Skip to main content
SQL, despite being a powerful language for database operations, has several common anti-patterns that can lead to performance issues, security vulnerabilities, and maintainability problems. Here are the most important anti-patterns to avoid when writing SQL code.
-- Anti-pattern: Using SELECT *
SELECT * FROM customers
JOIN orders ON customers.customer_id = orders.customer_id;

-- Better approach: Specify only needed columns
SELECT 
    customers.customer_id,
    customers.first_name,
    customers.last_name,
    orders.order_id,
    orders.order_date,
    orders.total_amount
FROM customers
JOIN orders ON customers.customer_id = orders.customer_id;
Avoid using SELECT * in production code. Instead, explicitly list the columns you need. This improves performance by reducing I/O and network traffic, prevents issues when table schemas change, and makes your code more maintainable by clearly showing which columns are being used.
-- Anti-pattern: String concatenation for dynamic SQL
-- (Pseudocode mixing application code and SQL)
query = "SELECT * FROM users WHERE username = '" + username + "' AND password = '" + password + "'";

-- Better approach: Use parameterized queries
-- In a prepared statement (syntax varies by language/framework)
query = "SELECT * FROM users WHERE username = ? AND password = ?";
execute_query(query, [username, password]);

-- Or using stored procedures
CREATE PROCEDURE authenticate_user
    @username VARCHAR(50),
    @password VARCHAR(100)
AS
BEGIN
    SELECT * FROM users WHERE username = @username AND password = @password;
END;
Always use parameterized queries or prepared statements instead of concatenating strings to build SQL queries. This prevents SQL injection attacks and improves performance through query plan caching.
-- Anti-pattern: Cartesian product (implicit cross join)
SELECT customers.name, orders.order_id
FROM customers, orders
WHERE customers.customer_id = orders.customer_id;

-- Better approach: Explicit JOIN syntax
SELECT customers.name, orders.order_id
FROM customers
JOIN orders ON customers.customer_id = orders.customer_id;

-- Anti-pattern: Joining unnecessary tables
SELECT customers.name, orders.order_id
FROM customers
JOIN orders ON customers.customer_id = orders.customer_id
JOIN order_items ON orders.order_id = order_items.order_id
JOIN products ON order_items.product_id = products.product_id;
-- ^ Products table isn't needed if we're not using any columns from it

-- Better approach: Only join necessary tables
SELECT customers.name, orders.order_id
FROM customers
JOIN orders ON customers.customer_id = orders.customer_id;
Use explicit JOIN syntax instead of implicit joins in the WHERE clause. Only join tables that are necessary for your query. Unnecessary joins can dramatically decrease performance, especially with large tables.
-- Anti-pattern: Not indexing columns used in WHERE, JOIN, or ORDER BY
SELECT * FROM orders WHERE customer_id = 123;
-- If customer_id isn't indexed, this will perform a full table scan

-- Better approach: Create appropriate indexes
CREATE INDEX idx_orders_customer_id ON orders(customer_id);

-- Anti-pattern: Using functions on indexed columns
SELECT * FROM customers WHERE LOWER(email) = 'user@example.com';
-- Index on email won't be used because of the LOWER function

-- Better approach: Avoid functions on indexed columns
SELECT * FROM customers WHERE email = 'user@example.com';
-- Or create a functional index if your DBMS supports it
CREATE INDEX idx_customers_email_lower ON customers(LOWER(email));
Create appropriate indexes for columns used in WHERE clauses, JOIN conditions, and ORDER BY statements. Avoid using functions on indexed columns in WHERE conditions, as this can prevent the optimizer from using the index.
-- Anti-pattern: Using DISTINCT to hide join problems
SELECT DISTINCT customers.customer_id, customers.name
FROM customers
JOIN orders ON customers.customer_id = orders.customer_id;

-- Better approach: Fix the join to avoid duplicates
SELECT customers.customer_id, customers.name
FROM customers
JOIN (
    SELECT DISTINCT customer_id FROM orders
) o ON customers.customer_id = o.customer_id;

-- Or even better, if you just want customers who have orders:
SELECT customers.customer_id, customers.name
FROM customers
WHERE EXISTS (
    SELECT 1 FROM orders WHERE orders.customer_id = customers.customer_id
);
Avoid using DISTINCT as a quick fix for duplicate rows. It can hide join problems and hurt performance. Instead, fix the underlying issue by using the correct join type or restructuring your query.
-- Anti-pattern: Using cursors for row-by-row processing
DECLARE @customer_id INT;
DECLARE @total_spent DECIMAL(10, 2);

