SQL Joins

Match rows across tables on a key; the join type decides which non-matches survive.

SQL · Beginner · ~4 min

In plain English

Stitch two tables together on a shared key. Inner keeps only matches; left keeps everything on the left and fills the gaps with NULL.

Why it's worth your time

It's the highest-frequency interview topic in SQL, and the most common source of quietly wrong numbers.

If you remember three things

INNER: matches only. LEFT: all of the left, NULLs on the right
A duplicate key on either side multiplies rows
Filtering a LEFT JOIN's right table in WHERE turns it into an INNER JOIN

Overview

A join stitches rows from two tables into one wider row, matched on a shared key — usually a primary key on one side and a foreign key on the other. INNER JOIN keeps only matched rows; LEFT, RIGHT, and FULL OUTER joins additionally keep unmatched rows from one or both sides, filling missing columns with NULL.

In an interview

Joins combine tables on a join condition. INNER keeps rows that match on both sides. LEFT keeps every left row and NULL-fills where the right has no match; RIGHT and FULL extend that to the other side. Index the join columns so matching is a lookup, not a scan.

Production defaults

Check cardinality: before joining. One-to-many is fine; many-to-many silently inflates every count
LEFT JOIN filters: put conditions on the right table in ON, not WHERE, or you lose the unmatched rows
Verify: compare row counts before and after. An unexpected increase means a fan-out

What breaks

Row count exploded after a join — Duplicate keys on one side. Deduplicate or aggregate before joining.
LEFT JOIN behaving like INNER — A WHERE condition on the right table's column. Move it into the ON clause.

Watch it explained

Inner Join, Left Join, Right Join and Full Outer Join in SQL Server | SQL Server Joins — Questpond, 8:11