CHOOSING THE RIGHT

CHART FOR YOUR IT DATA

Introduction
Visualizations are a great way to gain insight on IT data. They help you understand complex
data points, convey important findings, support theories, report progress, and correlate past
and present trends. However, getting visualizations right isn't easy. With so many types of
visualizations available, how do you choose which one's best for you?

Here at Analytics Plus, we follow two principles when designing a visualization:

1. Determine your visualization goal.

Unless you start out with a clear goal of what you want to achieve with a report, you're never
going to get it right. Typically, the crux of visualization falls under one of these categories:

• To inform
• To compare
• To show the distribution
• To establish relationships

2. Choose the best visualization for your data.

Once you have a goal in place, the next step is to choose the visualization that best conveys
your goals. For example, if your goal is to spot trouble areas, your visualization should help you
achieve that with great clarity. Alternatively, if your goal is to plan for the future, your
visualization should help you effectively look into your past trends.

Let's work through an example:

Consider the charts below as a means to understand the percentage of incoming requests by
category.
 Incoming request by category

Administrative

8.1% 7.1% Customer Services

7.7% Data Handling

7.2%
Faulty Hardware

8.2% 9.4% General Issues

ISP Link

Internet Link
8.6% 9.0%
MPLS Link

8.8% Network Access

8.5%
OS Corruption
8.8% 8.6%
Printer Problems

Software Request

Incoming requests by category

10%
9.36%
9%
9.01%
8.80%
8.60%
8.21%
8%
8.05%
7.74%
7.21% 7.15%
7%
Percentage of Requests

6%

5%

4%

3%

2%

1%

0
Data MPLS Hardware Network ISP Link General OS Printer Administrative

Category

Generally, pie charts are one of the most aesthetically appealing chart types; however, when
your goal is to compare small differences between different categories of data, the bar chart is
a better option to serve your goal.
Building great visualizations
There are plenty of ways to visualize data. Depending on your goal, some charts are better
than others. In this e-book, we'll explore some common types of charts and how you can use
them to gain better insights.

Funnel chart
The funnel chart is a great tool to visualize different stages of a process. Funnel charts are
great for depicting requests passing through the different stages of its life cycle as well as
change requests in different stages of implementation.

Due to its structure—wide at the top and narrow at the bottom—the funnel chart makes it easy
to spot where the maximum drop-offs occur, and it effectively illustrates transitions at each
stage. This is why the funnel chart is predominantly used to visualize processes.

Here's an example of the funnel chart showing the request journey from request creation to
request resolution.

Requests by stage

39.7%

Open

On-hold
21.3%
Resolved

Closed

14.6%

24.4%
The funnel chart can be used when you want to:

1. Display metrics across stages. The funnel chart works well in dashboards or presentations,
providing managers with a summary of request volumes in each stage without having to create
multiple reports.

2. Identify bottlenecks. The funnel chart is also a helpful aid if you're looking to identify hold-ups
in your process. The example above depicts a higher percentage of requests "On Hold," giving
you a clear picture of where the bulk of your requests are held up.

3. Understand transition rates. Need to look into how quickly requests move from one status to
the next? The funnel chart can help you visualize this with a single glance.

Line chart
When it comes to analyzing patterns in data trends, line charts are frequently the default
choice. They help you understand patterns, identify deviations, and spot outliers in your data.
The following line chart illustrates the alarm trends for the past year, starting with a sharp spike
in April and reaching a peak in December.

Alarm trends

5500
5115
5000

4500

4000

3500

3000
Alarms

2500
2102
2000 1827
1554
1500 1279
1005
1000
730
456
500
182
86 86 86
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Alarm Generation Time

The line chart can be used when you want to:

1. Study trends. It provides an easy way to visualize daily, weekly, monthly, or annual trends.
So, anytime you want to go back and see how the numbers have changed over time, the line
chart is your best option.

2. Compare data patterns. If you want a quick comparison of data trends between two or three
groups, the line chart is your best choice. However, you need to be careful not to include too
many lines, as it can clutter your graph. A quick comparison of alarm trends by severity
highlights a gradual increase in critical alarms over the past year.

Alarm trends by severity

4000
Attention
Critical
3500
Trouble
Warning
3000

2500
Alarms

2000

1500

1000

500

0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Alarm Generation Time

3. Forecast future trends. Although trend-forecasting is not a surefire way of predicting future
outcomes, the line chart can, to some extent, help you gauge how your metric is likely to
behave in the future.

Bar chart
The bar chart is suitable for comparing discrete categories of data alongside one another. Bar
charts allow you to effectively visualize the average request resolution times by site or by
technician, request volumes by categories, technician-wise ownership of alarms, technician
workload, etc.
 Technician-wise ownership of alarms

800
751 744

700 694
656
610
600 586
536
512
500 493
Alarms

400

300

200

100

0
Joe Williams Bruce Waters Brad Gessup Ed Holmes John Roberts Ed Grahhim Roger Wright Shawn Adams Steve Wilson

Technician

The bar chart can be used when you want to:

1. Compare discrete sets of data. Bar charts can be used to compare discrete data sets, such
as the number of alarms owned by each technician as shown in the graph above. Such a side-
by-side visual comparison allows you to spot anomalies right away without the need for further
analysis.

