Common Qlik Sense design mistakes: Not contextualizing data

In order to provide a richer and faster analysis experience for the user, a good tip is to find at least one comparative indicator for each main KPI. Choosing between these auxiliary information options will depend on the needs of your analysts.

In this first text about the 5 most common design mistakes for Qlik sense, we will show you with examples why it is a problem not to contextualize the data and we bring you our lessons learned that derived from this common mistake.

Error #1: Not contextualizing data

Let’s start with a short example to illustrate our point. Below we have a visualization that aims to cross sales and consumer experience indicators.

Right at the top of the page, in a prime area, we see the key indicators of the analysis in the form of KPIs. So far, so good.

 

This is a good position on the screen to show this type of data. KPIs highlight important numbers and the top left corner is where the user will look first. But upon closer inspection, we have some problems that end up impoverishing our analysis.

The first of these issues is the absence of the time period being shown. Am I seeing YTD data, last month data,  or MTD data? This is not made explicit anywhere.

Furthermore, there is a lack of information that helps me to understand whether these numbers are within the expected range or not. 

How are the indicators performing in relation to the target? Are they growing? How do they vary in relation to the previous year or month? 

There is a lack of context in order to analyze performance. Visually, we also don’t have any indication that helps understand these relationships.

The solution: find the context that makes sense for your analysts

In order to provide a richer and faster analysis experience for the user, a good tip is to find at least one comparative indicator for each main KPI. Choosing between these auxiliary information options will depend on the needs of your analysts. 

How do they track the performance of these metrics on a day-to-day basis? Are there goals? Is there a reference to the performance of the previous year? Are results monitored weekly? 

The important thing is to find some context that makes sense to whoever is going to analyze the data.

Below we see a good example of contextualized data. It is a dashboard, and a landing page that aims to show a summary of the most important information gathered from the entire dashboard.

In the header of the page we find an analysis period, as well as the possibility to choose other periods within a timeline. The option that is selected by default should be the one most used by your analysts.

In the body of the page we see that for each indicator of the current period (last 4 weeks), there is a reference to the previous period (4 previous weeks) and the variation between these two periods. 

The variation also has a color indicator and arrows that help me to quickly visualize which indicators are performing positively (green) or negatively (red).

Lessons learned:

• Make the analysis period explicit in a location that is easy to associate with the data;
• Bring context to indicators that allow understanding how they are performing;
• Use visuals like colors and arrows for faster performance reading;
• Always validate with dashboard users what context makes the most sense for the analyzes they need to do;

Data without context is nothing more than just numbers. They bring some information, sure, but do not communicate their full potential. If you want more meaningful analysis, contextualize your data. Your user will only benefit from this, and so will your business.

Note: The dashboard examples and images used in this series of posts are taken from the demos section of the Qlik website for educational purposes only.