Connect your PostgreSQL or MySQL database. Ask questions like you'd ask a colleague. Get instant charts, tables, and reports. No SQL required.
SELECT c.name AS category, SUM(oi.quantity * oi.unit_price) AS revenue
FROM order_items oi
JOIN products p ON oi.product_id = p.id
JOIN categories c ON p.category_id = c.id
WHERE oi.created_at >= DATE_TRUNC('quarter', NOW()) - INTERVAL '3 months'
AND oi.created_at < DATE_TRUNC('quarter', NOW())
GROUP BY c.name ORDER BY revenue DESC;
SELECT c.name, c.email, MAX(o.created_at) AS last_order,
COUNT(o.id) AS total_orders, SUM(o.total) AS lifetime_value
FROM customers c
JOIN orders o ON c.id = o.customer_id
GROUP BY c.id, c.name, c.email
HAVING MAX(o.created_at) < NOW() - INTERVAL '90 days'
ORDER BY lifetime_value DESC LIMIT 10;
| Customer | Last Order | Orders | Lifetime Value |
|---|---|---|---|
| Sarah Chen | 98 days ago | 47 | $12,450 |
| Marcus Johnson | 112 days ago | 34 | $8,920 |
| Aiko Tanaka | 91 days ago | 28 | $7,340 |
| David Park | 145 days ago | 22 | $5,680 |
SELECT DATE_TRUNC('month', o.created_at) AS month,
AVG(o.total) AS avg_order_value,
COUNT(o.id) AS order_count
FROM orders o
WHERE o.created_at >= DATE_TRUNC('year', NOW())
GROUP BY month ORDER BY month;
Stop waiting for your dev team to write SQL. Get answers in seconds.
Ask in plain English: "What were our top products last month?" SightQL writes the SQL, runs it, and shows the results — with charts.
PostgreSQL and MySQL support. Not just CSV uploads — connect to your production (read-only) or staging database for real-time answers.
Bar charts, line graphs, pie charts, tables — auto-selected based on your data. Export as PNG or share a live link with your team.
SightQL never modifies your data. Read-only connections only. Your database credentials are encrypted at rest. Only your schema (table/column names) is sent to AI — never your actual data. SOC2 in progress.
Upload your data dictionary or let SightQL auto-detect table relationships. The more you use it, the better it understands your business terms.
Set up daily/weekly reports delivered to your inbox or Slack. "Send me MRR numbers every Monday at 9am" — done.
The $50-100/mo sweet spot nobody else fills.
| Feature | Julius AI | Steep | Metabase | ThoughtSpot | SightQL |
|---|---|---|---|---|---|
| Natural language queries | ✓ | ✓ | ✗ | ✓ | ✓ |
| Live database connection | ✗ | ✓ | ✓ | ✓ | ✓ |
| Non-technical UX | ✓ | ✓ | ✗ | ✗ | ✓ |
| Charts & dashboards | ✓ | ✓ | ✓ | ✓ | ✓ |
| SMB price (/mo) | $20-49 | $50 | $85 | $1,250 | $79 |
| AI-generated insights | ✗ | ✗ | ✗ | ✓ | ✓ |
| Scheduled reports | ✗ | ✗ | ✓ | ✓ | ✓ |
ChatGPT can write SQL, but it can't connect to your database, run queries against your actual data, or generate charts from real results. Every time you use ChatGPT for data questions, you're copy-pasting schema definitions, pasting results back into a spreadsheet, and building charts manually. SightQL does all of this in one step: you ask a question, it connects to your DB, runs the query, and shows you a chart — instantly. It also remembers your schema, builds query history, and lets your whole team ask questions without sharing database credentials.
SightQL connects to your database with read-only access only. It never modifies, deletes, or exports your data. When you ask a question, only your schema (table and column names) is sent to the AI to generate SQL — your actual data never leaves your database. The generated SQL runs against your database locally. You can audit every query in your history.
SightQL uses schema introspection to understand your column names, data types, and relationships before generating SQL. You can mark which tables to include or exclude. While no NL-to-SQL tool is perfect on 100+ table schemas, SightQL handles most real-world SMB databases (5-50 tables) reliably. Start with the free tier and see for yourself.
Every query SightQL generates shows you the SQL before and after execution. You can see exactly what ran, verify the results, and edit the SQL if needed. SightQL runs in read-only mode, so a wrong query can't damage your data. Over time, SightQL learns from your corrections and gets better at understanding your specific schema.
No signup required. Start learning today.
Copy-paste queries for e-commerce, SaaS, and restaurant analytics.
10-question assessment. Get your analytics score and action plan.
One actionable data tip per day. Zero asks, pure value.
How much are manual reports costing you? Find out in 2 minutes.
Start free. Upgrade when you're ready.
SightQL launches soon. Join the waitlist and get free early access — first 10 users get Pro features free for 3 months.