Filtering rows with WHERE

In session b1 you ran SELECT ... FROM ... and got the whole table back. That is fine for eight products, hopeless for a million orders. WHERE sits right after FROM and keeps only the rows that pass a test you write - "price under 18", "country is Canada", "plan is Pro". Everything you already know still works; you are just adding one line that does the filtering.

A WHERE clause is a test applied to every row, one at a time. If the test is true, the row comes back; if false, it is dropped. The tests are built from comparison operators you already know from a calculator: = equals, != not equal, > greater, < less, >= and <=. Text goes in single quotes; numbers do not.

SELECT name, category, price
FROM products
WHERE category = 'Coffee';

SELECT name, price
FROM products
WHERE price < 18;

One test is useful; combining them is where filtering gets sharp. AND means a row must pass both tests. OR means it needs pass either. NOT flips a test. String enough together and you can describe almost any slice of customers in one line - "Pro plan, in the USA, signed up this year".

SELECT name, city
FROM customers
WHERE plan = 'Pro' AND country = 'USA';

SELECT name, country
FROM customers
WHERE country = 'Canada' OR country = 'UK';

Four specialist operators cover the filters that = cannot. LIKE matches text patterns with wildcards. IN checks membership in a list. BETWEEN tests a range, endpoints included. IS NULL finds missing values - and it is the one people get wrong most often.

SELECT name
FROM products
WHERE name LIKE '%Espresso%';

SELECT name, country
FROM customers
WHERE country IN ('Canada','UK');

SELECT name, price
FROM products
WHERE price BETWEEN 15 AND 20;

SELECT name
FROM products
WHERE roast IS NULL;

Daybreak's marketing lead wants a segment: "our Pro customers in the USA, newest signups first." That is two filters joined by AND, plus an ORDER BY from session b1. Let us assemble it one clause at a time.

SELECT name, city, signup_date
FROM customers
WHERE plan = 'Pro' AND country = 'USA'
ORDER BY signup_date DESC;

No hand-holding now. Each editor answers one question - run it, tweak it, break it, fix it. That run-fix-run loop is the whole skill, and errors are cheap here.

SELECT name, price
FROM products
WHERE category = 'Coffee' AND price < 18;

SELECT name, category, roast
FROM products
WHERE category = 'Equipment' AND roast IS NULL;

SELECT name, signup_date
FROM customers
WHERE signup_date LIKE '2026%'
ORDER BY signup_date;

• List every Dark-roast coffee by name and price. (Hint: two filters, roast = 'Dark'.)
• Find all customers who are not in the USA. Try it two ways: with != and with NOT country = 'USA'.
• Pull products priced BETWEEN 20 AND 30, most expensive first.
• Find every product whose name contains the word "Blend" using LIKE.
• Bring one segment question about Daybreak you cannot yet answer to session b3 - odds are it needs counting or summing, which is exactly what b3 delivers.

This session teaches the filtering core of the major free SQL curricula, run on live Daybreak data instead of screenshots. Certificates and graded problem sets stay on the official sites (linked below). This page covers: