Drive insight discovery with the data visualization app that anyone can use. With Qlik Sense, everyone in your organization can easily create flexible, interactive visualizations and make meaningful decisions.
Qlik Sense advantage
Most BI products let us prepare dedicated reports for those parts of business in which we are interested the most. Generally we design reports to answer question that are important for now. But what about follow-up questions? Qlik Sense as an in-memory analysis leader (by Gartner) which let us explore our data freely, learning at each step along the way and answer the question after question. And all of that with no advanced scripting.
Gartner's Magic Quadrant for Business Intelligence and Analytics Platforms - February 2016
For our case, we will use the business scenario for Textile Company which tries to find out their sales trend. Company is a wholesaler of fine clothing and has about 1200 customers in 21 countries. The company is looking for an easy to use and powerful platform that serves as the foundation for Self-Service Business discovery. We will use Qlik Sense as foundation for their reporting and analysis needs. Starting from connecting data repositories, loading data, creating associations and finally creating visualizations we will try to answer most recent question, which is: When and what buy our biggest customers?
As we can see in our client's environments - most of them use many different software systems which are provided by many different vendors. This is the best test for BI tool - try to combine all of different data sources and load the data with high level of quality. Qlik Sense is delivered with a list of providers out of the box, which we can use to load data from almost all leading types of repositories.
To simplify our scenario we will try to simulate loading Orders data from MS SQL Server and join it with additional Customer and Products details from XLS files.
Additional data from other systems (xls):
We just have selected the interesting data sources for load progress. There is one more thing that has to be prepared - Associations. Qlik Sense prepared innovative way of creating relations between data sources (tables) without any scripting.
Associations view of the data manager illustrates data using bubbles. Each bubble represents data table and size represents the amount of data in table. To make a new association (relation) between two tables simply drag one of them - you will be supported by assist which will try to find all matching objects.
Qlik Sense algorithm by checking data types and most recent data values recognizes similar columns and most likely keys for table associations. According to our source object - the best choice for [Support] will be [Products] table.
Doing so with all of tables we will get the final data model.
Data sources are ready to load. In loading process Qlik will prepare and execute all necessary scripts, prepare data model based on associations and load data into application file and RAM memory to support quick data calculations.
We are finally ready to start preparing visualizations. All data from data model is already in the application. Visualizations can be created from many types of charts (such as bar charts, pie charts, tables, gauges, treemaps), but to design an effective visualizations with clear purpose, you should thoroughly understand your data. If you know how it is organized and how its parts relate to each other, you should consider using well-known design patterns. For example: If you want to show how a measure behaves over time - you should use a line chart. If you need a summary of amounts by categories, the bar chart would be the best choice. Preparing visualizations is as simple as choosing proper dimensions and measures on different objects.
Before you start creating dashboards - try to predict what information are important for your users and who is going to make business decisions using your application. Making that research will make an easier understanding of how to prepare key performance indicators and how to design your dashboards. You also have to try avoid the most recent BI pitfalls of data visualization, which are: Color abuse, misuse of pie charts, visual clutter, poor design and the most important: bad data.
Answering business questions
The first dashboard should be always concentrated on the most important KPIs for last counting business period - for example: current month. The first view of app should deliver the most common values in which business users are interested in.
In our case - charts show the Sales amount in current month, compared to previous months and the same month in previous year. We can also see Top 5 lists of most sold products, most country export and the biggest buyers.
The next dashboard should be designed for some more analysis and let us drill down the data.
Trying to analyse what products were sold the most in Germany in the first quarter of 2012, we simply have to select: (1) Year and month, (2) Germany. All object in app will be automatically reduced to values associated with our selections.
Additionally we can check what was the sum of sales in that period, who were our best customers and in which cities we have had the biggest sales. Use your imagination to make the best selections and how to drill down the data to answer most recent questions in your company.
If we would like to share our findings in the company with other managers, we can easily prepare simple presentation inside the Qlik Sense.
Presentation is ready to export as PowerPoint file which can be easily send via email.
As you can see this simple data analysis was prepared with no scripting. Having minimum skills it will take about 30 min from a question to preparing an answer in your company. And when you start preparing Qlik apps - there is no way to miss important information hidden in your data.