Properties, listings and tenancies. Everything is time-boxed — a price, a lease, an availability window — so the schema stores periods rather than current values. Every one opens in the free drawDB editor in a click, and is public domain.
3 schemas here, from 13 to 13 tables (39 tables and 481 columns in total), written for Oracle, PostgreSQL. Each one ships an ER diagram you can open and edit, a column-by-column reference, design notes explaining the trade-offs, and SQL for all six dialects.
How a mortgage lender models a file in Oracle: borrowers and co-borrowers with income and liabilities, the subject property and its appraisals, rate locks that expire, underwriting decisions and conditions to clear, a document checklist, closing and disbursement.
How a property portal models its data: properties as durable things versus listings as repeatable campaigns, agents and agencies, ordered media, amenities, price history as rows, viewings, offers and saved searches with alerts.
Landlord operations modelled end to end: buildings and units, leases and renewals, co-tenants and guarantors, rent charges generated per period, payments allocated against them, deposits held and returned, maintenance with vendors, and inspections.
An experimentation platform end to end: experiments and weighted variants, deterministic hash bucketing, exposure events, metric definitions, per-variant aggregates with sufficient statistics, significance results and staged rollout.
How a short-term rental marketplace models its data: hosts and listings, a host-owned per-night calendar, request-to-book versus instant book, reservations, guest payments and host payouts, and two-way reviews that only reveal once both sides have written.
How a marketplace models many sellers listing the same product: a shared catalogue, competing offers, the buy box decision as a log, orders split into per-seller sub-orders, fulfilment, payouts and product-level reviews.
How a product analytics service stores its data: projects and a discovered event registry, raw events with JSONB properties, anonymous and identified users with identity merges, sessions, funnels, cohorts, retention reports and precomputed metric snapshots.
How an API platform is modelled: consumers and applications, hashed API keys with scopes and rotation, upstream services and routes, rate limit policies, durable usage counters, rolled-up aggregates, sampled request logs, plans, quotas and overage billing.
An ATS modelled end to end: jobs with openings and hiring teams, candidates separate from applications, per-job pipeline stages, interviews with panels, structured scorecards, offers with approval chains and a talent pool of near misses.
How an IT estate is tracked end to end: models versus serialised assets, custody history, software licences and seat allocation with compliance counts, warranties and support contracts, maintenance, depreciation, audits and disposal.
A double-entry bank ledger: accounts in a single currency each, balanced journal entries that are never updated, card authorizations that hold funds, transfers kept separate from postings, and frozen statements.
A content management schema that survives contact with editors: posts with an append-only revision history, categories and tags, threaded comments, a real media library, slug redirects and a double opt-in newsletter list.