headline »

sgf_01

Seize the Day! Submit an #SASGF Abstract by Oct 3

2016-09-21 – 6:32 AM

Are you considering submitting an abstract for SAS Global Forum 2017? Maybe you have already read the posts about how it helps your career and the excitement of sharing ideas. What is holding you back? Maybe you don’t have an idea, maybe you fear presentations, or maybe you just are unsure of the process.
I don’t have a idea
I bet this is the number one concern …

Read the full story »
BI Tools

Tips and tricks for building information maps, OLAP cubes, reports, and dashboards

BI Admin

Learn your way around a SAS BI installation.

Visual Analytics

Learn your way around the SAS Visual Analytics tool

Coding & Data

Extract, transform, and load your data into the SAS BI toolset

Stored Processes

Create and design stored processes like a rock star

Data Visualization »

Planning Your SAS Visual Analytics Dashboard

2016-06-06 – 6:41 AM

A dashboard starts with planning. In my last post, we talked about gauges and when you want to use them. In this post, we will apply the gauges to a dashboard. Gauges cannot live by themselves – they need supporting information. It must be clear how the information supports the gauges. I’ll use the same dashboard but with some different gauges. You can determine what worked the best and looked the best. Are those mutually exclusive?

Planning Your Dashboard

This is where you should always start – planning. Each of the gauges should have supporting information. This dashboard is for a Customer Support organization. There were 3 key performance indicators (KPIs) the management team established. The management team agreed if the organization could keep their eye on these KPIs, it would help maintain focus on what the organization wanted to accomplish.

KPI Supporting Info
Resolve 90% of tickets within 30 days The organization has to support service level agreements (SLA). This key SLA is 30 days to resolve a customer issue. Resolving a ticket depends on several things, such as properly trained and available staff and overall arrival rate. The organization has a trained staff, so they want to focus on the arrival rate.

To support the KPI, a dual bar-line chart that shows the Arrival Rate with the On-time percentage is used.

Resolve 25% of tickets during first contact This is a technical customer support organization. Many tickets require follow-up, such as logging into the customer system. Analysis discovered 30% of tickets could have been resolved within one phone conversation by a first-level agent. The management wants to understand severity level and how often tickets are sent to second level support.

To support the KPI, two bar charts are used: Active Incidents by Status and Ticket Severity.  There are other possibilities for this KPI as well.

Maintain customer satisfaction score of 4.5 A key part of a service organization is keeping the customers satisfied. The organization knows there is a relationship between count of tickets and satisfaction. Obviously the more tickets a customer opens the more buggy the product appears,so they will be unsatisfied.

To support this KPI, the customer rating is ranked two different ways: bottom 5 based on rating and top 5 customers based on open incidents.

Dashboard Layouts with Supporting Information

Now that I have my supporting objects, I need to determine a good layout. When the gauges are across the top, it was harder to understand how items were related. I decided to place the gauges along the left side and the supporting objects in rows. You can see how I planned the flows for the information somewhat based on role or interest. My thoughts are that different people in the organization want more information but in a different view. For instance, the product manager wants to see the product performance. The tech support manager wants to know about the agent performance.

Here’s the flow I used for the dashboard to reports.

dashboard layout

Trying out the Gauges

Here are my tries using each of the gauges.

Example 1: Using a Single Cell Table

This dashboard just uses a single cell table. There was no way to change the border on the table so it seems awkward. I tried using an icon but I could not control the spacing. [You can see the attempt a few images down.]

Click to enlarge image

Click to enlarge image

Example 2: Using a Dial

This dashboard uses a dial. It was hard to find a gauge that really worked or looked nice. The smaller image is more difficult so make sure that you view it enlarged.

dashboard

Click to view a larger image

Example 3: Using a Thermometer

I like the thermometer the best. It seems the easiest to read and the single color makes it more clear at a simple glance. I like this better than the number because you can also see how close/far you are from the goal.

Click to view the larger image

Click to view the larger image

Dealing with the Awkward Layout

I just didn’t like this dashboard above. The layout worried me. It wasn’t clear is that KPI on the left was related to the information on the right. If it wasn’t clear to me, it would not be clear to others. Then I thought I can use color and containers to assist the user with the association.

