> ## Documentation Index
> Fetch the complete documentation index at: https://parabola.io/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Average by group

The **Average by group** step calculates the average of all values in one or more columns. This step is similar to the AVERAGEIF function in Excel.

## Input/output

Our input data has 11 rows of order information showing how much of a particular food item was sold to a given customer.

<Frame>
  <img src="https://mintcdn.com/parabola-7119dfb0/55xSH1bSQs7FexjG/images/transforms/image-32.png?fit=max&auto=format&n=55xSH1bSQs7FexjG&q=85&s=bf4b7a719067bc356f4a42dbbdba56b6" alt="Input data showing 11 rows of order information with food items and quantities sold per customer" width="705" height="516" data-path="images/transforms/image-32.png" />
</Frame>

After we run our dataset through the **Average by group** step, we find the average number of each food item sold per order.

<Frame>
  <img src="https://mintcdn.com/parabola-7119dfb0/55xSH1bSQs7FexjG/images/transforms/image-33.png?fit=max&auto=format&n=55xSH1bSQs7FexjG&q=85&s=037d8277bdd4434d91a914a0e7539235" alt="Output data showing the average quantity sold per food item after grouping" width="700" height="460" data-path="images/transforms/image-33.png" />
</Frame>

## Default settings

When a data source is first connected, this step will auto-select a numeric column (like 'Quantity Sold') and a data-type column to take a first pass at averaging by group.

<Frame>
  <img src="https://mintcdn.com/parabola-7119dfb0/55xSH1bSQs7FexjG/images/transforms/image-34.png?fit=max&auto=format&n=55xSH1bSQs7FexjG&q=85&s=5ab78f11547ea7ea1bbf006e2a189401" alt="Default settings panel auto-selecting a numeric column and a grouping column" width="2180" height="894" data-path="images/transforms/image-34.png" />
</Frame>

## Custom settings

You may choose any column to average, and group the average according to any one column, or multiple columns, in your data set.

<Frame>
  <img src="https://mintcdn.com/parabola-7119dfb0/55xSH1bSQs7FexjG/images/transforms/image-35.png?fit=max&auto=format&n=55xSH1bSQs7FexjG&q=85&s=b0ba6857e1a04348004cf5b0bfed0232" alt="Custom settings panel selecting a column to average and one or more columns to group by" width="1030" height="448" data-path="images/transforms/image-35.png" />
</Frame>

You can optionally apply this to only the selected columns or apply to 'all columns except'. We'd recommend using the option that requires fewer column selections to save yourself time.

<Frame>
  <img src="https://mintcdn.com/parabola-7119dfb0/55xSH1bSQs7FexjG/images/transforms/image-36.png?fit=max&auto=format&n=55xSH1bSQs7FexjG&q=85&s=a9c837f72e523ea7b7fb20598c97b26c" alt="Settings panel showing the 'apply to selected columns' versus 'all columns except' toggle" width="1029" height="389" data-path="images/transforms/image-36.png" />
</Frame>

## Related steps

* [Sum by group](/product/transform/sum-by-group)
* [Count by group](/product/transform/count-by-group)
* [Find maximum by group](/product/transform/find-maximum-by-group)
* [Find minimum by group](/product/transform/find-minimum-by-group)
* [Merge duplicate rows](/product/transform/merge-duplicate-rows)