DECLARE customer_cursor CURSOR FOR
SELECT customer_id FROM customers;

OPEN customer_cursor;
FETCH NEXT FROM customer_cursor INTO @customer_id;

WHILE @@FETCH_STATUS = 0
BEGIN
    SELECT @total_spent = SUM(total_amount)
    FROM orders
    WHERE customer_id = @customer_id;
    
    UPDATE customers
    SET total_spent = @total_spent
    WHERE customer_id = @customer_id;
    
    FETCH NEXT FROM customer_cursor INTO @customer_id;
END

CLOSE customer_cursor;
DEALLOCATE customer_cursor;

-- Better approach: Use set-based operations
UPDATE customers
SET total_spent = (
    SELECT SUM(total_amount)
    FROM orders
    WHERE orders.customer_id = customers.customer_id
);
Avoid using cursors for row-by-row processing when set-based operations can accomplish the same task. Set-based operations are typically much faster and more efficient in SQL.
-- Anti-pattern: Using inappropriate data types
CREATE TABLE products (
    product_id VARCHAR(50),  -- Using VARCHAR for a numeric ID
    price VARCHAR(10),       -- Using VARCHAR for a numeric price
    created_at VARCHAR(20)   -- Using VARCHAR for a date/time
);

-- Better approach: Use appropriate data types
CREATE TABLE products (
    product_id INT PRIMARY KEY,
    price DECIMAL(10, 2),
    created_at TIMESTAMP
);
Use appropriate data types for your columns. Using the wrong data type can lead to unexpected behavior, poor performance, and increased storage requirements. For example, use numeric types for numbers, date/time types for dates, and VARCHAR only for variable-length strings.
-- Anti-pattern: Storing comma-separated values
CREATE TABLE users (
    user_id INT PRIMARY KEY,
    name VARCHAR(100),
    roles VARCHAR(255)  -- Stores roles like 'admin,editor,viewer'
);

-- Querying for users with a specific role is difficult
SELECT * FROM users WHERE roles LIKE '%admin%';
-- This could match 'admin', but also 'administrator' or 'badmin'

-- Better approach: Normalize the data
CREATE TABLE users (
    user_id INT PRIMARY KEY,
    name VARCHAR(100)
);

CREATE TABLE roles (
    role_id INT PRIMARY KEY,
    role_name VARCHAR(50)
);

CREATE TABLE user_roles (
    user_id INT,
    role_id INT,
    PRIMARY KEY (user_id, role_id),
    FOREIGN KEY (user_id) REFERENCES users(user_id),
    FOREIGN KEY (role_id) REFERENCES roles(role_id)
);

-- Now querying for users with a specific role is straightforward
SELECT u.user_id, u.name
FROM users u
JOIN user_roles ur ON u.user_id = ur.user_id
JOIN roles r ON ur.role_id = r.role_id
WHERE r.role_name = 'admin';
Avoid storing comma-separated values in a single column. This violates first normal form and makes it difficult to query, update, and maintain data integrity. Instead, normalize your data by creating appropriate related tables.
-- Anti-pattern: Not using transactions for multi-statement operations
UPDATE accounts SET balance = balance - 100 WHERE account_id = 123;
UPDATE accounts SET balance = balance + 100 WHERE account_id = 456;
-- If the second statement fails, the first one still goes through

-- Better approach: Use transactions
BEGIN TRANSACTION;

UPDATE accounts SET balance = balance - 100 WHERE account_id = 123;
UPDATE accounts SET balance = balance + 100 WHERE account_id = 456;

-- Check if both updates succeeded
IF @@ROWCOUNT > 0
    COMMIT TRANSACTION;
ELSE
    ROLLBACK TRANSACTION;
Use transactions for operations that require multiple statements to be executed as a single unit. This ensures data consistency and integrity, especially for operations like transfers or complex updates.
-- Anti-pattern: Using NOT IN with a subquery that might return NULL
SELECT * FROM customers
WHERE customer_id NOT IN (SELECT customer_id FROM orders WHERE total_amount > 1000);

