Unit-IV

Topics:
1. Introduction to data visualization
2. Dashboard design responsibilities of BI analysts
3. Importance of Data Visualization
4. Types of charts
5. Introduction to dash board
6. Types of dashboards
7. Performance Dash board architecture
8. Enterprise Dash board- characteristics, design
9. Pitfalls of dash board design

Introduction to Data Visualization:

● Data Visualization in Business Intelligence has become increasingly important in today‘s
business world
● Visual depictions of large datasets help businesses to make sense of their data, discover
trends and relationships that would be difficult to uncover through numerical
representations alone and gain powerful insights into consumer behavior and market
opportunities.
● Data Visualization is a powerful tool for converting complex data into easy-to-
understand graphics, allowing insights to be quickly derived from the information
presented.
● Visualizing data in various formats, such as pie charts, comparison bars, and demographic
maps, allows logical connections between units to be easily seen
● With interactive Data Visualization tools, companies can now bring their KPIs and
business metrics to life
● By combining these visuals with compelling storytelling, it‘s possible to provide a well-
rounded understanding of the critical elements driving success in any organization.
● Enterprises of all sizes and from diverse industries, such as retail, science, finance, and
healthcare, are leveraging the power of BI solutions to break down their data for
improved insights into business performance
Dashboard design responsibilities of BI Analysts:

Business Intelligence (BI) analysts play a crucial role in designing and maintaining
dashboards that provide valuable insights to an organization. The responsibilities of BI
analysts in dashboard design can vary depending on the specific organization and the
complexity of their data and reporting needs. However, here are some common
responsibilities of BI analysts in this context:

1. Data Gathering and Integration: BI analysts often work with various data sources,
including databases, spreadsheets, APIs, and more. They are responsible for
collecting, cleansing, and integrating data from these sources to create a unified
dataset.
2. Data Modeling: They create data models that define the structure and relationships
within the data. This is essential for creating accurate and meaningful dashboards.
3. Requirement Gathering: BI analysts work closely with stakeholders to understand
their reporting and analysis needs. They must have a clear understanding of the key
performance indicators (KPIs) and metrics that are important to the business.
4. Dashboard Design: This is a core responsibility. They design the layout, visuals, and
interactivity of dashboards. The design should be user-friendly, intuitive, and visually
appealing.
5. Data Visualization: They choose appropriate data visualization techniques (charts,
graphs, tables, etc.) to effectively communicate the insights. They ensure that the
visuals are clear and understandable.
6. Data Transformation and Manipulation: BI analysts often need to transform and
manipulate data to derive insights. This might involve calculations, aggregations,
filtering, and other data preparation tasks.
7. Performance Optimization: Dashboards should load quickly and respond smoothly.
BI analysts work to optimize the dashboard's performance, which may include
optimizing data queries and ensuring efficient data storage.
8. Data Security and Compliance: They are responsible for ensuring that sensitive or
confidential data is appropriately secured and that the dashboard complies with
relevant data privacy regulations.
9. Testing and Ǫuality Assurance: Before deploying a dashboard to end-users, BI
analysts should thoroughly test it to ensure accuracy and functionality. This includes
both data accuracy and visual correctness.
10. Documentation: Creating documentation for the dashboard is important for its
maintenance and support. This documentation may include data definitions, data
sources, and how to use the dashboard effectively.
11. User Training and Support: They may provide training to end-users on how to use
the dashboard and offer ongoing support to address user questions and issues.
12. Monitoring and Maintenance: Once a dashboard is live, BI analysts monitor its
performance, data quality, and user engagement. They also need to maintain and
update the dashboard as the organization's reporting needs evolve.
 13. Feedback and Iteration: They should actively seek feedback from users and
stakeholders to continuously improve and refine the dashboard. Regularly updating
and enhancing the dashboard is essential for its effectiveness.
14. Collaboration: Collaborate with other teams or departments to align dashboards
with business goals and ensure that the data presented is relevant and actionable.
15. Stakeholder Communication: BI analysts often act as intermediaries between
technical and non-technical stakeholders, translating business needs into technical
solutions and vice versa.

