Last week, I gave the business case for using an exponential distribution to predict a customer’s purchase frequency and detect at-risk and lost customers, Sales Analytics in Qlik: From the Basics to Statistical Modeling. In this week’s post, we go over the details of calculating and visualizing the exponential distribution in QlikView using the following chart. I also add some final thoughts on how to use the information it tells us in a real-world sales dashboard.
The most common Qlik application involves sales data analysis. Period.
Well, I don’t have enough information to back that up, and since data analysis is my life, I can’t make unsupported claims without some major nervous facial twitching (or so my wife says). However, I would bet based on personal experience that it is one of the most common applications, and moreover, I’d go as far to say that it is often the first analytical application that businesses develop after buying Qlik.
One the most obvious reasons that this could be true is that sales data is huge, low-hanging fruit. Sales is what drives most businesses and the data trail it leaves is usually the most readily available data to analyze. When a company purchases Qlik, it is often after continuous investments in an ERP (Enterprise Resource Planning) system, a CRM (Customer Relationship Management) systems, a customer portal and/or numerous Excel reports – all of which make sales data ripe for harvest.
Many of you are probably familiar with the following sales metrics.
|Gross Sales Revenue||Sales before discounts measured in monetary units|
|Net Sales Revenue||Sales after discounts measured in monetary units|
|Sales Volume||Sales measured in non-monetary units such as an individual item, boxes, pallets, kilograms, tons, etc.|
|Unit Sales||Also referred to as Average Price, it can be defined as Net Sales Revenue divided by Sales Volume|
|Hits||Number of sales transactions or invoices|
|Gross Profit||Net Sales Revenue minus the Cost of Goods Sold (COGS)|
|Gross Profit Margin||Gross Profit divided by Net Sales Revenue|
If you are familiar with these metrics, you’ll also be well acquainted with the series of dimensions that often slice and dice them. Catalogs that describe customers, sales representatives, products, dates, branches, stores, promotion codes, and channels answer the questions of who, what, when, where, why, and how that surround the sales numbers. You’ll also recognize the importance of using some type of reference data that comes in the form of a budget, a forecast, or at the very least, historical data.
Almost immediately after businesses start to take advantage of this basic, yet powerful, sales data analysis, they start to adjust their business questions. We can answer some new questions by adding a new metric, a new dimension, or a new visualization, but some new questions involve more advanced analysis techniques. One of the most popular questions that I encounter and that requires a more sophisticated approach is the evaluation of customer retention and the detection of customers that the business is endanger of losing. For here on, we’ll refer to this type of analysis as customer churn.
As Qlik Sense comes of age, I anxiously wait for the day when we will talk about the good ol’ QlikView days when we used to make map charts using a background image and a scatterplot chart. In the meantime, I’ve been finishing up one last, great QlikView adventure with Mastering QlikView Data Visualization and seeing how far ol’ QlikView can still go. Although, we have not seen a major update to its native visualizations in 5 years, I’ve been pleasantly surprised that there are still a few new tricks to be learned and boundaries to be pushed.
Stephen Redmond was the first to make a cookbook on tips and tricks, and he inspired us to look for ways to squeeze the most out of QlikView. More recently, I was inspired by a QlikFix blog post Barry Harmsen wrote on macros to create a my own macro that generates design layout grids and I called it the QlikView Grid System Tool. In turn, Barry was inspired to create a improved GridMaker. (Yes, he even improved upon the name.)
That, and to add insult to injury, he has a uncanny knack for making the most hilarious memes. I’ll leave the meme business to my fellow consultant Qlik Freak Julian Villafuerte, but as far as QlikView is concerned, I thought I’d return the favor and continue the chain of inspiration.
While writing my book, I was recently inspired by an excellent QlikFix blog post written by Frédérique Verhagen about creating bar chart target lines in QlikView. This post was the piece of the puzzle I was missing to finally create a single object, native bullet graph in QlikView. If you convert Frédérique’s bar chart into a combo chart with stacked bars, an error bar and a stock chart expression then you have a native bullet graph that is as robust as any other normal QlikView chart.
By the way, you can learn the trick on how to add a stock chart expression in Stephen Redmond’s cookbook. I’ve left the rest of the details in my book because otherwise Packt will blame me for plagiarizing. No kidding, they’ve already done it once. I bet some of you can figure it out by yourselves, so I’ll also keep the rest of the tricks for the book.
And of course, the bullet graph isn’t perfect. It would be impeccable if only we could change the line width of the stock expression. If you like the idea then vote to add this feature in this Qlik Community idea. Though, in all honesty, what I’m really hoping to do is to inspire one of you to take it that little bit further.
Recently, I’ve been reviewing the financial analysis part of my new book and I’m reminded how hard it is to create data visualization for an area obsessed with calculating every amount to the exact cent. When we develop our visual analysis it is so easy to get trapped in this labyrinth of detailed tables and numbers. We hit our heads against the wall and try to imagine one omniscient visualization that captures every detail and we only succeed in creating something more impenetrable than the table we were trying to replace. So, let’s step back and think about what we need to do to create a successful visual analysis for an accounting department. Continue reading “Data Visualization for Accounting and Unicorns”
Qlik Sense and QlikView are two data discovery tools that make it easy to go from raw data to data visualization. This in contrast to Tableau that is limited to combining data tables with joins. If your data source is, for example, an OLTP system that contains a long list of table with complex relationships, you will probably have to invest time to transform and model the data in another tool before you can use Tableau for data discovery.
The couple times I tried to implement Tableau for a company, I had to work all night in SQL Server Express to create a data model that made Tableau easy to use. It was this experience that made me realize that data discovery does not depend on great data visualization alone. It also depends on being able to easily extract, transform, and model data.
When we work with data visualization in Qlik Sense, we work within a grid system. Even though Qlik Sense gets some slack because it does not allow you the freedom to escape the grid, using a grid system a good practice to follow. We should even apply the same practice when we design QlikView applications. Continue reading “QlikView Grid System Tool”
Lately, consultancies with little QlikView experience have asked me to review the feasibility of using QlikView for a variety of projects. It was obvious after only a quick glance of the projects’ goals that they did not take into consideration the strengths and necessities of QlikView. I had come to believe people understood the concept of data discovery and that we were past the idea that QlikView was a just a quick reporting tool, but I was mistaken.
Many still believe QlikView only stands for fast implementation time, ease of use and a visual display. They try to adapt what they understand of BI to what they’ve heard about QlikView. Of course, you can’t blame them because we humans naturally interpret new information based on past experiences.
So, I’ve decided to write a series of blog posts that explain the strengths of QlikView so that we can understand how to use it effectively in our organizations. I will then conclude the series by detailing the reasons why QlikView projects sometimes go awry. We’ll add an extra part in each post about how Qlik Sense may or may not change how we use QlikView.
First, let’s explain the concept of data discovery and how we should go about implementing data discovery projects. Continue reading “Data Discovery in QlikView – Part 1 – Agile Implementation”