2. Arrange categories by rank. A typical use case for the bar chart is to build ranked reports,
for example: top 10 alarm categories, top 10 popular knowledge base solutions, most
expensive projects, and top performing technicians. These ranked reports can provide top-
level management with visibility into vital help desk metrics, empowering them to make
educated decisions backed by data.
 Technician-wise ownership of alarms

800
751 744

700 694
656
610
600 586
536
512
500 493
Alarms

400

300

200

100

0
Roger Wright Steve Wilson Brad Gessup John Roberts Ed Holmes Bruce Waters Shawn Adams Joe Williams Ed Grahhim

Technician

3. Compare multiple categories. In addition to the horizontal and vertical chart types, the bar
chart has many nuanced versions. The grouped bar chart can help you perform complex
comparisons between categories in your data sets. For example, if you wish to compare the
"number of alarms owned by technicians" by the status of the alarm, a group bar chart like the
one given below provides greater clarity into who owns what. Using the report below, help desk
managers can see that all technicians own a greater number of alarms in "Trouble" status.
 Technician-wise ownership of alarms
400
Attention
Critical
357
350
Service down
Trouble
305
300
283
267
252 248
250 242
241
Alarms

200

163
154
150
132

108
102
100
84
72
61 61
55 54
50

19

0
Steve Wilson Brad Gessup John Roberts Bruce Waters Robert Woodman
Technician

Pie chart
The pie chart is useful for communicating the percentage distribution of data among different
categories. For example, the pie chart below shows the current status of help desk requests
each day, making it easy for your help desk to see where they need to focus, so they can
reassign their workforce and resolve incoming requests efficiently.
 Incoming tickets by category

28%
Open

45% On-hold

Closed

Resolved
14%

12%

For a pie chart to be effective:

• Keep the number of categories to a minimum.

• In addition to the in-chart text, always have a legend by the side to offer an explanation
for your pie chart.

The pie chart can be used when you want to:

1. Understand percentage distribution. If you're looking to understand how the total is split
among different categories, the pie chart is your best bet. It's visually appealing and helps the
audience quickly understand the contribution from each category.

2. Compare percentage distribution. One look at the pie chart above and it's easy to deduce
that close to half of the incoming requests are still unassigned or open, while on-hold requests
make up one-third of the total requests. Using this chart, it's easy to deduce the statuses that
have the lowest percentage of requests.
Bubble chart
Bubble charts are versatile, simple to create, and easy to interpret. For example, the bubble
chart below clearly shows how much work each technician has on their plate. In this case, the
color of the bubbles represent the different technicians, and the size of the bubble represents
each technician's relative workload.

Technician load

140
Brad Gessup

Brad Gessup Roger Wright Steve Wilson Bruce Waters

120 Ed Holmes John Roberts Ed Holmes
Robert Whitman
Howard Stern Watson Oz
Watson Oz Lynn Hendricks
100 Bruce Waters
Heather Lee
Eve Adams
Heather Lee Joe Williams Howard Stern
Requests

Max Wood
80
Joe Williams
John Roberts
60
Lynn Hendricks
Robert Whitman
40
Max Wood
Roger Wright
20
Eve Adams
Steve Wilson
0
Ed

R
Ly
H

M
H

Jo

Ed
Br

Br

Jo

St
ob
ow
ea
at

ax

og
nn

ev
e
ad

uc

hn
H

Ad
so

er
th

er
ar

e
ol

W
e

H
G

tW
n

am
d

illi
m

W
W

en

oo

W
rL
es

ob
O

St
es

am
at

ils
rig
dr

s
hi

d
z
su

er
ee

en
er

on
tm
ic

ht
s

ts
p

ks

an

Technician

A popular spin-off of the bubble chart, the packed bubble chart, is often used to present the
keyword cloud—a cluster of bubbles—each representing a keyword used by end users to
register their request. In this instance again, the relative size of the bubble indicates the
frequency of occurence of the keyword.
 Keyword cloud report

attachment boot
extension
53
58 44 Adobe Access
41
Microsoft
Firewall 53 ERP AD
53 Linux storage 57
58 60 Adobe
wifi
Windows
Datacenter
62
Download Antivirus
44
46 AD 42
MPLS 56
Application
access 169
Network 106 website
42 sharepoint
Attachment
52 Telephone
51
Reset
41
Backup
56
WAN
25
database Boot
Router 61
58 Hardware
external
Crashes
ISP 102
42 Desktop
Login
155 59 Database
57 Directory
41 Email Datacenter
114 DHCP
52
crashes Desktop
48
backup Internet
54 Redhat
Printing
52
139 DHCP
61 PC
22
server Application Directory
60 48
OS
Password 25 Download
Mouse Monitor
55
52 MPLS
41
Antivirus
Email
upgrade
55 56 59
ERP
Installation
Keyboard
56
50
External
Firewall
+ 30 more…

Bubble charts can be used when you want to:

1. Draw comparisons. Bubble charts are a good way to draw comparisons among values by
comparing the size of the bubbles. In the above report, you can infer that the keywords
"password" and "network" are more frequently used by end users simply by looking at the size
of those bubbles.