Overall, BI analysts play a critical role in bridging the gap between data and actionable
insights, and their dashboard design responsibilities are integral to helping organizations
make data-driven decisions.

Beneťts & Importance of Data Visualization:

Beneťts:
● Helps You See Trends and Patterns
● Helps You Make Better Decisions
● Helps You Communicate Your Findings More Effectively
● Data Visualization Helps Users See Outliers
● Data visualization is not only powerful; it is also easy to use and learn.

Importance of Data Visualization:

1. In Health Care:
Healthcare is an industry that relies heavily on data. From patient medical records to insurance
claims, there is a lot of data that needs to be collected, analyzed, and interpreted. Data
visualization plays an important role in healthcare by allowing doctors and other medical
professionals to make better-informed decisions.

For example, let's say a hospital wanted to reduce the number of patient readmissions. They
could use data visualization to look at readmission rates over time, identify which types of
patients are more likely to be readmitted, and develop interventions to target those high-risk
patients.

2. In Analytics
Analytics is the process of turning data into insights. Data visualization is a key part of analytics
because it allows analysts to take a huge dataset and distill it down into something that can be
easily understood and interpreted.

Without data visualization, analysts would be stuck looking at raw data all day long without
being able to identify any patterns or trends. But with data visualization, analysts can quickly
see relationships between variables and make better-informed decisions.

3. In Business Intelligence
Business intelligence (BI) is the process of transforming raw data into actionable insights. Data
visualization plays an important role in BI because it allows businesses to see their data in a
new light and make better-informed decisions about their overall strategy.

For example, let's say a company wanted to increase sales by 10% this year. They could use
data visualization to track sales over time, identify which products are selling well and which
ones are not, and develop marketing campaigns and targeted promotions accordingly.

4. In Data Science
Data science is all about extracting insights from large datasets. Data visualization plays an
important role in data science by allowing scientists to visualize their data and find patterns that
are otherwise impossible to discover.

For example, let's say a scientist was studying a disease and wanted to find out which genetic
factors were associated with it. They could use data visualization techniques like cluster
analysis or heat maps to try to find groups of genes that are similar to the disease group and
then further investigate those genes.

5. In Machine Learning
Machine learning is a subfield of artificial intelligence that deals with the design and
development of algorithms that can learn from data. Data visualizations are important in
machine learning because they can be used to understand complex datasets and identify
patterns. Machine learning algorithms can then be developed to automatically detect these
patterns.

For example: Google's PageSpeed Insights tool uses machine learning to automatically analyze
website performances and provides recommendations on how to improve them.

6. In Big Data Analytics

Big data analytics can be very overwhelming because of the sheer volume of data that is
involved. Data visualization can help us make sense of all this data by helping us identify
correlations and structures that we would not be able to see otherwise.

For example, if we are looking at a dataset with millions of rows, it would be very difficult to
find patterns without data visualization. But by utilizing a visualization tool, we can easily find
patterns in the data.

Types of Charts:
Power BI includes many different types of visualizations to help you analyze your data.
Bar Chart:
 ● Used to compare data across categories.
● Types: Clustered bar charts, stacked bar charts, and grouped bar charts.

Column and Bar Charts – The most common charts are bar and column charts which can also be
configured as clustered or stacked. Currently the bar chart is the only chart which displays data
horizontally by inverting the axes, so the x-axis shows the chart values and the y-axis shows the
category values.Bar Chart: Used to compare data across categories. Types: Clustered bar charts,
stacked bar charts, and grouped bar charts.

2. Line and Area Charts – The best way to display linear data would be through line charts. Line
Charts and Area Charts display points connected with a line. These charts are commonly used to
show periods over time. Line Chart: Shows data trends over a continuous period or interval.Useful
for displaying changes or patterns over time.
3. Combination Chart – The line chart and column chart can be combined. This type of chart is
useful if you wanted to display data on different axes. This can be quite useful when comparing
values of the same range.

4. Scatter and Bubble Charts – When you want to analyze a correlation between two variables a
scatter chart could be used. The chart displays the intersection of an x and y value. If you want to
analyze the correlation between three variables use a bubble chart. To create a bubble chart in
Power BI simply put the third variable in the Size field of a scatter chart.