-- Better approach: Use NOT EXISTS or LEFT JOIN / IS NULL
-- Option 1: NOT EXISTS
SELECT * FROM customers c
WHERE NOT EXISTS (
    SELECT 1 FROM orders o
    WHERE o.customer_id = c.customer_id AND o.total_amount > 1000
);

-- Option 2: LEFT JOIN / IS NULL
SELECT c.*
FROM customers c
LEFT JOIN orders o ON c.customer_id = o.customer_id AND o.total_amount > 1000
WHERE o.order_id IS NULL;
Avoid using NOT IN with a subquery that might return NULL values. In SQL, comparing anything to NULL using NOT IN returns UNKNOWN, which is treated as FALSE in a WHERE clause. Use NOT EXISTS or LEFT JOIN / IS NULL instead.
-- Anti-pattern: Using COUNT(*) to check existence
IF (SELECT COUNT(*) FROM orders WHERE customer_id = 123) > 0
    -- Customer has orders
ELSE
    -- Customer has no orders

-- Better approach: Use EXISTS
IF EXISTS (SELECT 1 FROM orders WHERE customer_id = 123)
    -- Customer has orders
ELSE
    -- Customer has no orders
Use EXISTS instead of COUNT() to check if records exist. EXISTS is more efficient because it stops scanning as soon as it finds a matching record, while COUNT() needs to scan all matching records.
-- Anti-pattern: Using OFFSET for pagination with large offsets
SELECT * FROM products
ORDER BY product_id
LIMIT 10 OFFSET 10000;

-- Better approach: Use keyset pagination
-- Assuming the last product_id from the previous page was 10000
SELECT * FROM products
WHERE product_id > 10000
ORDER BY product_id
LIMIT 10;
Avoid using OFFSET for pagination with large offset values. OFFSET requires the database to scan and discard all rows up to the offset, which becomes increasingly inefficient as the offset grows. Use keyset pagination (also known as cursor-based pagination) instead, which uses a filter on the last seen value.
-- Anti-pattern: Using OR conditions on different columns
SELECT * FROM products
WHERE category = 'Electronics' OR brand = 'Samsung';

-- Better approach: Use UNION
SELECT * FROM products WHERE category = 'Electronics'
UNION
SELECT * FROM products WHERE brand = 'Samsung';
Consider using UNION instead of OR conditions on different columns, especially if you have indexes on those columns. The optimizer can use different indexes for each part of the UNION, potentially improving performance.
-- Anti-pattern: Repeating complex subqueries
SELECT
    customer_id,
    (
        SELECT COUNT(*) FROM orders
        WHERE orders.customer_id = customers.customer_id
    ) AS order_count,
    (
        SELECT SUM(total_amount) FROM orders
        WHERE orders.customer_id = customers.customer_id
    ) AS total_spent
FROM customers;

-- Better approach: Use Common Table Expressions (CTEs)
WITH customer_orders AS (
    SELECT
        customer_id,
        COUNT(*) AS order_count,
        SUM(total_amount) AS total_spent
    FROM orders
    GROUP BY customer_id
)
SELECT
    c.customer_id,
    co.order_count,
    co.total_spent
FROM customers c
LEFT JOIN customer_orders co ON c.customer_id = co.customer_id;
Use Common Table Expressions (CTEs) to simplify complex queries, improve readability, and avoid repeating subqueries. CTEs can also be recursive, which is useful for hierarchical data.
-- Anti-pattern: Implicit conversions
SELECT * FROM orders WHERE order_id = '1000';
-- order_id is an INT, but we're comparing it to a string

-- Better approach: Use explicit conversions or correct types
SELECT * FROM orders WHERE order_id = 1000;
-- Or if you need to convert
SELECT * FROM orders WHERE order_id = CAST('1000' AS INT);
Avoid implicit conversions between different data types. They can prevent the optimizer from using indexes and lead to unexpected results. Use the correct data type in your queries or explicit conversions when necessary.
-- Anti-pattern: Using scalar UDFs in WHERE clauses
CREATE FUNCTION dbo.GetFullName(@FirstName VARCHAR(50), @LastName VARCHAR(50))
RETURNS VARCHAR(101)
AS
BEGIN
    RETURN @FirstName + ' ' + @LastName;
END;

SELECT * FROM employees
WHERE dbo.GetFullName(first_name, last_name) = 'John Doe';