My first try was putting color behind the gauges. This seemed to highlight the dissociation of the objects more. It must be clear how the objects were related.

gauge_17

Use Color in the Horizontal Container

I did not need the gauges in a colored container. I needed a colored horizontal container. Here’s an example with a light blue background. I used the modern style for the dial. I like this dashboard and don’t find the gauges distracting. They do draw attention, which is their purpose. [Enlarge the image for a life-size view.] But the user can use the color bands as a visual cue that the object are related.  [Here’s how I moved everything into containers.]

gauge_14

Click to view larger image

Here’s the bullet chart that Stephen Few finds so pleasing. I don’t mind it. I tried all the styles and this basic one seem to work the best. Notice that I change the background to a light bluish gray so as not to compete with the objects. I don’t know if it matters and it may just be personal preference. What’s important is that it doesn’t compete or draw attention to itself. [Otherwise It could be a distracting character.]

gauge_15

Click to enlarge image

Here’s the same dashboard with a thermometer. The bullet chart almost competes with the other colors while this one helps you zero in on what is important. The manager for Team A can see how far/close they are to reach the goal.

Click to enlarge image

Click to enlarge image

Another Dashboard with Gauges

This dashboard uses the number and the gauges in the table. The table uses a mixture of icons, bullet charts, and display rules to signal the user about the company status. When the user clicks on the YTD or MTD metrics, it controls the charts on the right.  This way the user can learn which customers and products are contributing to the revenue.

I used a bullet chart in the top left but I don’t like it. It would have been better to have everything an icon.

dashboard_example_02

Click to enlarge

Conclusion

I think dashboard gauges take design consideration for their use. There’s no reason to leave them out of your dashboard but you have to be thoughtful. You must consider the layout and how to assist the user with understanding their role.

Never miss a BI Notes post!

Click here for free subscription. Once you subscribe you'll be asked to confirm your subscription through your email account. You email address is kept private and you can unsubscribe anytime.

Why Dashboard Gauges Aren’t as Bad as You Think

2016-05-22 – 2:57 PM
gauge_thumb

Dashboard gauges were once the darling of the executive suite. Some designer took inspiration from a car dashboard and its gauges. A gauge is useful because it tells us how fast we are going or how much gas is in the tank. My newer model dashboard still has gauges but it also states how many miles until empty. That information is more useful than knowing how much gas is in the …

SAS Visual Analytics: Bubble Plots for Less Toil and Trouble

2016-04-24 – 12:55 PM
va_bubble_06

You cannot research bubble plots from any data visualization expert without reading a reference to the Hans Rosling’s Ted Talk Let My Dataset Change Your Mindset. It may have been the first time a bubble plot gained a mainstream appeal noted Stephen Few in his Show Me the Numbers book. Rosling provided an intense amount of information in one data visualization and he changed my mindset. Here’s some …

SAS VA Data Builder: Add a New Data Item

2016-02-05 – 9:25 AM
va_db_00

The SAS Visual Analytics Data Builder allows you to prepare and load data to the SAS LASR Server so you can complete analysis or build reports and dashboards. It’s a little more tricky to use than some of the other parts of SAS Visual Analytics but can usually be mastered within a day. If you need to add a data item to your data table, …

3 Methods to Remove Data from the HDFS VAPUBLIC Directory

2015-12-20 – 3:53 PM
hdfs_delete_data_thumb

When working with HDFS, you are often loading data to the various directories to upload into SAS Visual Analytics. However, sometimes you want to remove that data. Here’s three different ways to remove the data.
Method 1: Use the SAS VA Administrator Tools
If you have access to the Manage Environment area, you can use the Explore HDFS tab to interact with the data in the HDFS. You …

SAS VA: Dealing with Missing Dates

2015-12-20 – 1:27 PM
va_helper_dsn_thumb

It’s nearing the new year and you may have already started preparing reports that have the organization’s goals and the progress toward those goals. However your chart may not be displaying correctly. If your data is not setup properly then you’ll notice that you only have one or two months when you want to see the entire year even if there is no data yet. …