5. Shape Chart – To display values of percentages a shape chart is commonly used. The size of the
shape is then used to display the proportion of the whole. Shape charts include Pie, Doughnut, and
Funnel charts.Pie Chart: Represents parts of a whole.Useful for showing the distribution of a single
data set.

6. Treemap and Waterfall Charts – To view hierarchical data use a Treemap. This visualization
should be used when you have a large amounts of hierarchical data that doesn‘t fit in columns or
bar charts.Usually Waterfall charts are used to show running totals when they are added or
subtracted over time or a total.

7. Table and Matrix – These visualizations are used to show text across a table. Both visualizations
can be formatted with conditional formatting, which increases the ease of use. Although the matrix
visualization can be used to pivot data by one or more columns.
8. Maps – Maps are obviously used for geospatial visualization. You can use maps to configure
categorical and quantitative information within locations.

9. Gauges and KPIs – Gauges and KPIs(Key Performance Indicators) are a quick way to display a
value compared to a target value. They can both be used to display a measure of progress or

distance very quickly.

fi0. Cards – Cards can be used for a single value or multi values depending on your preference. Most
of the time I prefer the single cards but if you had many sets of data to display, a Multi Row Card
visualization could be a benefit.

1fi. Slicgrs – These are best used when you want to filter down a report with data that is related. By
default, slicers affect all visualizations on the page that are related to each other. A simple trick to
make a visualization not filtered through a slicer is to Edit Interactions for that chart. Slicers can also
be viewed as a list of data or a dropdown.

What are dashboards?

Dashboards are business intelligence (BI) reporting tools that aggregate and display critical
metrics and key performance indicators (KPIs) in a single screen, enabling users to monitor and
examine business performance at a glance. Dashboards extract and communicate high-level
insights such as anomalies, issues and trends for end-users of all skill levels, before they choose
to analyze data in more detail using advanced mechanisms like contextual analytics.

As a high-level reporting mechanism, dashboards ultimately provide fast ‗big picture‘ answers
to critical business questions and assist and benefit decision-making in several ways:

● Communicating how a business is performing according to defined targets

● Improving informational awareness for everyone
 ● Organizing operational data into a well managed format
● Visualizing complex relationships in a easy-to-understand way

Types of dash board

Opgrational Dashboards:
Purpose: Monitor and manage day-to-day operations.
Users: Front-line employees, managers, and operational teams.
Content: Real-time data, key performance indicators (KPIs), and metrics.
Timeframe: Immediate or short-term.

Stratggic Dashboards:
Purpose: Support long-term planning and decision-making.
Users: Executives, strategic planners, and senior management.
Content: High-level KPIs, trends, and performance against strategic goals.
Timeframe: Medium to long-term.

Analytical Dashboards:
Purpose: Facilitate in-depth data analysis.
Users: Data analysts, business analysts, and researchers.
Content: Interactive data exploration tools, drill-down capabilities, and data manipulation
features.
Timeframe: Varies depending on the analysis needs.

Tactical Dashboards:
Purpose: Aid in achieving specific, short- to mid-term objectives.
Users: Middle managers, project managers, and teams with specific goals.
Content: KPIs and metrics related to the defined tactical objectives.
Timeframe: Short to mid-term.

Examplgs:
In the sales domain, you can create specific types of dashboards to cater to various needs and
objectives. Here are operational, strategic, analytical, and tactical dashboards in the sales
domain:

1. Operational Sales Dashboard:

● Purpose: To monitor and manage day-to-day sales operations in real-time.

● Users: Sales representatives, sales managers, and team leads.
● Content:
● Current sales performance, including sales revenue, number of deals closed, and
quotas.
● Real-time sales pipeline and lead tracking.
 ● Recent sales activities, such as calls, emails, and meetings.
● Timeframe: Immediate or short-term, focusing on current sales activities and targets.

2. Strategic Sales Dashboard:

● Purpose: To provide insights for long-term sales planning and decision-making.

● Users: Sales directors, executives, and strategic planners.
● Content:
● High-level KPIs and metrics related to sales goals and objectives.
● Historical sales performance trends and comparisons.
● Market analysis and competitor benchmarking.
● Timeframe: Medium to long-term, emphasizing the achievement of strategic sales
objectives.

