A record in a database is a single collection of related values stored together as one unit. In a relational database, a record is usually represented as one row in a table. If a table stores customers, each customer row is one record.
That definition sounds simple, but records matter because they are the level where most real work happens. Applications create records, update records, join records to related records, and delete records that are no longer needed. Once you understand what a record is, it becomes much easier to understand tables, fields, primary keys, and SQL operations.
Quick Answer
A record is one row of related data in a table.
customer_id
name
email
city
101
Ana Ruiz
[email protected]
Madrid
102
Marcus Lee
[email protected]
Chicago
103
Sana Khan
[email protected]
Toronto
Each row above is a separate record. The table defines the structure. The record holds the actual values for one entity.
What Is a Record?
A record stores all the values that describe one thing. That thing might be a customer, an order, a product, a hospital visit, or a support ticket. The database groups those values into fields, and the full set of field values becomes the record.
A spreadsheet analogy helps. If the spreadsheet is the table, the columns are the fields, and one filled-in row is the record. Databases are more powerful than spreadsheets, but this mental model is useful when you are first learning the concept.
In practical database design, records should represent one clear business entity. If a table mixes several entities together, the records become hard to search, validate, and maintain.
Record Structure: Fields, Data Types, and Primary Keys
A record is made up of fields. A field is one piece of data, such as a name, date, quantity, or status. Each field usually has a data type, which tells the database what kind of value is allowed.
Field
Example Value
Typical Data Type
order_id
5001
INTEGER
order_date
2026-04-10
DATE
total_amount
149.99
DECIMAL
is_paid
true
BOOLEAN
Most production tables also define a primary key. A primary key uniquely identifies each record so the database can refer to the exact row without ambiguity. Without a reliable primary key, updates and relationships quickly become error-prone.
If you need a wider foundation before going deeper, read What Is a Database? Definition, Types, Uses, and Examples.
How Records Relate to Each Other
Records rarely live in isolation. A customer record might relate to many order records. A teacher record might relate to many course records. A patient record might relate to many visit records.
These links are usually implemented with foreign keys. A foreign key stores the primary key from another table so the database can connect related records.
orders.order_id
orders.customer_id
orders.total
5001
101
149.99
5002
101
39.00
5003
102
220.00
Here, the order records point back to customer records through customer_id. That creates a one-to-many relationship: one customer can have many orders.
Records in SQL: Create, Read, Update, and Delete
Most database work with records falls into four actions: create, read, update, and delete. These are often shortened to CRUD.
Create a Record
INSERT INTO customers (customer_id, name, email, city)
VALUES (104, 'Priya Shah', '[email protected]', 'London');This inserts one new customer record into the table.
Read Records
SELECT customer_id, name, city
FROM customers
WHERE city = 'Chicago';This retrieves records that match the filter condition.
Update a Record
UPDATE customers
SET city = 'Boston'
WHERE customer_id = 102;This changes a single record identified by its primary key.
Delete a Record
DELETE FROM customers
WHERE customer_id = 104;This removes the matching record from the table.
If you want to understand how the database finds records quickly, pair this topic with What Is Database Indexing? Examples for MySQL and Postgres.
Record vs. Row vs. Tuple
In everyday conversation, people often use record and row interchangeably. In relational databases, that is usually fine. A row in a table is the practical implementation of a record.
The word tuple is more theoretical and often appears in database theory or academic writing. In many contexts, a tuple also means a row. For most working developers, analysts, and administrators, “record” and “row” are the useful terms.
Term
Common Meaning
Record
One stored unit of related data
Row
The table representation of a record
Tuple
Theoretical term often equivalent to a row
Real-World Examples
CRM system: each contact is one record with fields like contact_id, company, title, phone, and lifecycle_stage.
E-commerce platform: each order is one record with order_id, customer_id, order_date, subtotal, tax, and shipping_status.
Healthcare system: each patient visit is one record with visit_id, patient_id, clinician_id, diagnosis_code, and visit_timestamp.
In all three examples, the structure stays consistent, but the values differ from record to record.
Common Mistakes When Thinking About Records
One mistake is assuming a record is just “any line of data.” In a well-designed database, a record belongs to a defined table structure and should represent one coherent entity.
Another mistake is confusing a field with a record. A field is one attribute, such as email. The record is the complete set of field values for one row.
A third mistake is ignoring keys. Without a primary key and clear relationships, records become hard to deduplicate, join, and maintain safely.
Key Takeaways
A record in a database is one row of related data in a table. It is made of fields, constrained by data types, and usually identified by a primary key. Records become useful when they can be created, queried, updated, and related to other records through SQL and sound schema design.
If you understand what a record is, you already understand one of the most practical building blocks of database work.