COUNT(*), COUNT(column), and COUNT DISTINCT
The three forms answer row count, non-NULL value count, and unique-value count.
Problem and reasoning
COUNT(*) counts result rows, COUNT(column) ignores rows where that column is NULL, and COUNT(DISTINCT column) also removes duplicate non-NULL values. Choosing the wrong form changes the metric, not just its syntax.
On an orders table, COUNT(*) returns orders, while COUNT(DISTINCT customer_id) is closer to customers. After JOIN, first check whether the relationship multiplied rows.
Use business-specific aliases such as order_count, known_phone_count, or customer_count instead of naming every aggregate total.
Examples
Sample table
customers| nickname |
|---|
| NULL |
| Bobby |
| NULL |
Before / Not recommended
SELECT COUNT(nickname) AS nickname_count FROM customers;Before result
| nickname_count |
|---|
| 1 |
After / Recommended
SELECT COUNT(*) AS total_rows, COUNT(nickname) AS known_nicknames, COUNT(DISTINCT city) AS cities FROM customers;After result
| total_rows | known_nicknames | cities |
|---|---|---|
| 3 | 1 | 3 |
Why it is better：Showing total rows, non-NULL nicknames, and distinct cities together prevents three business metrics from being confused as one COUNT.
Was this explanation clear?
Thanks. This feedback will inform the next content review.