3. Analytical Sales Dashboard:

● Purpose: To support in-depth data analysis and exploration of sales data.

● Users: Data analysts, sales analysts, and business intelligence teams.
● Content:
● Interactive data visualizations, charts, and graphs.
● Sales forecasts and predictive analytics.
● Drill-down capabilities for exploring specific sales trends and patterns.
● Timeframe: Varies depending on the analysis needs, ranging from short-term to
long-term.

4. Tactical Sales Dashboard:

● Purpose: To assist in achieving specific short- to mid-term sales objectives and goals.
● Users: Sales managers, team leaders, and teams responsible for specific sales initiatives.
● Content:
● KPIs and metrics related to tactical sales goals, such as lead generation and
conversion rates.
● Progress toward achieving short-term sales targets.
● Data on specific sales campaigns and promotions.
● Timeframe: Short to mid-term, focusing on the duration of specific sales initiatives and
objectives.

The choice of dashboard type depends on the specific goals and objectives of the sales team or

organization. These dashboards can help sales professionals and management make informed

decisions, track performance, and drive sales growth. Additionally, they can be customized to
include other key metrics and indicators that are relevant to your specific sales processes and

strategies.

In thg hgalthcarg domain, dashboards play a crucial role in improving patient care, optimizing
operations, and managing resources effectively. Here are operational, strategic, analytical, and
tactical dashboards tailored to the healthcare sector:

1. Operational Healthcare Dashboard:

● Purpose: To monitor and manage day-to-day healthcare operations in real-time.

● Users: Clinical staff, nurses, doctors, and hospital administrators.
● Content:
● Real-time patient data, such as bed availability, patient admissions, and
discharges.
● Current status of operating rooms, emergency departments, and ICU.
● Staffing levels and shift assignments.
● Timeframe: Immediate or short-term, focusing on current patient care and hospital
operations.

2. Strategic Healthcare Dashboard:

● Purpose: To support long-term healthcare planning and decision-making.

● Users: Hospital executives, healthcare administrators, and strategic planners.
● Content:
● High-level KPIs and metrics related to healthcare facility performance.
● Historical trends in patient admissions, readmissions, and patient outcomes.
● Resource allocation and capacity planning.
● Timeframe: Medium to long-term, emphasizing long-range planning and quality
improvement.

3. Analytical Healthcare Dashboard:

● Purpose: To facilitate in-depth data analysis and exploration of healthcare data.

● Users: Healthcare analysts, data scientists, and researchers.
● Content:
● Interactive visualizations of patient outcomes, disease trends, and clinical
pathways.
● Predictive analytics for resource allocation and patient risk assessment.
● Comparative data for benchmarking against other healthcare facilities.
● Timeframe: Varies depending on the analysis needs, from short-term to long-term
research.
4. Tactical Healthcare Dashboard:

● Purpose: To help healthcare managers and teams achieve specific short- to mid-term
healthcare objectives.
● Users: Department heads, care teams, and quality improvement teams.
● Content:
● KPIs related to specific healthcare initiatives, such as reducing patient wait times.
● Progress toward achieving quality improvement goals.
● Data on specific clinical programs, such as vaccination campaigns.
● Timeframe: Short to mid-term, focusing on the duration of specific healthcare initiatives
and goals.

Dashboard Design Best Practices

fi. Know Your Audience
The first critical step in designing your dashboard is to develop a user persona that defines who
your audience is, what information they need, and how they will use this information. Some
dashboards have broad distribution but most are tailored to specific people or roles such as
executives, managers or analysts. These roles have very different needs and expectations when
it comes to data.

For example, an analyst will want many data views and the ability to dig into the data to
explore more deeply. In contrast, a busy operations manager needs to know at a glance if there
are significant deviations from the norm that require immediate action.

2. Choose the Right Type of Dashboard Design

Once you have defined the audience and purpose for your dashboard, make sure to choose the
type of dashboard that aligns. There are four main types of business intelligence dashboards
based on their purpose: strategic, operational, tactical and analytical. Below we discuss the
main attributes and give dashboard examples of each type as inspiration.

