Fuzzy Matching: LIKE vs. Equals

The Vague Request

Customer support pinged me. *"We're looking for a specific item. The customer says it had 'Ultra' in the name, but they can't remember the full title. Can you find it?"*

If I used `WHERE product_name = 'Ultra'`, I would get zero results. Most products aren't named *just* 'Ultra'; they are 'Ultra Pro Max' or 'The Ultra Desk'. I needed a "Fuzzy" search.

The Quest: The Wildcard Search

While `=` is for exact matches, the `LIKE` operator is for patterns. It works with a special symbol: **%** (The Wildcard).

`%` stands for "Zero or more characters of any kind."

The Implementation: The Pattern Match

I used `LIKE` with percentages on both sides to find 'Ultra' anywhere in the name.

-- Finding any product with 'Ultra' anywhere in its title

SELECT product_name

FROM products

WHERE product_name LIKE '%Ultra%';

Pattern variations

`'A%'`: Starts with A.

`'%A'`: Ends with A.

`'%Ultra%'`: Contains Ultra anywhere.

`'_ltra'`: (The underscore) matches exactly one character (e.g., 'Ultra' but not 'Uultra').

The "Oops" Moment

I once forgot that `LIKE` is often **Case-Sensitive** depending on the database. I searched for `'%ultra%'` and found nothing, because all the titles were capitalized: `'Ultra'`.

**Pro Tip**: Many databases offer `ILIKE` (In-sensitive LIKE) which ignores capitalization. If yours doesn't, use `LOWER(name) LIKE '%ultra%'`.

The Victory

I found 4 products with 'Ultra' in the name. Support identified the right one, and the customer was happy. I realized that a good analyst isn't a robot looking for exact bits—they are a detective looking for traces.

Your Task for Today

Use `LIKE` to find all customers whose names start with the letter 'S'. Then, try to find all products that have the word 'Blue' in their description.

*Day 10: Membership Logic—IN vs. EXISTS.*