Trend analysis plays a vital role in understanding context and relevance of business data. Hence, visualizing Time Series has to be quick, easy, and intuitive. This is particularly true, with tools such as SAP Lumira which enable the business user to self-service his or her analytics. In this situation, the right kind of visualization and easy customization play critical roles in the ability to make quick decisions.

This blog is based on a presentation I gave at  SAP Analytics Infoday 2015, where we had built a special type of chart using Lumira SDK to analyse and visualize Time Series data in SAP Lumira. This chart type is one of a kind for SAP Lumira. It is a smart combination of categorical analysis which drills down to a Time Series analysis within a single component.

The custom chart consists of a bar chart for categorical comparison, which acts as  filter for a calendar heatmap which then facilitates the day-after-day analysis. This calendar heatmap can accommodate the entire year within a scroll-less canvas. In this blog, I will explain the chart properties while providing a glimpse into the SAP Analytics Infoday 2015 session  conducted cooperatively by SAP and Visual BI Solutions.

As an example, I have day-wise sales data for different products of a store for three years and I would like to analyze the sales patterns. The following visualizations make an attempt in accommodating all of this data into a single viz.

What it was before

SAP Lumira visualization

SAP Lumira visualizations

Looks murky doesn’t it?  It looks that way because there are so many data points to be captured. But not so anymore! Because we now have our BewilderBeast, which can accommodate all of this data in a slick manner. Take a look at it below:

What it is now

Visual BI BewilderBeast Extension

Here, you can see that I have captured the entire chart canvas. There are three core elements to this chart:

1. Bar chart as a Filter

Bar chart, acting as a filter, shows the total sales (measures) for each city (dimensions). This gives the total summary of the sales, based on the length of each bar in the bar chart. This is a clear way to see which city is performing well or not without having to  create two separate charts. From this we can see that city Kent does not have good overall sales. The chart also provides a “tooltip” for additional information (as shown below).

Bar chart as a Filter

2. Calendar as a Heatmap

On selecting the city from the bar chart, the calendar heatmap appears on the right, showing the sales of each day for that city. Here, I am selecting Kent to figure out what happened to sales performance in that city.

Calendar as a Heatmap

The chart splits the data as Year/Month/ Day and shows the sales as heat map in calendar. The more intense the colour, the more sales there were on that particular date.

Want to find out which day of the month had the highest sales? Not a problem.  Simply hover on top of the chart and the tool tip shows which day it is. From a design perspective, the colour of the chart can be changed by using SAP Lumira’s colour palette.

3. Changing Calendar Year

The year in the calendar heatmap can be changed by navigating using the arrows highlighted below. In this case, the chart lists data for the last two years. Here I am analysing the sales for the year 2014.

Visual BI Changing Calendar Year

Further Use Case

This chart can visualize not only sales performance; if you want to find out on what days the stock prices were high, you can use this to visualize the stock and get daily insights at a glance.

Consider the scenario where the management of a retail chain prefers to monitor daily sales during the time between Thanksgiving and New Year, or when the Head of marketing prefers to closely track the status of critical campaigns on a daily basis. This chart comes to aid in visualizing the right information on a daily basis. SPLY (Same Period Last Year) analysis can also be easily done using this chart.

What will be next

The following are some of the features that can be added to enhance this chart.

  • Support for two measures in bar chart filter (length and color as measure)
  • Rolled up aggregates for Month, Quarter and Year hierarchies
  • Drill down to Hours, Minutes and Seconds for analyzing live data (useful for analyzing stocks)
  • Support for running calculations in SAP Lumira

Great time at Infoday 2015

It was a great fun filled and learning time at SAP Infoday 2015 with people from SAP and Visual BI attending it. I hope to present again next time with something new. Here are some of the pics from the Infoday.

Visual BI-SAP Infoday 2015

To report this post you need to login first.

3 Comments

You must be Logged on to comment or reply to a post.

  1. Meng Wang

    Hi Shankar,

    To combine calendar view with Heatmap is really cool to get an overview of the whole year situation.

    Will you share the source code?

    Best Regards.

    Meng

    (0) 

Leave a Reply