Module 1: Business intelligence visualizations

When a business intelligence
professional works on a project, they have to juggle many tasks. They apply effective BI practices and
tools in order to make a positive impact; they engage with stakeholders and
manage their expectations; they extract data,
transform it, and load it. They optimize and maintain databases. Then they get to begin the next exciting
part of a BI professional’s role: monitoring and reporting. These activities, bring together all
aspects of BI to produce a usable tool, visually represented data that can be
shared with and understood by others. After all,
sharing data is what makes it powerful. It’s one of the key ways in which
information becomes intelligence. Hello, welcome to the final course of the
Google Business Intelligence Certificate. I’m your instructor, Terence and I’m a Business Intelligence Analyst
here at Google. I’m excited to work with you as
we explore data monitoring and reporting as a BI analyst at Google. I’m responsible for developing a deep
understanding of the problems that my stakeholders are trying to solve and further, leveraging our data to
provide solutions to those problems. Sometimes these solutions are simple data
queries or grids of numbers in a report. But the best solutions are often
dashboards with interactive visualizations that can immediately draw attention to
the most important trends and insights. This can save hours of work for your team. As we work together through this course,
we’ll focus on building visualizations and reporting to stakeholders. To do this BI professionals
create visualizations and dashboards that monitor data over time and
continuously answer business questions. Then they present these dashboards
to their stakeholders and improve on them as time goes on. This enables businesses to make quick
informed decisions using specially built monitoring tools. As a reminder a dashboard is
an interactive visualization tool that monitors live, incoming data. You can create a dashboard for
a variety of purposes, but BI professionals mainly use them to
display the data they’ve collected, analyzed and tracked for
a specific business purpose. In this course, you’ll learn to design
dashboards, communicate about their scope, then improve on the visualizations
as business needs evolve. First, you’ll discover what kinds of
information you need to know about a project in order to prepare
the right kind of dashboard. you will explore how to empower
stakeholders by designing interactivity into the dashboard. You’ll also gain key tools and
techniques to design dashboards yourself. First, you’ll make a plan to
create dashboard mockups. Then you’ll explore tableau a commonly
used data visualization tool which will enable you to design your own dashboards
and presentations in upcoming exercises. You’ll learn to make appropriate
design tradeoffs and whether to pre-aggregate your data in SQL,
then you’ll move on to designing charts and organizing them into simple and
effective dashboards. You’ll refine your visualizations to
adapt to evolving business needs. In addition, you’ll gain some great
strategies for presenting to and communicating with stakeholders. Having clear conversations with users
about business needs and project scope prepares you to create the best tools for
your stakeholder client or user. And you’ll definitely use these
communication skills to share your work and
launch your career as a BI professional. Once you learn each of these skills, you’ll have the chance to apply them
in a realistic business scenario. You’ll evaluate your stakeholders
needs and decide how best to plan, build and iterate on a dashboard. So let’s get going, I can’t wait to
share the exciting world of BI with you.

Hi, I’m Terrence and I’m a
BI analyst at Google. My team at Google,
we’re a central data and analytics team in the
finance organization. We help make decisions
related to headcount, how we’re investing
Google’s money, how basic sales revenue
is doing, profitability, all of those metrics to help make decisions within
the finance org. With business intelligence,
I didn’t have any formal training in school
or anything like that. I started my career as more of a traditional
financial analyst. Now I found myself being
drawn to writing SQL, creating pipelines that then took that data and
made it powerful for everyone. That BI role
was coming into its own, and I found myself
really drawn to that. In my day-to-day when
it comes to BI tools, I’m on more of the
project management side. It’s really about having a really good familiarity
with the tools that we use that ultimately
our data engineers will be using to
construct the pipelines, whether that’s Google
Cloud Platform and BigQuery, or all the
way to dashboarding. I have my hands in all
of these things as I see projects all the way
through from conceptualization, understand where the
data is coming from. Is this a feasible project? Does the data exist
in a way that we can ingest it and transform it? One of my biggest
learnings is that the simple solutions are
often the most powerful. The only way to really understand what
those solutions are is to sit and really understand
from your stakeholders, your users, what they’re
trying to solve for. You have to have a curiosity. You have to have a
burning passion to go out there and
want to understand, “Hey, how does your
business work? What are the problems
you’re facing? What questions are
you trying to answer? What’s your day-to-day like?” You have to have that
really base understanding and then deep
understanding to really go and take the data that may be available to you and take
it that additional mile. Really pull out
important insights. Allow them to make
decisions on data that they didn’t even
know existed before. You need to have
that passion about understanding the
business, be curious. Understand what problems
they’re facing. And ultimately
that’s what I find really fun about this field. Being able to have
that opportunity, being able to be empowered to go understand the problems
that the business is facing and come up with really creative solutions to
those problems with data.

