Communicating

Data Stories to
Executives:
Top 5 Best
Practices for
Dashboard Design
Creating dashboards isn’t just about storing information. benefit from having a more effective visual format for their
They’re useful for monitoring, analyzing and sharing insight in everyday use, clear information for easy onboarding of new and
an efficient way. Useful dashboards can elevate data analysis future team members, and important data and visuals that will
tasks, and bridge the gap between data and action. Viewers make cross-functional collaboration easier.
should be able to look at a dashboard and go, “I understand
what’s going on and exactly what I need to do now.”
Sometimes, dashboards, like the one below, aren’t created to
their full potential, which masks not only the work that went
on behind the scenes – but also makes the insight gathered
obsolete to other viewers due to the level of comprehension.
This guide is a breakdown of our top 5 best practices for
dashboard design to communicate data stories.

While this guide will focus on displaying technical data in a

digestible and impactful way, primarily benefiting executives
and business users, these tips are best practices that will
strengthen every dashboard. The security operations center
(SOC) or network operating center (NOC) analyst-at-large can This dashboard lacks proper labeling, confusing metrics, misguided coloring and distracting
visuals – all of which draw attention away from the important insight it should communicate.

Communicating Data Stories to Executives: Top 5 Best Practices for Dashboard Design | Splunk 2
01
Understand the story
you want to tell
Dashboards should be designed with the intended audience in mind.
An investigative dashboard for experienced viewers would turn out differently
than a simple overview for management.

You can enhance your dashboard to meet the needs of non-expert viewers, but
it’s important to keep that in mind as you’re creating. Write down the purpose of
your dashboard and determine the:

1. Pre-existing experience and knowledge the audience has — How expert

will they be about the subject and/or display conventions?
2. Information they need to gather from the page — Is your goal to explore,
document, and/or explain your data?
3. Actions they need to take after acquiring insight from the page.

Communicating Data Stories to Executives: Top 5 Best Practices for Dashboard Design | Splunk 3
02
Arrange relevant content
in an appropriate layout
Consider the importance of each piece of data you place on the dashboard.
Is it something that the viewer needs to read? Is it necessary to help tell the
story? If it’s optional or supplementary information, consider hiding it behind
an interaction, such as a tooltip on hover or drilldown on click. Adding too many
visuals to a page can be extremely distracting and draw attention away from the
actual insight you want to highlight.

Display modes can vary across your audience. Design your content size, layout
and interactions based on the devices you expect viewers to be on. For example,
a large TV monitor will need larger font sizes and may not have hover capabilities.

A good dashboard should provide:

1. Brief context on the task at hand to give an overview so that any viewer can
understand what they’re supposed to look out for.
2. Key information with common workflows set up to minimize clicks and
simplify repetitive tasks.
3. Indication of action and how to do so if necessary.

Communicating Data Stories to Executives: Top 5 Best Practices for Dashboard Design | Splunk 4
 Within one screen view, these examples: 1) include descriptions and metrics
at the top of the dashboard to give brief context on the task at hand so that
any viewer can understand what they’re supposed to look out for, 2) provide
quick context and insight on these metrics over time, by category, or geographically as relevant
metrics, recognize information to fully understand the current state, and 3) have detailed, further
issues that investigable data at the bottom if action is necessary with corresponding
need attention
descriptions on how viewers can take action.

more info and

breakdown to Read more about dashboard layouts in Dashboard Design: Getting
fully understand
the current state Started With Best Practices (Part 1).

investigate further
if action is needed,
and links out of
the dashboard

quick context and

metrics, recognize
issues that
need attention

more info and

breakdown to
fully understand
the current state

investigate further
if action is needed,
and links out of
the dashboard

Communicating Data Stories to Executives: Top 5 Best Practices for Dashboard Design | Splunk 5
03
Choose suitable
visualizations and
configurations
Visualizations transform data into a meaningful story with the right type and There are many different ways to visualize data. Select the visualization most
configurations. Every visualization should be: appropriate to convey the desired message and underlying information as
quickly as possible. A visualization, like the one below, is typically composed of a
1. Meaningful — Choose a visualization, style, and layout that accurately couple of aspects:
portrays the story you want to tell. Every visualization should convey
a message. Ensure the data visualization is clear, accurate and 1. Data and display
comprehensible. 2. Color and style
2. Essential — Prioritize information that will add to the story rather than 3. Axis formatting
distract. Highlight the most important pieces. Utilize space efficiently and
apply semantic notation when possible. 4. Interactions

3. Usable — Help users of all levels and devices view and understand your
visualization. Adapt your design for different platforms, languages and
access needs.

Communicating Data Stories to Executives: Top 5 Best Practices for Dashboard Design | Splunk 6
 Data and display

Color and style

Axis formatting

Interactions

Data and display can be chosen based on the data available and the insight you
wish to display. For example, a line chart based on the horizontal axis typically
requires time series data and represents data over a period of time, which is
useful to understand trends, highlight anomalies, and possibly compare multiple
series. Do you want to show pre-existing, distinct categories? Or do you want
to identify new, previously unknown patterns? Leverage display methods like
outlier visuals or column stacking to best communicate your data.

