adoption-report

Adoption Report

Context

Measures whether the system is actually used, keeping coverage and adoption strictly separate (see adoption-measurement). On the first run it establishes a baseline rather than inventing a trend. At-risk areas are framed as enablement opportunities, not team failings.

Configuration

Reads (all optional): target_repo; stack.component_roots; output.trend_dir (for baseline/trend via session-memory); severity_overrides.

Auto-pull integrations

None required. If GitHub is configured, MAY use history to compute the adoption trend; without it, report coverage only and note the limitation.

Step 0 — Baseline check

If there's no prior snapshot, this is a baseline run: compute coverage, save it (via session-memory), and say explicitly that trend will appear next run. Don't fabricate a trend.

Step 1 — Coverage (structural)

Of the surfaces that could use the system, how many do: shared-component usage vs page-local one-offs that duplicate a shared atom; token usage vs raw values. Break down by area (route/feature folder).

Step 2 — Adoption (trend) — only with history

Direction over time per area: moving toward or away from the system. Requires prior snapshots or git history; otherwise mark N/A.

Step 3 — At-risk areas

Areas with low shared usage and high local one-offs — most likely to drift, best ROI for enablement (docs, onboarding, comms).

Produce the report

## Adoption Report — <repo>  [baseline | trend since <date>]

**<🟢/🟡/🟠/🔴> Headline.** <Coverage posture + the area to help first.>

### Coverage (today)
| Area | Shared usage | One-offs | Coverage |
|------|--------------|----------|----------|
| jobs | 41 imports | 3 | 🟢 |
| care-journeys | 12 | 9 | 🟠 |

### Adoption (trend)
<per-area direction — or "baseline run, no trend yet">

### At-risk areas
- <area> — low shared usage + high one-offs → enablement: <docs/onboarding/comms>

### Scope
- Inspected: <areas, import scan>
- Not inspected: runtime usage analytics
- Assumed: baseline on first run; trend needs history

Closing note

Coverage and adoption answer different questions — a high-coverage area still drifting is a different problem from a low-coverage area improving. Keep them apart, and treat at-risk areas as places to help, not to name and shame.

Quality checks

• Coverage and adoption reported separately, each with its own label.
• First run is a labelled baseline; no fabricated trend.
• At-risk areas framed as enablement, with the specific intervention.
• Output-discipline self-check passed.