Revenue Dashboard — The Queens Hotel Gladstone

Replaces the Excel revenue tracking workbook: daily actuals, targets, and historic reporting across the venue's revenue lines (Gaming, Main Bar, Sports Bar, Bistro, South Gladstone, Gladstone Phillip St, Accommodation, Golf).

Stack

• Next.js (App Router) + TypeScript, deployed on Vercel
• Supabase (Postgres + Auth + Storage) — the existing "The Queens" project, shared with the Functions Manager app. Revenue tables are prefixed rev_ and isolated by row-level security; see supabase/migrations for schema.
• pdfjs-dist for PDF text extraction, exceljs for the Golf ledger xlsx export

Report parsers (src/lib/parsers)

Four source reports, sniffed by file content rather than trusted by extension (sniff.ts):

• SwiftPOS "Master Group Sales by Location" (PDF) — trade date is the start of the Reporting Period, not the print/filename date.
• Maxgaming Daily Report (Gaming) — the venue's sole gaming source; Turnover/Revenue plus Card Usage % (gaming floor and POS) and New Members. Dated directly, no offset.
• RMS "Occupancy By No Group" (PDF) — one report, run weekly, gives the full Accommodation picture per day: Occ %, Revenue, and ADR (Nett Avg RevPOR). Each row is matched to the database by its own literal date, never assumed to be "this week".
• Golf booking ledger export (CSV or genuine xlsx) — venue revenue is only the credits to account 4000.

All four were validated against real sample reports.

Getting started

npm install
cp .env.example .env.local   # fill in from the Supabase project settings
npm run dev

Upload flow

Every parsed report lands in a review screen (/review/[documentId]) before anything is written to rev_daily_actuals — extracted figures are editable, and the upload can be discarded instead of committed. Nothing is auto-committed.