Why does your SQL alias error? The execution order

You wrote the query top to bottom. The engine runs it in a completely different order. That single fact explains the most common beginner error in SQL, the WHERE vs HAVING confusion, and why ORDER BY is special.

The error everyone hits

SELECT price * quantity AS revenue
FROM sales
WHERE revenue > 1000;

-- ERROR: column 'revenue' does not exist

The alias exists in your head, not yet in the engine. WHERE runs before SELECT, so at filter time revenue has not been created.

Written order vs run order

ORDER BY is the only clause that can use your aliases: it is the only one that runs after SELECT.

The two fixes

-- 1. repeat the expression
SELECT price * quantity AS revenue
FROM sales
WHERE price * quantity > 1000;

-- 2. or name it in a CTE first
WITH t AS (
  SELECT *, price * quantity AS revenue
  FROM sales
)
SELECT * FROM t WHERE revenue > 1000;

Some engines (BigQuery, DuckDB) also offer QUALIFY and lateral column aliases, but the CTE works everywhere.

The trap: WHERE vs HAVING

WHERE filters rows before grouping. HAVING filters groups after.

-- wrong: aggregate in WHERE
SELECT city, SUM(amount)
FROM sales
WHERE SUM(amount) > 100   -- ERROR
GROUP BY city;

-- right: aggregates live in HAVING
SELECT city, SUM(amount)
FROM sales
GROUP BY city
HAVING SUM(amount) > 100;

Same root cause: WHERE runs before GROUP BY, so no aggregate exists yet when it fires. Once the execution order clicks, half of SQL's error messages start making sense.