One-to-One Relationship in ERD Design

When One Record Pairs with Exactly One Other

A one-to-one (1:1) relationship means that each record in table A is associated with at most one record in table B, and vice versa. This is less common than one-to-many but important in specific design scenarios.

Why Split into Two Tables?

You might wonder: "If it's one-to-one, why not put everything in one table?" There are several valid reasons:

1. Performance: Separate rarely-accessed columns (e.g., user profile details) from frequently-accessed ones (e.g., login credentials)
2. Security: Isolate sensitive data (e.g., SSN, payment info) into a separate table with different access controls
3. Optional data: When the related data doesn't exist for every record (e.g., not every user has a profile)
4. Legacy integration: When merging data from different systems

Implementation

The foreign key goes in the "dependent" table (the one that can't exist without the other), and it must have a UNIQUE constraint:

CREATE TABLE users (
  id SERIAL PRIMARY KEY,
  email VARCHAR(320) NOT NULL UNIQUE
);

CREATE TABLE user_profiles (
  id SERIAL PRIMARY KEY,
  user_id INT NOT NULL UNIQUE,
  bio TEXT,
  avatar_url VARCHAR(512),
  FOREIGN KEY (user_id) REFERENCES users(id)
);

The UNIQUE constraint on user_id is what enforces the "one" on the profile side. Without it, the relationship would be one-to-many (one user could have multiple profiles).

Alternative: Shared Primary Key

Another pattern uses the same primary key value in both tables:

CREATE TABLE user_profiles (
  user_id INT PRIMARY KEY,
  bio TEXT,
  FOREIGN KEY (user_id) REFERENCES users(id)
);

Here user_id is both the primary key and the foreign key, guaranteeing a strict 1:1 mapping.

In the ERD Editor

Create both entities, then add a one-to-one relationship between them. The diagram shows single line markers on both sides of the relationship line, visually distinct from the crow's foot used in one-to-many relationships.