Hello, again. Welcome to the first section
of this course. In the next few lessons, we’ll be exploring
the foundations of BI dashboards and how they’re different
from some other kinds of data analytics dashboards. We’ll also discover
many reasons why dynamic visualizations
are an excellent way to answer business questions, especially when the answers to those questions
change over time. You’ll learn how a
dashboard’s interactivity empowers the stakeholders
who you’re building it for. Also, we’ll get into the specific steps to take
while planning a dashboard. To do this, you’ll even get
to create your own mockup, an essential step in
dashboard building. Finally, you’ll cover
the tools you might use to build BI visualizations
in a professional setting. If you haven’t already, you’ll get set up with Tableau, the visualization tool you’ll
be using in this course. The next few lessons, we’ll lay the foundation
for your dashboard journey. I hope you’re as excited
as I am to get started.

Data Visualization in Business Intelligence

Overview

Data visualization is the graphical representation of data. It is a powerful tool for business intelligence (BI) professionals, as it can help them to communicate insights to decision-makers in a clear and concise way.

Benefits of data visualization in BI

Data visualization offers a number of benefits for BI professionals, including:

• Improved communication: Data visualization can help BI professionals to communicate complex data insights to decision-makers in a way that is easy to understand and interpret.
• Increased engagement: Data visualizations are often more engaging than traditional text-based reports, which can lead to decision-makers paying more attention to the data and insights.
• Faster decision-making: Data visualizations can help decision-makers to quickly identify patterns and trends in data, which can lead to faster and more informed decision-making.

Types of data visualizations in BI

There are many different types of data visualizations that can be used in BI, including:

• Charts: Charts are a common type of data visualization that can be used to represent a variety of data types, such as trends, comparisons, and distributions.
• Graphs: Graphs are another common type of data visualization that can be used to represent relationships between different data points.
• Maps: Maps can be used to visualize data that is geographically distributed.
• Dashboards: Dashboards are collections of data visualizations that are often used to provide a comprehensive overview of key performance indicators (KPIs) and other important metrics.

How to create effective data visualizations in BI

To create effective data visualizations in BI, it is important to consider the following:

• Audience: Who is the target audience for the data visualization? What is their level of technical expertise?
• Purpose: What is the purpose of the data visualization? What insights should it communicate?
• Data: What data is available to create the data visualization? Is the data clean and accurate?
• Design: How should the data visualization be designed to be effective and engaging?

Best practices for data visualization in BI

Here are some best practices for data visualization in BI:

• Use clear and concise labels and titles.
• Choose the right visualization type for the data and the audience.
• Use color and formatting to highlight important data points.
• Avoid cluttering the visualization with too much data.
• Make the visualization interactive, if possible.

Conclusion

Data visualization is a powerful tool that BI professionals can use to communicate insights to decision-makers in a clear and concise way. By following the tips above, BI professionals can create effective data visualizations that help decision-makers to make informed decisions and solve problems.

Here are some additional tips for creating effective data visualizations in BI:

• Use storytelling to help communicate the insights from the data.
• Consider the cultural context of the audience when designing the visualization.
• Get feedback from others on the visualization before sharing it with a wider audience.

By following these tips, BI professionals can create data visualizations that are both informative and engaging.

Why might a business intelligence professional choose to use a dynamic visualization in a dashboard?

They want the visualization to be interactive or change over time.

A business intelligence professional might choose to use a dynamic visualization if they want it to be interactive or change over time.

