Skip to main content
Announcements
Qlik Connect 2024! Seize endless possibilities! LEARN MORE
cancel
Showing results for 
Search instead for 
Did you mean: 
parahvars
Contributor
Contributor

qilk Excel Sheet vs SQL data source

I am fairly new to QLIK I used it back in 2017 to 2019 then we switched to Power BI and Spot Fire once QLIK took away free access. The company I work for now is using QLIK and instead of using a SQL data source they are using Excel spreadsheets. I haven't received my license yet but I wanted to know if there are limitations with Excel spreadsheets as a data source or if there are any tips or tricks when using Excel data?

Although I am completely against this type of data source it looks like I will have to deal with it. Their current developer says that the filters don't work correctly and may not be able to work in QLIK. I find this hard to believe and once I get access I'm hoping this is not the case. I was just curious if there are limitations with QLIK and an Excel data source when compared to programs like Power BI.

Labels (1)
3 Replies
Ray_Strother
Support
Support

Hello,

I'm not sure of limitation they are speaking of. What specific problem are they facing? Below are some help site links that may help:


1. Loading data from Microsoft Excel spreadsheets

https://help.qlik.com/en-US/sense/November2023/Subsystems/Hub/Content/Sense_Hub/DataSource/load-data...



2. Guidelines for data and fields
https://help.qlik.com/en-US/sense/November2023/Subsystems/Hub/Content/Sense_Hub/Introduction/guideli....
marcus_sommer

Excel files are an easy to use source in Qlik and there are no real limitations. The load-performance of xlsx will be lower as xls and those lower as from text-files but accessing data from a data-base isn't mandatory faster - it will depend on various parameters. Are the data-sets are rather larger with hundreds of thousands of records in the files and/or hundreds or thousands of files it may need some time - but regardless from the source this a good reason to apply an incremental approach.

Nevertheless there are a few specialities. Excel as source has no real NULL else NOTHING will be treated as EMPTY in default which has benefits as well as disadvantages and could be of course adjusted in various ways. Further complete empty records will be skipped and identically header get a continuously number added. Nearly everything what may occur is easy to handle.

Beside this you may have a more or less good data-quality but again this mustn't be more worse as by a data-base because they could also lack for mandatory fields and/or any validity checks as Excel could be manually manipulated.

steeefan
Luminary
Luminary

You could build a two-stage approach. If you have data in the Excel files that does not change, such as from previous years, you could extract that once and write to to QVD files, one for each year for instance. In your actual app, you then load from these QVDs combined with data from the Excel files directly. That should significantly increase your load performance.