Data Visualization

Visualizing Structured Data Using Dashboards

Data Visualization

Table of Contents
Introduction ............................................................................................................................................................................. 4

1. Characteristics of Good Dashboards ................................................................................................................................ 5

1. An Overview of Good Dashboards .................................................................................................................................... 5

1.1. Types of Dashboards .................................................................................................................................................. 5

1.2. Building a Dashboard ................................................................................................................................................. 8

1.3. Characteristics of a good dashboard? ....................................................................................................................... 9

2.1. Assessing dashboard requirements ......................................................................................................................... 10

2.2. Perspectives on dashboard design ......................................................................................................................... 10

2.3. Suggested Best Practices ........................................................................................................................................ 11

2.4. Pitfalls to avoid ......................................................................................................................................................... 12

Summary ............................................................................................................................................................................... 13
Data Visualization
Data Visualization

Introduction
The topic focuses on describing information dashboards which is the most widely used
medium for visualizing structured business data. To monitor and manage the wide array of
business functions and processes within organisations, dashboards provide a summarised
and constantly updated view of the key influencing parameters. Information dashboards
gather data from multiple sources and display them on one reporting screen, which
significantly reduces the time spent on compiling and sharing reports, and analysis findings
across the organisation.

Designing the dashboard involves in-depth discussion with users to identify the dashboard
requirements. There are multiple perspectives involved in analysing these user needs and
arriving at the final design specifications.

Learning Objectives
Upon completion of this topic, you will be able to:
• Define dashboard and its types
• Describe the guidelines to build dashboards
• Summarise the charecteritics of good dashboards
• Determine the key considerations for designing dashboards
Data Visualization

1. Characteristics of Good Dashboards

1. An Overview of Good Dashboards

Stephen Few, one of the famous luminaries of data visualization
today, defines a dashboard as below:

“A dashboard is a visual display of the most important

information needed to achieve one or more objectives that has
been consolidated on a single computer screen so it can be
monitored at a glance.

The highlighted words convey the chief characteristics of an

enterprise dashboard.

Image 1.1.- Stephen Few

1.1. Types of Dashboards

Dashboards can be categorised into the following 3 basic types:
• Strategic: These provide an overview of the high‐level measures of organisation
performance against its strategic goals.
Typical examples of these are executive dashboards which give the quick overview
which leaders require to monitor the health of the business.
Data Visualization

Figure 1.2. - Example of an executive dashboard

Source - www.diskel.co.uk

• Operational: These are updated in real-time and used primarily for monitoring
business operations.
In comparison to strategic dashboards, operational dashboards need to grab the
manager’s attention immediately if the operational parameters cross the accepted
limits. The details required are also higher than in strategic dashboards, and the
ability to drill down further is an important feature.
Data Visualization

1.3. - Example of an operational dashboard

Source - www.inetsoft.com

• Analytical: They support interactions with the data such as drilling down to enable
exploration. They are mainly used for forward-looking evaluation and planning.
Data Visualization

1.4. - Example of an analytical dashboard

Source - i0.wp.com/parasdoshi1989.files.wordpress.com/2012/11/excel-powerpivot-twitter-social-media-analytics-
dashboard-1.png

1.2. Building a Dashboard

Given below is a set of fundamental guidelines for building a dashboard.
1. Visual displays: Information needs to be presented as a combination of words and
graphics, but with a focus on the latter.
2. Information to achieve specific objectives: Data from sources related to several
organisational functions needs to be gathered to provide the user the information
required to do their job.
3. Single screen: All the information must be visible within the user’s eye-span to view
within a glance. If the user has to scroll up and down or switch screens, it defeats the
objective.
4. Monitor at a glance: The dashboard must instantly highlight anything which needs
attention and requires action. The focus is on giving an overview rather than too much
detail.
5. Customised: The dashboard must be built for a specific role, group or function.
Data Visualization

6. Effective display media: The display media chosen to represent information needs to
be clear and descriptive, but at the same time compact in nature to fit into the single
screen format of the dashboard.