You probably know that data visualization
is the graphical representation of data. Maybe you’ve even had some experience
creating charts, graphs and other types of data visualizations. If you’ve earned your Google Data
Analytics Certificate, you explored many key data viz concepts, including
accessibility and design thinking. If you’d like, feel free to spend some time
going over these lessons again before moving forward. So, you have some familiarity
with data visualization. But did you know that data analytics
visualizations are not exactly the same as business intelligence visualizations?
In many ways, they’re pretty similar. Both are visual representations of data
that communicate insights to the people who interpret them. They aim to share data in a clear and
easy-to-digest manner. However, they differ in context. A DA visualization aims to answer
a business question, such as, “How is our company revenue
trending this year?” This chart uses only existing data.
A BI visualization, on the other hand, seeks to answer that
question in the long term. So, BI Professionals might ask: “How is
revenue trending year over year?” and “What other metrics related to revenue
might indicate an upcoming change in revenue trends?” A BI visualization tracks, or monitors, data that is relevant
to an ongoing business question. This means that sometimes it’s built for
data that doesn’t even exist yet. In addition, the insights that come from
a DA visualization would typically be used to make an important business decision. However, BI visualizations are used
to make one decision now, but then another decision in the future, as the track data updates and lends new
insights. In this way, BI visualizations are generally more dynamic than static,
which means they are interactive, or change over time. When
creating a dashboard, a BI Professional takes a different
approach than a data analyst would. This is because their work
goes beyond building and designing the dashboard to also
maintaining it as time goes on. Here’s an example: say a BI Professional
at an airline designed a dashboard for stakeholders who want to monitor new
airplane demand over the next 10 years. They include six different chart types, so the users can easily identify important
metrics and KPIs related to consumers’ travel frequency, the useful product
life cycle of airplanes, demand for new crafts, and lots more. But perhaps something in
the industry changes. The stakeholders might ask for new KPIs, different arrangements of visual
elements, or similar updates. It’s the BI professional’s job
to interpret their requests, determine if they’re reasonable and
feasible, and implement them into the dash. In this way, the dashboard keeps informing
decisions, solving problems, and answering key questions for
the long term. As you’ve learned, the ability to proactively
respond to change and remain continuously useful
is an essential part of BI. Next time we’ll go deeper into empowering
stakeholders with data visualizations!

As a BI professional, you will build tools
that monitor and showcase important
business data. As you’re learning
the main purpose of tracking this data
and building it into a dashboard is to enable stakeholders to answer
their own data questions. At its core a
dashboard is a tool. For example, a
dashboard that tracks annual revenue can
tell you a story about how a small business went viral overnight or a
dashboard monitoring the global marketplace
can tell you that competition has increased
in a particular market. The stories told by
dashboards provide important information
about what’s currently happening with
the company’s data, then these stories provide
opportunities for action. Maybe a company facing new competition now
knows that it needs to introduce a new product line or improve an existing one, and visualizations are more
helpful than just raw data because they help make
the data insights far more clear and
understandable. Sometimes a significant
trend might be nearly impossible to identify
without a visualization. For example, in this
spreadsheet about a company’s international
sales all of the data is here, but its meaning is not
immediately apparent. On the other hand, a
dashboard about this data can much more clearly
represent what’s going on. After all the primary purpose of a dashboard isn’t just to
be aesthetically pleasing, it’s to answer a question
or solve a problem. Imagine that you need to find a company’s total sales
for the year 2019. This chart from a dashboard answers that question
very simply. You can easily find
the data point for the total of 2019 sales. Dashboards are also
interactive making them extremely
flexible tools that put the power in
the user’s hands. A dashboard doesn’t answer one question and then
lose its usefulness, instead it can answer
multiple questions based on what the user is searching for and how it’s being used. Going back to our example. Maybe a stakeholder wants to use the dashboard to find out which country has the
greatest total sales across all years. They can simply
change the setting from within the
dashboard environment to focus on different data or the same data in
a different way. By creating and maintaining
a multifunctional tool, BI professionals can reduce the amount of work
people need to do. Instead of building
10 separate charts to answer 10 questions, just one interactive
dashboard does the job. As you know, your stakeholders
use the tools you build to make informed
business decisions. Your role as a BI
professional will rarely involve interpreting the data that your dashboard displays, instead, you’ll
create a dash that empowers users to interpret
the data on their own. You’ll determine what the
dashboard needs to do, build the appropriate tool, and maintain and improve
on it over time. Coming up, we’ll
investigate more about stakeholder goals and create a mockup from a dashboard plan. You’ll soon be on your way to planning for the perfect dash.

