fte-dashboard

FTE Capacity Dashboard

An interactive dashboard for the FTE (Level of Effort) / FTE capacity reports produced by the fte-report GitHub Action. It turns the generated CSVs into a spreadsheet-style capacity matrix, charts, filterable tables, and an in-browser what-if editor so PMs can visualize over-allocation and model re-balances without touching GitHub.

Static single-page app (Vite + React + TypeScript). No backend, no secrets.

Two tabs (switch bottom-left)

1. Capacity Matrix (default, read-only) — a people × objectives pivot. Objective columns are grouped into collapsible Initiative groups (click the header to fold a group into a single Σ Initiative subtotal column); a Project dropdown scopes the view. A person who holds more than one role gets one row per role: the Per role column sums that role and the bold Per person column (a merged cell spanning the person's role rows) sums all of them. Footer rows give Total per objective and Total per initiative. Each objective header links to its GitHub ticket.
2. What-if Dashboard — the original headline cards, per-person / per-role charts, and the editable allocations table (below). Reset/Export live in the header on this tab only.

Grouping comes from the project / initiative / team columns in fte_allocations.csv (read from the board by the generator). CSVs written before those columns existed group under "Unspecified" rather than breaking.

How it gets data

The reports are published by the Action to the public branch fte-report/all-pis as reports/fte_*.csv. This site fetches those files at runtime from raw.githubusercontent.com (with a cache-buster), so it always shows the latest report without a redeploy. If that branch/file is unavailable, it falls back to the snapshot bundled in public/data/. A badge in the header shows which source is live (● live vs ● snapshot) and the report's generation time.

The fte_allocations.csv rows are the source of truth; fte_by_person.csv / fte_by_role.csv are the generator's aggregates, used only to validate that the in-browser recompute reproduces the Python numbers exactly (the ✓ matches baseline badge).

Features

• Headline cards — objectives, people, over-allocated (person·PI) pairs, total raw & weighted FTE. Recompute live as you edit.
• Capacity by person — raw vs weighted FTE per person with a dashed 1.0 full-time cap; bars over 1.0 turn red.
• Capacity by role — FTE demand per role (staffing signal).
• Allocations — what-if editor — filter by person / role / PI / over-allocated, click through to the GitHub objective, and edit any FTE to model a re-balance. Person, role and headline totals update instantly. Edits stay in your browser and are never written back to GitHub.
• Export — download the (adjusted) allocations, by-person, or by-role CSV. Files are suffixed _whatif when edits are active.

Local dev

cd fte-dashboard
npm install
npm run dev        # http://localhost:5173
npm run build      # -> dist/
npm run typecheck  # tsc --noEmit

Deploy on Netlify (one-time)

Build config lives in ../netlify.toml (base fte-dashboard, publish dist). No env vars or secrets are needed — the data is public.

1. In Netlify, Add new site → Import from Git, pick Disasters-Learning-Portal/disasters-aws-conversion, branch main. netlify.toml supplies the build settings. Netlify auto-deploys on push to main.
2. Seed the live source once: run the FTE Capacity Report Action via Run workflow with pi = "All PIs" so the fte-report/all-pis branch exists. Until then the site serves the bundled snapshot automatically.

After that, each weekly report run updates the numbers on next page load — no redeploy.

Keeping the recompute in sync

The aggregation in src/compute.ts mirrors .github/scripts/generate_fte_report.py. Unedited rows reuse the generator's authoritative per-row weighted_fte (computed from the full-precision PI fraction, which the CSV stores only rounded), so the dashboard reconciles exactly with the report. If the generator's columns or FTE math change, update compute.ts and re-check the ✓ matches baseline badge.