1.3. Characteristics of a good dashboard?

A well-designed dashboard will have the following characteristics:
• Well Organised
It provides a high-level view of the ground situation, and at the same time enables
quick access to additional information which may be required beyond the dashboard.
• Condensed Information
To communicate a large amount of information, summarized numbers are used to
represent the data conveyed by a large set of numbers as a single number e.g. sums,
averages and rates.
• Customisable
The dashboard is designed to provide the information required to meet an objective or
set of objectives.
• Clear and Concise Display Element
The display element chosen would be the most commonly used one to display that
particular kind of information in most dashboards. Even if the display element is
compressed to fit into the dashboard, the message conveyed would remain clear.
Data Visualization

2. Designing the Dashboard

The process of designing the dashboard commences with conducting in-depth discussions
with the user community to arrive at dashboard requirements.

2.1. Assessing dashboard requirements

To come up with a good dashboard design, it is essential to ascertain from the final users
what they really need from it. Some of the typical questions which need to be answered at
this stage are:
• Who will use the dashboard i.e. whether it is for a person, group or departmental
function?
• What objective does it support?
• What will the users monitor with the dashboard?
• What are the actions which users will take based on dashboard alerts?

Additionally, specific information on aspects such as user expertise level, technology

platform and display modes is collected which can have an influence on the dashboard.

2.2. Perspectives on dashboard design

Given below are six key perspectives which should be evaluated while designing the
dashboard. For each of these perspectives, there are a set of critical questions to be
addressed, which will clarify the various design aspects influencing the dashboard.

1. Business Perspective
a. What is the role of the target business user (executive, manager, data analyst)?
b. What are the business questions that the target user will need to answer?
c. What decisions does the target user make in his/her role?
d. Who are the other business stakeholders who will use the information?
e. What Key Performance Indicators (KPIs) are essential to make the critical
decisions?
2. User Perspective
a. What is the type of business user? - novice/expert/analyst
Data Visualization

b. What type of devices do users use?

c. What are the personalisation and internationalisation needs?
d. Are there any accessibility considerations?
e. What are some of the information display preferences of these users?
3. Data Visualization Perspective
a. Dashboard layout
b. Selection of graphing medium such as charts, images, icons.
c. Selection of appropriate navigation and action medium (smart devices)
d. Appropreate font selection to ensure attractiveness and readability
e. Table formats for display
f. Interactivity options
4. Usability Perspective
a. Emphasise graphs and reduce text
b. Dashboard elements order – proximity, similarity, grouping, connections
c. Values and text – contrast, labels and color variants
d. Table format
e. Navigation design
5. Technology Perspective
a. Device support
b. Data volume, online and offline processing
c. Interface standards
d. Data update latency
e. Enterprise roll-out requirements
f. Dashboard release cycles
2.3. Suggested Best Practices
Some of the best practices to be followed during design are:
• Involve users and obtain decision-maker perspectives
• Develop alternative dashboard designs
• Use agile/iterative development approach
• Develop a standard-style guide early in the design cycle
• Provide business insight and not just facts
Data Visualization

• Ensure trustworthy data sources and currency of data

• Ensure that the dashboard promotes Action!

2.4. Pitfalls to avoid

Some of the typical mistakes which need to be avoided during dashboard design and
development are:
• Filling up the dashboard with too many unfamiliar metrics
• Not aligning the dashboard to business goals
• Ineffective graphical elements usage in terms of colour/text/navigation
• Adding too much visual clutter and complexity
• Insufficient focus on implementation planning, user education and support.
Data Visualization

Summary
Dashboards are the most widely used and effective mediums of visualization. They are used
to display structured business data visually to key decision-makers in enterprises.
Dashboards are of three basic types – strategic, operational and analytical.

While building the dashboard, a few important guidelines need to be followed such as:
• More graphical media than text
• Achieve specific objectives
• Displayed in one single screen
• Customised for a specific set of user(s)

Before designing the dashboard, high-level needs of the user has to be ascertained. These
needs would be both in terms of functional business needs as well as non-functional
technical requirements.

Designing the dashboard involves critically analysing the needs, captured from users for the
dashboard and arriving at the right specifications for the build to take place. This activity is
approached from the perspectives of business, user, metrics and KPI, data visualization,
usability and technology.