Like any important professional endeavor, dashboards require a lot
of planning to get right. They also require a lot of collaboration. You’ve already worked on
many aspects of BI planning, including a stakeholder requirements
document, a project requirements document, and a strategy document. In this video, we’re going to
learn more about planning and gain some strategies for
dashboards specifically. Typically, a BI professional works with
their stakeholders to figure out what needs to be included in the dash. The process usually begins with
a stakeholder describing their needs. For example, a sales or marketing executive may want a dashboard
that tracks customer spending habits. They could ask for visualizations
that demonstrate how often customers buy the same product again, or how much
revenue increases at peak shopping times, such as during certain holidays. After stakeholders explain their needs, a BI professional follows up
with their own questions. This helps determine
the level of detail involved. Some questions I like to ask are:
What’s the dashboard supposed to do? And what KPIs and
dimensions is it tracking? It’s helpful to also ask
who the dashboard is being built for. Is it someone who needs the most
specific level of detail or someone who just needs
high-level insights? Lastly, the BI professional must consider
where to find the necessary data and the timeline it should represent. Of course, there are a limitless number of
questions that can help shed light about any given project. As you continue to gain context, you might
identify some gaps in your understanding or find some pain points that you, and
even the stakeholders, weren’t expecting. Following up on these new discoveries
will help you ensure you address your stakeholder’s needs more directly and
create an effective tool for them to use. In addition to asking questions, many BI professionals will observe
their stakeholders at work. This can be actually sitting alongside
them at their workstations or sharing their screens in order to
demonstrate a particular workflow. This is extremely helpful because, often,
a stakeholder is so close to their own processes that they won’t notice where or
how a dashboard can help them. This exercise requires time, but
it yields a ton of great insights. Likewise, it’s valuable to iterate
alongside your stakeholders. We talk about iteration a lot in BI. As a refresher, this just means
repeating a procedure over and over again in order to keep getting
closer to the desired result. And as you’ve learned,
desired results are always changing, which is why BI professionals
iterate all the time. The first step of the dashboard iteration
process is to create a low-fidelity mockup. A low-fidelity mockup is a simple draft
of a visualization that is used for planning a dashboard and
evaluating its progress. It may be a pen-and-paper model
of how it’s organized, or a dashboard with a very
limited amount of data. The stakeholder provides
feedback, which you act on. Sharing a low-fidelity mockup is an
excellent way to include your stakeholders and make the most of your collaboration. Also, it’s very effective at
avoiding potential errors. If you make an assumption that turns out
to be incorrect, the mockup can help you and your users identify it and
implement the necessary fixes. Soon, we’ll explore even more
about dashboard planning. After all, the best way to reach a goal
is with a clear and thoughtful plan.

A company’s servers are overwhelmed because they are being used to track too many key performance indicators. What type of problem is this company experiencing?

Tool problem

This company is experiencing a tool problem, which is a dashboard issue involving its hardware or software.

If you don’t properly plan before creating
a dashboard, it’s only common sense that you might make mistakes. But what
kinds of problems can occur. In this video, we will explore some of the most
common consequences of not adequately planning for a dashboard. There are three main types of dashboard
issues that BI professionals encounter. Put simply, they are data problems,
tool problems, and audience problems. Let’s consider each of these more closely. First, when it comes to a BI dashboard, a data problem is an issue caused by the
data being used that’s in the dashboard. Maybe you’ve included more data than
you need just as you’ve discovered with pipelines. This can cause dashboards to run slowly. On the other hand, if you don’t include
enough data, you might be missing crucial information that your stakeholders
could really benefit from having. Perhaps you incorporated incorrect data
sets or data from the wrong time period. This can cause your stakeholders
to come to inaccurate conclusions. And if you don’t understand what data
is actually available, you might waste time and resources trying to obtain
data that isn’t there. Avoiding data problems requires you to be very specific
before you begin building any BI tool. Some key points to consider include
making sure you know what data should be displayed and
where that data should come from. Also set up a plan for
how you will access the data and import it into your dashboard. And
confirm that you have permission to access the data and
you know what file type you’ll use. Now let’s explore tool problems. These are dashboard issues involving
the hardware or software being used. The most basic type of tool problem is simply
selecting the wrong tool. As you know, each company uses
particular coding languages and data visualization programs. And in most cases, the BI professional must
use their organization’s preferred tools. As you’ve been learning, many of the skills
you apply to one tool easily transfers to another. But you still need to understand
how these tools are different so you can navigate them effectively. Other tool problems relate
to processing limitations. Some programs or servers may become overwhelmed if they’re
tracking too many KPIs. Sometimes, they can even crash if too many people are
making edits to the same visualization. If you run into a hardware limitation
during your building phase, you may need to reduce the project scope
or find a more creative solution, which you will learn more about
later on in this course. Okay, finally, we have audience problems. These are dashboard issues caused by
failing to adequately consider the needs of the user. For instance, imagine you’re tasked
with creating a dashboard for a company’s human resources team. The tool needs to keep track of how
many employees the company has hired, or will hire, and how many have resigned. If your dashboard also includes financial
and operational metrics, these won’t be relevant to your audience and may clutter
the dash causing complexity or confusion. Which means your tool won’t be as helpful
to stakeholders as it could be. Another kind of audience problem
concerns privacy and permissions. It’s your responsibility to ensure that
your dashboard only includes information that the users are permitted to access,
rather than something sensitive or private. It’s especially important to consider the
needs of different audiences when you’re building a single dashboard for
multiple users. For example, say you’re creating a dash to be used by
a company’s marketing team across Africa, Asia, and Europe,
as well as by the chief marketing officer. The CMO would need data about all
regions, where the regional teams may only require access to their own data. In this case, you would configure
the privacy settings to account for multiple permissions. Later on, you’ll learn more about
how privacy permissions work, but for now you’ve explored some
of the most common problems that occur when building dashboards. More importantly,
you understand how to avoid them. And being proactive is an essential skill for
any BI professional.

