Edilitics | Data to Decisions

Aggregations: Complete Reference for Edilitics Visualize

Complete aggregation reference for Edilitics Visualize: numerical, categorical, and datetime options by field type, defaults, available formatters, and what belongs in Transform instead.

Every field assigned to Rows or Columns carries an aggregation. Edilitics sets a default based on field type, and you can change it at any time. This page covers what the defaults are, what options are available per field type, and what cannot be done in the viz layer at all.


How Aggregations Work

Each field on the Columns or Rows panel shows an aggregation label in parentheses. Click that label to open a dropdown and change it. The chart re-renders immediately.

The options available depend on two things: the field type (numerical, categorical, or datetime) and whether the field is in Columns or Rows.

Columns hold grouping dimensions: the categories or time periods shown on the chart axis. The default aggregation for Columns is Group By, which groups by each distinct value of the field.

Rows hold measures: what is computed for each group. At least one Row field is required to generate a chart.

All aggregations run server-side. Edilitics translates each selection into the appropriate query expression for your connected database before executing. No values are computed in the browser and no raw rows leave your database.


Aggregations by Field Type

Numerical fields carry a 123 badge. When added to Rows, the default aggregation is Sum.

AggregationLabel in UINotes
SumSumTotal of all non-null values.
MeanMeanArithmetic average. Nulls excluded.
MedianMedianMiddle value. Nulls excluded.
MinimumMinSmallest non-null value.
MaximumMaxLargest non-null value.
CountCountCount of non-null rows.
Count DistinctCount DistCount of unique non-null values.
Sample standard deviationStd DevStandard deviation using N-1 denominator.
Population standard deviationStd Dev (Pop)Standard deviation using N denominator.
Sample varianceVarVariance using N-1 denominator.
Population varianceVar (Pop)Variance using N denominator.
PercentilePercentileSubmenu: 5th, 10th, 25th, 50th, 75th, 90th, 95th.

When added to Columns, the default aggregation is Group By. The available options in Columns are a reduced set: Sum, Mean, Median, Min, Max, Count, Count Dist, Std Dev, Std Dev (Pop), and Group By. Percentile and variance options are not available in Columns.

You can add the same numerical field to Rows more than once, as long as each instance uses a different aggregation. For example, revenue as Sum and revenue as Mean are two valid Rows entries.

Categorical fields carry an Abc badge. When added to Columns, the default aggregation is Group By, which places each distinct value on the chart axis.

When added to Rows, the default aggregation is Count. The available options in Rows are:

AggregationLabel in UINotes
CountCountCount of non-null rows in each group.
Count DistinctCount DistCount of unique non-null values in each group.

You can add the same categorical field to Rows twice: once as Count and once as Count Dist. No other combinations are available for categorical fields in Rows.

Datetime fields carry a calendar badge. When added to Columns, the default aggregation is Group By with Year formatter, which groups the chart by calendar year.

When added to Rows, the default aggregation is Count with Year formatter.

All three aggregations (Group By, Count, Count Dist) are available in both Columns and Rows. Each aggregation requires a formatter that specifies the time unit to extract from the raw timestamp.

Available formatters:

FormatterLabel in UIExample output
YearYear2024
QuarterQuarterQ1
MonthMonthJan
Week numberWeek Number12
Day of weekWeek DayMon
Full dateDate2024-03-15
Day of monthDay15
TimeTime14:30:00
HourHour14
MinuteMinute30
SecondSecond45
Raw valueValues(no truncation)

You can add the same datetime field to Rows more than once. Each instance must use a different aggregation and formatter combination. For example, order_date as Count with Year and order_date as Count with Month are two valid Rows entries. order_date as Count with Year added twice is not allowed.


What Belongs in Transform, Not Here

In most BI tools, anyone can define a metric anywhere. Over time, "revenue" means something different in every chart. Edilitics prevents this by design: metrics are defined once in Transform, versioned, and reused. The viz layer applies them. It does not create them.

The aggregation panel only applies fixed aggregation functions to existing columns at query time. It does not support:

  • Custom expressions or formulas (e.g., revenue / units_sold)
  • Derived columns that persist across charts
  • Level-of-detail calculations
  • Window functions or running totals

If the metric you need does not exist as a raw column, define it in Transform first. Transform outputs a new table with the derived column, which you connect as an integration and visualize like any other field. If you want AI to suggest which charts to build from your table, see Auto-Generate Charts.


FAQs


Need help? Email support@edilitics.com with your workspace, job ID, and context. We reply within one business day.

Last updated on

On this page