2. Create word graphs. Words graphs are a great way to keep track of how many times
specific keywords were used by your customers. These graphs can help you monitor the
status of requests.

Dial chart
Also known as a gauge chart or a speedometer chart, the dial chart is best used for target-
based performance tracking, including tracking for CPU performance, the amount of storage
device memory used in a month, or the percentage of requests within SLAs as well as SLA
compliance rates.
 Percentage of requests within SLA

98%

0% 100%

The dial chart can be used when you want to:

1. Indicate performance. Dial charts provide instant insights into data, and they're often used in
executive dashboards to present performance results. Dial charts are great for highlighting
technician performance and focusing on compliance levels.

Map chart
Geographical data looks best when visualized on a map. In the map chart, we assign values
to a specific location based on continent, country, state/province, county, or latitude and
longitude.

The map chart provides two key pieces of information; it shows where things are happening
and how much is happening. In the graph below, you can easily spot the regions where the bulk
of your incoming requests come from (North or South America) and see the numbers recorded
by each site. Strategic insights like this can help you instantly identify soft spots in your agent
assignment time across different locations.
The map chart can be used when you want to:

1. Study geo-spatial spread. With map charts, it's easy to pin down and study data points
against geographical locations. For example, you can easily examine SLA violations by site,
alarms by site, total requests by site, or cost per request by site.

2. Create a combination of charts. Map charts can be used in combination with other chart
types. Let's look at an example. Say I need to visualize geographical data multidimensionally.
Examining the size of the bubbles in the above map, it's easy to conclude that the Cape Town
office witnessed the highest number of requests for the month.

Area chart
The area chart can be used when you want to emphasize the magnitude of fluctuations. For
example, in the area chart below, you can easily make out the fluctuations in alarm volumes
over the past eight months.
 Fluctuation in alarm volume

80% 80%
74% 72%

60%

45%
40%

20%
Alarms

6%
0%

-19%
-20%
-17%
-37%
-40%

-45% -47%
-60%
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Alarm Generation Time

The area chart can be used when you want to:

1. View data trends across time. Area charts allow you to visualize data trends of different
categories over time with remarkable clarity.

2. Understand the magnitude of change. Although you can spot fluctuations in your data trends
by using a line chart, the filling between the line segments in the area chart helps you
understand the magnitude of changes with greater clarity.

Pivot chart
The pivot chart, or pivot table, allows you to dynamically group and summarize large sets of
data for easy analysis. It helps you transform tabular data into meaningful grid summaries.

In the example below, data for the complex change approval stage and status information are
presented in the form of a grid that makes it easy to find the data you are looking for. For
example, if you wanted to look at the number of change requests rejected in the approval stage
in the month of October 2015, it would be much easier to get this information using a pivot
chart, as opposed to looking through the entire change table.
 Scheduled Start Time Status Planning Implementation Review Approval

1 Oct 2018 Accepted 4 3 5 5

2 Rejected 4 2 1 3

3 Requested 3 2 4

4 Requested for Information 7 2 1 3

5 Submitted for Authorization 2 2 1 2

Oct 2018 20 11 8 17

6 Nov 2018 Accepted 6 2 3 3

7 Rejected 1 2 6 1

8 Requested 1 3 1 2

9 Requested for Information 5 1 4 4

10 Submitted for Authorization 6 1 7 1

Nov 2018 19 9 21 11

11 Dec 2018 Accepted 8 6 4 9

12 Failed 1

13 Rejected 2 3 5 4

14 Requested for Information 2 2 4 2

15 Submitted for Authorization 1 1 3 2

Dec 2018 13 13 16 17

One clear advantage of using the pivot table is that it allows you to meaningfully summarize
text-based rows and columns into grids. In the above example, you can see the total number of
change requests approved or rejected each month as well as the number of change requests
that were submitted or were in the planning, review, or approval stages.

The pivot chart can be used when you want to:

1. Perform multidimensional data analysis. The pivot chart provides granularity to complex data
sets, and it allows you to view the data from different perspectives at a glance, without having
to go through multiple reports or charts.

Summary
Building great visualizations doesn't have to be a headache. With the right charts, it can be fun
and eye-opening; you might even discover a new opportunity simply by gaining a fresh
perspective on your data.

Can't wait to get started? Download a free, 30-day trial of Analytics Plus now, and start creating
stunning visuals today: https://www.manageengine.com/analytics-plus/download.html
About Analytics Plus
Analytics Plus is self-service IT analytics software that lets you visualize your IT data in the
form of colorful charts, reports, and dashboards. It offers out-of-the-box integrations with
ServiceDesk Plus and other ManageEngine tools that help you get an in-depth look at your IT
infrastructure. It features a simple drag-and-drop reporting interface that eliminates the need for
a data analyst to help your helpdesk managers optimize operations and improve service
delivery.

Check us out at https://www.manageengine.com/analytics-plus/