You’ve been investigating various BI visualization
tools and how they may have different
user interfaces or UIs and workstreams. In this video, we’ll learn more about the
different tools you might use to visualize data
and create dashboards, as well as how your skills in one program can
transfer to another. As you know, BI visualization
tools enable you to visually represent the data you collect and prepare
for a dashboard. With some programs,
you can transform data and make a dashboard
using the same software. Others are specifically
dedicated to visualization and have limited functionality
for transformation. In this course, we’ll be
working in Tableau most often. It’s a powerful visual analytics platform
that enables you to create BI dashboards
and data visualizations. Tableau is used by
many industry experts, and has a straightforward
and friendly UI. If you earned your Google
data analytics certificate, you already have some
experience with Tableau. To prepare for what’s to come, feel free to take some time
to review those lessons. Some other programs
you might use as a BI professional include
Google Data Studio, Looker, and Power BI. Let’s explore them briefly now. Google Data Studio is a reporting tool that’s
all about simplicity. It focuses more on reporting
than data transformation, but it can still
be used to create a variety of dashboards. Looker is a data aggregation and dashboarding tool
similar to Tableau. It uses its own modeling
language called Look ML. Power BI is Microsoft’s
dashboarding and reporting tool, which is similar to Tableau. Some of its functions may offer different capabilities
depending on the volume of data
you’re working with. These programs behave
fairly similarly. You may use any
of these tools to complete the activities
in this course. But keep in mind that most
of the instructions will be tailored to Tableau’s
UI and functionality. If you do choose to work
in another program, you may need to take
additional or different steps to finish the tasks. To start, in the next activity, you will set up your
Tableau public account. Tableau public is a free
platform to explore, create, and publicly shared
data visualizations online. You can use it to
perform most of Tableau’s functions on any
operating system at no cost. The second activity is optional. In it, you can download the Tableau public desktop edition application
to your computer. While in browser functionality
is nearly identical, some may prefer the
desktop version. Because it is only available
with Windows and MacOS, it is not required to
complete this course. Tableau public desktop does have functions that the in-browser
Tableau public does not. For example it
enables you to use Tableau without an
Internet connection. But again, it’s not required to complete the activities
in this course. One important thing to
keep in mind is that these programs are constantly
being updated and refined. They might appear differently
as changes are made. If this happens, take the opportunity to troubleshoot
the issue yourself. This is excellent practice
for actual business scenarios you’ll encounter
a lot as a BI professional? You can hone your skills
as a data detective as you practice the art
of trial and error. If you truly get stuck, you can always refer
to the programs, help resources, our forums, and the programs
customer support. One last thing, when you create your Tableau
public account, remember that potential
employers can view your public profile as part
of your work portfolio. This is helpful to remember after you’ve completed
this program. Tableau also has a
feature where you can let potential employers know
you’re available for hire. This can be enabled by
checking the available for hire box on your
Tableau public profile.

Throughout this
course, you’ve been discovering so much
about the world of BI. You now know how interactive BI visualizations
empower stakeholders to make decisions. You’ve explored how to
effectively plan for a dashboard, as well as how to build
a low-fidelity mockup. You’ve also set up
the tools you will use to build dynamic
visualizations. Once you’re in a
professional setting, you’ll be able to use these experiences to plan
for excellent dashboards. But for now, it’s time to turn your attention to
the next assessment. As a reminder, these challenges
are here to help you evaluate your learning and progress throughout the course. Think of them as an
essential measurement of your professional growth. Take some time to review
any material you like, including the latest glossary, which is full of new vocabulary, definitions, and concepts. Good luck! You’re
going to do great.