Distinct Count

Count the number of unique values in a field using field.count.

How It Works

TPL has two forms of count:

Syntax	Meaning	SQL Equivalent
n or count	Row count	COUNT(*)
field.count	Distinct count of field values	COUNT(DISTINCT field)

When you bind count to a field (e.g., occupation.count), it counts the number of unique values of that field within each group. Standalone count or n counts the total number of rows.

Interactive Example

Try It

TABLE ROWS education * (n | occupation.count);

Dataset:samplesoccupation, education, gender, income

Try Variations

Query Breakdown

• n - Row count: how many rows exist for each education level
• occupation.count - Distinct count: how many unique occupations appear in each education level
• The row count will always be >= the distinct count

This is useful for understanding data cardinality, e.g.:

• How many unique users per region
• How many distinct products per category
• How many unique event types per session

Ordering by Distinct Count

You can order and limit rows by distinct count, just like any other aggregate:

-- Top 5 education levels by number of distinct occupations
TABLE ROWS education[-5@occupation.count] * occupation.count;

-- All education levels ordered by distinct occupation count (descending)
TABLE ROWS education DESC@occupation.count * occupation.count;

Distinct Count with Percentages

Combine distinct count with ACROSS to see each group's share of unique values:

TABLE ROWS education * (occupation.count ACROSS);

Related Examples

• Multiple Aggregates - Side-by-side aggregations
• Measure Binding - Bind multiple aggs to a measure
• Limit by Value - Top-N ordering