Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions powerbi-docs/guidance/import-modeling-data-reduction.md
Original file line number Diff line number Diff line change
Expand Up @@ -58,6 +58,8 @@ In some specific instances, you can convert source text data into numeric values

In this example, we recommend that you set the column default summarization property to `Do Not Summarize`. It helps to avoid inappropriate summarization of the order number values.

Converting the text data to numeric values does not help if the column is being used in a relationship, because it will always be hash encoded.
Copy link

Copilot AI Dec 3, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The sentence structure is awkward with "does not help." Consider revising to: "Converting text data to numeric values doesn't help if the column is used in a relationship, because it's always hash encoded."

This improves readability by:

  1. Using the contraction "doesn't" (consistent with Microsoft Writing Style Guide conversational tone)
  2. Changing "the column is being used" to "the column is used" (simpler present tense)
  3. Changing "it will always be" to "it's always" (more concise and uses present tense)
Suggested change
Converting the text data to numeric values does not help if the column is being used in a relationship, because it will always be hash encoded.
Converting text data to numeric values doesn't help if the column is used in a relationship, because it's always hash encoded.

Copilot uses AI. Check for mistakes.

## Preference for custom columns

The VertiPaq storage engine stores model [calculated columns](../transform-model/desktop-calculated-columns.md) (defined in DAX) just like regular Power Query-sourced columns. However, the internal data structures are stored slightly differently, and typically achieve less efficient compression. Also, the data structures are built once all Power Query tables are loaded, which can result in extended data refresh times. It's therefore less efficient to add table columns as _calculated_ columns than Power Query _computed_ columns (defined in M).
Expand Down