Executive Dashboards: These dashboards allow executives to track key performance indicators
(KPIs) over time. The purpose here is to analyze bigger-picture trends over longer time periods
rather than focusing on immediate, short term actions.

The example below shows how hospital executives can use an executive dashboard to visualize
strategic KPIs relevant to management, patient experience, and staff dynamics.
 1. High-level KPIs give executives a big-picture view of critical success factors.
2. Predictions indicate whether key indicators are performing better or worse than
expected.
3. Time-based trends provide a view of overall activity, and can be shared with
higher-level management as needed

Operational Dashboards
The purpose of operational dashboards is to quickly tell the user what‘s happening in the
moment and to highlight when and where critical issues occur. Use this type of dashboard for
time-sensitive information and to highlight deviations in the data so that users can take action.

For example, a police force can use an operational dashboard to track the location and volume
of police reports and provide a snapshot of incidents teams should be prepared to deal with.

Heatmaps make it easy to quickly spot areas where activity levels are high.
1. Log tracking over time helps teams predict what time of day workloads will be
highest.
 2. Log types provide a quick snapshot of the kind of incidents teams should be prepared
to deal with.

Tactical Dashboards
Tactical dashboards help users track progress toward specific goals or time periods. Users
should be able to click on the dashboard to quickly and easily drill in to the data on the
dashboard itself to explore problems and opportunities.

In this example, a software company can use a tactical dashboard to make faster, better
informed decisions with information on browsing and purchase behavior, interactions across
digital channels, and more.

Key data points are enhanced by trending information; how much this number changed
from the prior month.
1. More detailed insights can be found by drilling down into additional data.
2. A day-to-day view allows teams to look more closely at activity and what caused it.

Analytical Dashboards

Analytical dashboards are typically created and used by analysts as an interactive tool to help
them better support business people in analyzing trends and identifying issues. They usually
have minimal graphic elements and instead strive to present as many data views as possible.
Here especially, the user needs to be able to drill down into additional data for more detailed
analysis.

The example below showcases how a CPG company can use an analytics dashboard to track
trends, forecast market share, and compare sales against competitors.
 Viewing market valuation over time helps guide business decisions.
1. Forecasts can be used alongside other data to drive business planning and resourcing.
2. Comparison data shows sales versus key competitors in an easy-to-read, side-by-side

3. Identify Your Key Metrics

Your audience may tell you they want everything and that every data point is important.
Your job in dashboard design is to help them choose the essential metrics to best help them
align behaviors, refine strategy, and measure success. For example, a marketing team that‘s
rewarded based on the number of people reached needs different metrics than one
rewarded for qualified leads. When in doubt, stick to the 80/20 rule. Don‘t try to account for
everything but instead focus on the 20% that‘s most strategic— and that can deliver 80%
of the value.

4. Tell a story with your data

Once you‘ve identified which KPIs matter most to your audience for a given dashboard,
organize how you present these KPIs in a way that guides your audience as if you were
telling a story. For example, if this quarter‘s customer acquisition cost is a critical KPI, your
dashboard design would emphasize that KPI and then, in smaller font, present data from
prior quarters and breakouts for different acquisition channels. Ideally, you can also provide
a summary and/or key takeaways from the data. Get inspired by the ten best modern data
visualization examples.

5. Choose the Right Types of Charts

It‘s important to apply the right type of charts to represent your data. You should start by
keeping in mind what information your audience needs to see. Here are the four primary chart
types and recommendations on when to use them as you design your dashboard:
Comparison Charts
Use these simple visualizations to help users compare values over time, recognize trends, and
identify high and low values. Common examples are bar charts, line charts and circular area
charts. Time-based charts should represent time periods on the X axis and bar and line charts
should ideally be limited to 7 data values.

Composition Charts
This type of data presentation helps users see parts of a whole, either over time or for a static
period. Stacked area charts and stacked bar charts represent the changing relationship between
data points over time. Pie charts show the relative difference between parts of a whole. Limiting
how many data components you apply will make composition charts much easier to read.

Distribution Charts
This visualization type allows users to more easily see the shape or tendency of their data and
quickly spot outliers and commonalities. Bar histograms, line histograms and scatter plots are
common examples of distribution charts.