-- Better approach: Inline the function logic
SELECT * FROM employees
WHERE first_name + ' ' + last_name = 'John Doe';
Avoid using scalar user-defined functions in WHERE clauses. They can prevent the optimizer from using indexes and are executed once per row, which can severely impact performance. Inline the function logic directly in your query when possible.
-- Anti-pattern: Not using schema names
SELECT * FROM customers;

-- Better approach: Always use schema names
SELECT * FROM dbo.customers;
Always use schema names when referencing database objects. This improves query plan caching, prevents ambiguity if multiple schemas have objects with the same name, and makes your code more maintainable.
-- Anti-pattern: Using table variables for large result sets
DECLARE @LargeResults TABLE (
    ID INT,
    Name VARCHAR(100),
    Value DECIMAL(10, 2)
);

INSERT INTO @LargeResults
SELECT ID, Name, Value FROM SomeLargeTable;

-- Better approach: Use temporary tables for large result sets
CREATE TABLE #LargeResults (
    ID INT,
    Name VARCHAR(100),
    Value DECIMAL(10, 2)
);

CREATE INDEX IX_LargeResults_ID ON #LargeResults(ID);

INSERT INTO #LargeResults
SELECT ID, Name, Value FROM SomeLargeTable;
Use temporary tables (#temp) instead of table variables (@table) for large result sets or when you need indexes. Table variables store statistics only when they’re created and don’t support indexes (in most database systems), which can lead to poor query plans for large datasets.
-- Anti-pattern: Using DELETE without WHERE
DELETE FROM orders;

-- Better approach: Always include a WHERE clause
DELETE FROM orders WHERE order_date < '2020-01-01';

-- If you really want to delete all rows, be explicit
TRUNCATE TABLE orders; -- Faster than DELETE without WHERE
Always include a WHERE clause with DELETE statements unless you explicitly want to delete all rows. If you do want to delete all rows, use TRUNCATE TABLE, which is faster and uses fewer system resources.
-- Anti-pattern: Using separate INSERT/UPDATE for upserts
IF EXISTS (SELECT 1 FROM customers WHERE customer_id = 123)
    UPDATE customers
    SET name = 'John Doe', email = 'john@example.com'
    WHERE customer_id = 123;
ELSE
    INSERT INTO customers (customer_id, name, email)
    VALUES (123, 'John Doe', 'john@example.com');

-- Better approach: Use MERGE for upserts
MERGE INTO customers AS target
USING (SELECT 123 AS customer_id, 'John Doe' AS name, 'john@example.com' AS email) AS source
ON target.customer_id = source.customer_id
WHEN MATCHED THEN
    UPDATE SET name = source.name, email = source.email
WHEN NOT MATCHED THEN
    INSERT (customer_id, name, email)
    VALUES (source.customer_id, source.name, source.email);
Use the MERGE statement for upsert operations (insert if not exists, update if exists) instead of separate INSERT and UPDATE statements. MERGE is more efficient and atomic, reducing the risk of race conditions.
-- Anti-pattern: Using database-specific features without necessity
-- SQL Server specific:
SELECT TOP 10 * FROM products ORDER BY price DESC;

-- Better approach: Use standard SQL when possible
-- Standard SQL (works in most databases):
SELECT * FROM products ORDER BY price DESC LIMIT 10;
Use standard SQL features when possible to make your code more portable across different database systems. If you do need to use database-specific features, document them clearly and consider abstracting them in a way that makes them easier to change if needed.
-- Anti-pattern: Not using appropriate constraints
CREATE TABLE orders (
    order_id INT,
    customer_id INT,
    order_date DATE,
    total_amount DECIMAL(10, 2)
);

-- Better approach: Use appropriate constraints
CREATE TABLE orders (
    order_id INT PRIMARY KEY,
    customer_id INT NOT NULL,
    order_date DATE NOT NULL,
    total_amount DECIMAL(10, 2) NOT NULL CHECK (total_amount >= 0),
    FOREIGN KEY (customer_id) REFERENCES customers(customer_id)
);
Use appropriate constraints (PRIMARY KEY, FOREIGN KEY, NOT NULL, UNIQUE, CHECK) to enforce data integrity at the database level. This prevents invalid data from being inserted and makes your database more reliable.
⌘I