Qlik Said ‘Trust Me’... Then $ Turned Into 1 (And No One Noticed)

Read time: 2.5 minutes

Ways To Stop Putting Your Faith In Broken Qlik Applications

1. Your associations create distorted realities.

• The combination of improperly joined records can lead people to believe things are connected when they aren’t.

• Some problems with keys, if not fixed through the data model, cause even more problems elsewhere.

Solution:

• Ensure all associations are correct before building any visuals to display your data in the Qlik applications.

• Use consistent and valid keys across all tables.

2. Currency and Data Types Are Silent Killers

• If your record types don’t match ($ vs number, for example), this will break the aggregation.

• Problems may not appear until they are reviewed because no errors are present.

Solution:

• Before loading data into your Qlik application, standardize on your data types.

• Ensure all measures are validated across multiple displays or Qlik applications.

3. Synthetic Keys Destroy Trust Quickly

• Auto-generated Synthetic Keys can create links between unrelated records.

• Different selections will yield different results.

Solution:

• Completely avoid using synthetic keys.

• Create real, explicit relationships.

4. More Data Will Not Equal More Insight

• Too many fields can confuse users, resulting in misinterpreted data relationships.

Solution:

• Intentional Curation of fields for user insight.

• Provide clear instructions using dimensions and measures.

5. Visual Display May Not Equal Data Accuracy

• Visually appealing charts can mislead or hide logic errors.

• Users believe what they see.

Solution:

• Regularly validate the output from Qlik Applications to the source data.

• Integrate validation checks within the development process.

💡Key Takeaway: 

AIf your data model is incorrect, you cannot rely on Qlik to alert you to the problem… it will represent the information with absolute certainty, but it will be incorrect.