Relationship Charts
These visualizations help users see correlations and relationships in data. Scatter plots are
effective for analyzing distribution and scatter plots with different bubble sizes adds a third data
dimension

6. Apply Essential Dashboard UI Design Principles

Well-designed dashboards adhere to user interface (UI) design principles that guide an
audience toward the information they need. The four principles below will help you transform
messy data into well-informed decisions.
Organize the Information
When designing your dashboards, use classic UI design and information hierarchy
principles to add clarity and guide your audience through the data. People usually scan
horizontally across a page, and then down, so put the most important information at the
top-left.
Keep it Simple
Busy business people need information and insights at a glance. Limit the number of charts
and data points on the dashboard to avoid clutter and draw attention to key metrics by
using color, font size and placement
Provide Visual Cues
When designing for modern, interactive dashboards, use icons, buttons or text to indicate
when an element links to another page or provides contextual information. Add clear calls
to action if you want users to do something.
Get Color Right
First, make sure your dashboard design is consistent with your organization‘s brand
identity (colors, fonts, and graphics). Second, limit your use of color in general to keep
your dashboard‘s appearance open and light. Third, balance your use of color with the use
of shapes and contrast.

7. Iterate and evolve

Before you roll out your dashboards to a large group, share them with a few users from the
audience, listen to their feedback and make any revisions. Once you‘ve rolled it out to the
broader team, again ask for input and suggestions. Ideally, you could watch people actually
using your dashboard. Plus, the KPIs most important to your audience now may change next
month as the market changes, company strategy shifts, or new initiatives are launched.
Pitfalls in dashboard design: Designing effective dashboards is crucial for conveying
information and insights in a clear and user-friendly manner. However, there are common
pitfalls in dashboard design that can hinder their effectiveness. Here are some of the most
common pitfalls:

1. Information Overload: Including too much information or too many data points on a single
dashboard can overwhelm users. It's essential to focus on key metrics and data that are
relevant to the dashboard's purpose.
2. Lack of Clarity: Cluttered and confusing layouts make it challenging for users to interpret
data. Dashboards should have a clean and intuitive design, with clear labels and headings.
3. Inadequate Data Visualization: Choosing the wrong type of chart or graph for displaying
data can lead to misinterpretation. Ensure that the data visualization chosen is appropriate
for the data and the message you want to convey.
4. Not Considering User Needs: Failing to understand the specific needs and preferences of
the dashboard's users can result in a design that doesn't effectively address their
requirements. Engage with potential users to gather input and feedback during the design
process.
5. Poor Color Choices: Overusing or misusing colors can make a dashboard visually
overwhelming. Use colors sparingly and with purpose, such as to highlight important data
points or trends.
6. Neglecting Mobile Responsiveness: In today's mobile-driven world, dashboards should be
designed with responsiveness in mind. Ensure that the dashboard is accessible and
functional on various devices and screen sizes.
7. Inconsistent Design Elements: Inconsistent fonts, colors, and layout elements can create a
disjointed and unprofessional appearance. Maintain a cohesive design throughout the
dashboard.
8. Lack of Context: Data without context is less meaningful. Provide clear titles, descriptions,
and explanations for the data and metrics presented on the dashboard.
9. Ignoring User Feedback: After the initial design and deployment, failing to incorporate user
feedback for improvements can lead to an ineffective dashboard. Continuous user feedback
and iterative design are essential for dashboard success.
10. Security and Privacy Concerns: Failing to address security and privacy concerns when
displaying sensitive data can lead to data breaches and compliance issues. Ensure that the
dashboard design considers data security and access control.
11. Failure to Prioritize Data Ǫuality: Inaccurate or outdated data can undermine the usefulness
of a dashboard. Regularly monitor data sources and maintain data quality.
12. Overlooking Performance: Dashboards that load slowly or are unresponsive can frustrate
users. Optimize the performance of the dashboard to ensure it operates efficiently,
especially when handling large datasets.
13. Not Providing User Training: Assuming that users will intuitively understand how to use the
dashboard can lead to frustration and underutilization. Provide training or documentation to
help users effectively navigate and interact with the dashboard.