5 Brutal Truths: Your Qlik Sense App Still Doesn’t Show Root Cause

If executives keep asking for “one more breakdown,” your app isn’t lacking data — it’s lacking causality.

Read time: 2.5 minutes

The meeting began with initial confidence because the Qlik Sense dashboard loaded with polished KPI graphics and responsive filter functionality. "Why is the revenue $1.2M down?" was the first question.

Everyone was clicking and filtering through Product, Region, SKU, etc. After 10 minutes of filtering and pivoting, they all had the numbers but could not provide an explanation for them. The numbers were present but the insight was absent. The app was designed for exploration and not for accountability.

5 Painful Realities About Qlik Sense Root Cause:

1. A Drilldown Is Not The Same As A Cause.
Without calculating the contribution, you are just guessing.
Solution: Create a Root Cause sheet with:
• Variance = Sum(Actual) - Sum(Plan),
• Contribution % = Variance / Sum(Total Variance),
• A Waterfall chart showing variance visually.

2. Your Load Script Is The Actual Problem.
An associative model is not the same as a clean model.
Solution: Remove synthetic keys from your load script. Use a Master Calendar. Pre-aggregate heavy fact tables into QVD layers. Use Fact tables and Dimension tables.

3. Absolute KPI’S Are Executive Theater.
Absence of Abs will not show the result of the Movement.
Solution: Standardize Variance, Variance %, Contribution % using TOTAL . Every KPI must show you at least one delta.

4. Your Sheet Does Not Have A Story.
You have selections everywhere, but have no insight.
Solution: Include dynamic text such as:
“-$1.2M driven by $(=Only(TopDriver)) (-62%).”

5. If It’s Slow You’ll Die Trying To Explore.
Delays are going to prevent you from exploring.
Solution: Use QVD staging, pre-calculate flags, reduce high-cardinality fields and monitor object calculation times.

💡Key Takeaway: 

A dashboard gives you numbers but a root cause analysis gives you ownership of how this happened. If your application can’t give an explanation in a single sentence as to why something has occurred, then you are not done developing your application yet.

👉 LIKE this if you have created a dashboard and still cannot answer the question “why?”

👉 SUBSCRIBE now for very sharp and pragmatic copies of Qlik Sense and Lee Analytics.

👉 Follow Glenda Carnate for more updates where we break down our dashboard into a decision-making tool.

👉 COMMENT with the hardest root cause question your dashboard failed to answer.

👉 SHARE this with someone who thinks drilldown equals insight.

Reply

or to participate.