Communicating Data Stories to Executives: Top 5 Best Practices for Dashboard Design | Splunk 7
Popular visualization types used include: More advanced visualizations can be used to enhance your dashboard. You can
use a Choropleth SVG image to display anything from room capacity within a
Visualization % of Dashboards Common Uses building and electricity usage across various appliances in your home too.

Table 60 Useful for straightforward

investigation and analysis.

Single value 35 A quick and simple way

to communicate metrics.
It’s important to provide
titles and/or descriptions
to ensure appropriate
representation. Link graphs can be used to efficiently analyze connections between distinct
entities, such as relationships between threat objects and risk objects in an
Column chart 29 Charts based on the alerted risk event.
horizontal axis typically
Line chart 28 display time-series data.

Column charts are useful

to highlight discrete points,
while line charts represent
overall trends.

Pie 20 Useful for single series to

highlight how the partial
categorical elements add up
to a whole.

These % are based on usage within dashboards in Splunk Enterprise. Tables are useful for simple
investigation and analysis. Single values are the quickest way to communicate metrics.

Communicating Data Stories to Executives: Top 5 Best Practices for Dashboard Design | Splunk 8
Color and style should be configured with their possible semantic meanings
in consideration. Categorical coloring is helpful for comparison series, whereas
color palettes that sequentially increase in luminance and saturation are more
effective for ordinal data tied to a specific count or value.

Luminance
effective for ordinal data
(limited accuracy in perception)
Saturation

Hue very effective for categorical data

Furthermore, specific colors like green and red often mean “good” and “bad”,
respectively (or Christmas, if you’re a regular celebrator).

Axis formatting is helpful for customizing the frame of reference for the data
represented, including contextual information like axis titles, corresponding
units, labels, appropriate scales and possible overlay fields. It’s also useful
for additional formatting details such as grid lines and tick marks, which are
reference points that help people visually locate the position of values along the
axis. Charts may also display grid lines that each extend from a tick (or in replace
of one) across the plot area to help people visually estimate a data value when
its mark isn’t near an axis.

Interactions can be added to a visualization to highlight specific connections or

to provide secondary information on hover. Add additional information in tooltips
if it’s not critical to display at first glance to understand the overall visualization.

Read more about visualizations in Dashboard Design: Visualization Choices and

Configurations (Part 2).

Communicating Data Stories to Executives: Top 5 Best Practices for Dashboard Design | Splunk 9
04
Enhance details
for your team
Supplementary metadata provides background information on a subject
and can include relevant links to documentation, related data, and contact
information. Ensure it’s intentionally selected information and leverage links
to hide extensive URLs behind a summary word, phrase, or image viewers
can click on.

It’s important with any design that you gather feedback and iterate! Keep in mind
that although it may work for one person, it might need to be adapted slightly
for another.

1. Dashboards should feel collaborative — Design dashboards with other

team members updating as a first-class activity. Updating dashboards is
critical to keeping your data healthy and you want to find ways to make
dashboards feel like they belong to a team. Make it easy for users to add
context, leave feedback and update when necessary.
2. Dashboards should inspire action — If a chart indicates something that has
changed, there should be clear actions and places to go from there. Think
about how the user will notice they need to take action and where to click to
do so. What are the next steps the user will need to be aware of? Will users
understand how various dashboards and actions are related to each other?

Communicating Data Stories to Executives: Top 5 Best Practices for Dashboard Design | Splunk 10
05
Add applicable
interactivity and
connectivity
For further investigation, leverage highlighted points on visualizations to indicate
interactivity and pass selected values to another search and visualization
for further investigation. Filters are extremely helpful to focus on relevant
data such as a certain time range or category. Other interactions such as
drilldowns or settings tokens are great ways to engage your audience and
provide opportunities for critical thinking. It’s important to ensure however, that
there’s enough context on the page for non-expert viewers to recognize and
understand the interaction, such as additional descriptions and/or styles to
show where connections lie.

Turning data into action can be difficult, but you can use these dashboard design
best practices to communicate your data insights to any audience in your
organization. You can effectively share the value of your insights to executives,
visualize the impact and context of your work to cross-functional teams, and
create more effective, nuanced dashboards for your growing team. Splunk’s
Dashboard Studio offers an intuitive editing experience to easily build powerful,
storytelling dashboards with a point-and-click editor, customizable formats,
flexible layouts and advanced visualization tools.

Communicating Data Stories to Executives: Top 5 Best Practices for Dashboard Design | Splunk 11
Start a free Splunk Cloud Platform or Splunk Enterprise trial
today to explore what Dashboard Studio can do for you.

Get Started

Splunk, Splunk> and Turn Data Into Doing are trademarks and registered trademarks of Splunk Inc. in the United States and
other countries. All other brand names, product names or trademarks belong to their respective owners. © 2023 Splunk Inc.
All rights reserved.

23-Splunk-Top 5 Dashboard Design Listicle-SS_104