Do not input private or sensitive data. View Qlik Privacy & Cookie Policy.
Skip to main content

Announcements
Now accepting applications for the Qlik Luminary and Partner Ambassador Programs: Apply by July 6!
cancel
Showing results for 
Search instead for 
Did you mean: 
Pri_07_03
Partner - Contributor III
Partner - Contributor III

Using Multiple Excel Sheets with Different Granularity in Qlik Sense

Can Qlik Sense use multiple sheets from the same Excel file as data sources for different visualizations (for example, a chart from one sheet and a table from another) within the same application?

What is the possible way if we have data with different granularities:

Monthly detail data in one sheet
Aggregated Period data (H1, H2, Annual, Recent3) in another sheet

Any recommendations or examples would be appreciated.

Labels (4)
3 Replies
marcus_sommer
MVP
MVP

Just concatenate them into a single table - see: Fact Table with Mixed Granularity - Qlik Community - 1468238

Pri_07_03
Partner - Contributor III
Partner - Contributor III
Author

As per the shared logic and requirements, I created the fact table accordingly. However, the table and chart are based on different source data and calculation logic.

The H1 chart is created from the Period/Snapshot sheet. The H1 judgment is calculated using only the September snapshot data (9th month). Therefore, H1 contains only the judgments that exist in the September snapshot, such as 〇 (Maru) and △ (Triangle).

The detail table is created from the Monthly sheet. For H1, the table displays records from April to September based on the H1 flag. As a result, the table can contain judgments such as 〇 (Maru), △ (Triangle), and ✕ (Cross) that occurred in any month within the H1 period.

When a user selects ✕ (Cross) in the table, Qlik Sense applies that selection across all visualizations. However, the H1 chart is based only on the September snapshot data, where no ✕ (Cross) judgment exists. Therefore, no matching records remain for the H1 chart, causing the H1 bar to disappear.

marcus_sommer
MVP
MVP

By adding an appropriate source-information within each of the loads, like:

'detail' as Source

as well as:

'aggregated' as Source

the parts could be differentiated in dimensions, selections and conditions and/or states.

For the majority of views it would be sufficient to select the wanted Source part or adding it within a set analysis, like: sum({< Source = {'detail'}>} Value).