any change to a file we identify as test code: *_test.go, *.spec.ts, *.test.tsx, __tests__/ and tests/ trees, snapshots, fixtures and test-only helpers. Classification is per file, so a commit that adds a feature and its tests splits between Features and Tests.
How shares are computed
For each day we sum the Engineering Throughput of merged commits, split by category, and divide by the daily total across all five categories. The rolling-average overlay is a 90-day trailing mean. The most recent day is dropped if it looks like a partial sync.
How the change is computed
We compare the first and the most recent point of the 90-day rolling series — the leftmost and rightmost values of the bold line on the chart — and report the difference in percentage points. Both endpoints are single 90-day rolling readings, not calendar-quarter averages, so they stay aligned with the rest of the site's rolling-90d convention.