Sooq Analytics event store

Large tables for query-performance lessons: 20,000 users, 300,000 tracking events and 100,000 orders during 2025, plus a small product list. Only primary keys are indexed, so filters and joins on other columns start as sequential scans; you create the indexes and compare plans. Columns are chosen to show selectivity: email is almost unique, status is dominated by one value, and event_type and user activity are heavily skewed.

The documentation is open to everyone. Querying this dataset in the playground is included with Formula Quest membership; lessons that use it follow their own access.

EXPLAIN
SELECT *
FROM users
WHERE email = 'user12345@example.net';

EXPLAIN ANALYZE
SELECT event_type, count(*)
FROM events
WHERE user_id = 42
GROUP BY event_type;

EXPLAIN
SELECT order_id, user_id, amount
FROM orders
WHERE status = 'pending';

SELECT event_type, count(*) AS events,
       round(100.0 * count(*) / sum(count(*)) OVER (), 1) AS pct
FROM events
GROUP BY event_type
ORDER BY events DESC;

SELECT occurred_at::date AS day, count(*) AS purchases
FROM events
WHERE event_type = 'purchase'
  AND occurred_at >= '2025-03-01' AND occurred_at < '2025-04-01'
GROUP BY day
ORDER BY day;