---
title: "UNIQUE constraint"
description: "A UNIQUE constraint ensures that all values in a column, or combination of columns, are distinct across every row in a table."
canonical: "https://motherduck.com/glossary/unique-constraint/"
related:
  - title: "Constraints | MotherDuck Docs"
    url: "https://motherduck.com/docs/sql-reference/duckdb-sql-reference/constraints/"
  - title: "Analyze JSON Data Using SQL and DuckDB"
    url: "https://motherduck.com/blog/analyze-json-data-using-sql/"
  - title: "CREATE INDEX | MotherDuck Docs"
    url: "https://motherduck.com/docs/sql-reference/duckdb-sql-reference/duckdb-statements/create-index/"
---

# UNIQUE constraint

> A UNIQUE constraint ensures that all values in a column, or combination of columns, are distinct across every row in a table.

## Overview

A `UNIQUE` constraint prevents duplicate values from being inserted into a column or set of columns. It's commonly used for things like email addresses, usernames, or external identifiers where duplicates would indicate a data error, even though the column isn't the table's primary key.

<glossary-callout guide="duckdb-cheatsheet-full" />

## UNIQUE vs. PRIMARY KEY

A table can have only one `PRIMARY KEY`, and primary key columns implicitly disallow `NULL`. A table can have any number of `UNIQUE` constraints, and unlike primary keys, most databases (following the SQL standard) treat `NULL` as unequal to any other value — including another `NULL` — so multiple `NULL`s are typically allowed in a `UNIQUE` column.

## Composite Uniqueness

`UNIQUE` can also apply across a combination of columns, so that the combination must be distinct even if individual columns repeat:

```sql
CREATE TABLE enrollments (
    student_id INTEGER,
    course_id INTEGER,
    UNIQUE (student_id, course_id)
);
```

## DuckDB Support

DuckDB supports `UNIQUE` constraints on a single column or a combination of columns:

```sql
CREATE TABLE users (
    id INTEGER PRIMARY KEY,
    email VARCHAR UNIQUE
);
```

Under the hood, DuckDB automatically builds an Adaptive Radix Tree (ART) index for any column with a `PRIMARY KEY` or `UNIQUE` constraint, which is also what makes point lookups on that column fast in addition to enforcing uniqueness.