WEEK 1

INTRODUCTION
TO DATA ANALYSIS
AND POWER BI

05 May 2025
[email protected]
OUTLINE

• Introduction
• Discover Data Analysis
• Get started building with Power BI
• Get data in Power BI
INTRODUCTION

Email: [email protected]
Please fill in the form: https://forms.office.com/r/gpk2FZU4nV

-Module guide
-Prescribed material:
https://learn.microsoft.com/en-us/certifications/power-bi-data-analyst-associate/

-myLMS
DISCOVER DATA ANALYSIS
What is the role of a data analyst?
Your role is to turn massive amounts of data into insights that can
drive action and change

Why is it important?
Businesses struggle to understand and use their data effectively.
Having data isn't enough—action must follow insights. Data must be
crafted into a clear and accurate story through reports so business
leaders can make informed decisions quickly. Data analysts are vital
for uncovering insights, evaluating performance, and driving
strategic decisions.
DISCOVER DATA ANALYSIS
• Data is everywhere - Organizations collect various types of data think about loyalty cards
Retailers use data to track customer behavior, sales patterns, and operations like:
• Inventory tracking
• Detecting buying habits and trends
• Recommending products
• Pricing optimization
• Analysis involves more than just reporting - You must clean, model, and transform
data before it's ready to be turned into insights or visual stories.
• Collaboration is essential - Work closely with data engineers and scientists to access and
prepare the right data. Make sure that your insights are shared with the right people in your
organization and are easy to find and understand.
• Trusted data = better decisions - Leaders need accurate, trusted data to make smart,
impactful decisions.
 DISCOVER DATA ANALYSIS
Clean Model Visualise

Imagine you have a big Once your data is clean, After you've cleaned and
messy pile of information. modeling is like building a modeled your data,
Cleaning is like tidying it up. simplified version or a "map" visualization is like turning
You throw out the obviously of it to understand how that information into
wrong stuff (like typos or things are connected or to pictures, charts, or graphs.
impossible numbers), fix predict what might happen. This makes it much easier
inconsistencies (like different You use different tools and for people to see trends,
ways of writing the same techniques (like math or patterns, and insights that
address), and fill in any logic) to find patterns and would be hard to spot in just
missing pieces so the whole relationships within the data. a table of numbers. Think of
pile is neat and ready to be It's like creating a recipe it as presenting your cooked
used. Think of it as making from your ingredients – dish in an appealing way so
sure your ingredients are all you're combining them in a everyone can easily see and
fresh and properly prepared specific way to get a desired understand what it is.
before you start cooking. outcome or understanding.
OVERVIEW OF DATA ANALYSIS

Data Analysis: More Than Just Processing

• Data analysis is crucial for business.
• It goes beyond cleaning, modeling, and
visualizing data.
• It's essential for informed decision-making.
OVERVIEW OF DATA ANALYSIS

Five Core Types of Analytics:

1. Descriptive
2. Diagnostic
3. Predictive
4. Prescriptive
5. Cognitive
DESCRIPTIVE:
WHAT
HAPPENED?
• Looks at what has happened in
the past
• It’s like looking in the rearview
mirror. You're just seeing what
already happened, without
asking why.
• Employs Key Performance
Indicators (KPIs) like ROI.
• Uses historical data.
• Example: A store reviews last
month's sales report and sees
that they sold 500 shirts.
DIAGNOSTIC:
WHY DID IT HAPPEN?
• Tries to understand why something
happened.
• It’s like being a detective—you're not
just looking at what happened, but
digging into the reasons behind it.
• Collects related data.
• Uses statistical techniques to explain
anomalies
• Example: After seeing that shirt
sales dropped, the store finds out it
was because a competitor had a big
discount sale.
PREDICTIVE ANALYTICS:
FORECASTING THE
FUTURE • Looks at the past to predict what might
happen next.
• It’s like using a crystal ball—based on
past trends, it guesses what will probably
happen in the future.
• Employs statistical and machine learning
techniques.
• Example: The store notices that shirt
sales always go up before Christmas, so
they expect higher sales next December.
PRESCRIPTIVE
ANALYTICS:
GUIDING ACTION
• Suggests what actions should
be taken for the best results.
• It’s like getting expert advice—
after predicting the future, it
tells you the best way to
respond.
• Relies on machine learning.
• Example: Based on predictions
of high Christmas sales, the
store gets advice to stock more
shirts and offer a small
discount.
 COGNITIVE ANALYTICS:
LEARNING AND ADAPTING

• Uses AI, machine learning, and

natural language processing to mimic
human thinking and make smart
decisions.
• It’s like having a smart assistant who
learns, understands, and helps make
decisions, just like a human would.
• Draws inferences from data and patterns.
• Creates a self-learning knowledge base.
• Example: An AI assistant at the store
chats with customers, learns their
preferences, and recommends clothes
they’ll probably like.
OVERVIEW OF DATA ANALYSIS

• Trustworthy data is fundamental: The data analysis process relies on

capturing data from reliable sources and transforming it into understandable
information for decision-making.
• Data analysis empowers data-driven decisions: It enables businesses to
fully comprehend their data, leading to more confident choices.
• The demand for data analysts is increasing: As data volumes grow, the
need for professionals who can organize, interpret, and extract meaningful
insights from data becomes more critical.
ROLES IN DATA
• Data storytelling is a collaborative journey: It doesn't typically begin with the
individual telling the story; the data originates elsewhere.
• Data acquisition and preparation are significant undertakings: Getting data
into a usable state often requires substantial effort and is usually outside the
scope of a single individual, especially in larger organizations.
• Modern data landscapes are complex: Applications and projects are intricate
and necessitate diverse skills and knowledge from multiple individuals working
together.
• Traditional data roles have evolved: The growth in data volume and variety has
led to the specialization of roles like business analysts and business intelligence
developers into more focused areas like data engineering and analysis.
ROLES IN DATA

• Distinct data roles exist with

specific responsibilities
BUSINESS ANALYSTS –
HELPS BUSINESSES SOLVE PROBLEMS AND IMPROVE
PROCESSES BY UNDERSTANDING WHAT THE BUSINESS
NEEDS.

Closer to the business, a

Focus on interpreting specialist in interpreting data
visualized data: They bridge that comes from visualization.
the gap between data and They differentiate from data
business understanding. analysts by their focus on what
to do with the visualized data.
DATA ANALYST –
LOOKS AT DATA (NUMBERS, FACTS, ETC.) TO FIND

USEFUL INFORMATION, LIKE TRENDS OR PATTERNS .

• Focus on data visualization, reporting, and modeling:
They transform raw data into meaningful insights
using tools like Power BI and manage related assets.
• Enables businesses to maximize data value through
visualization and reporting tools like Power BI.
Responsible for profiling, cleaning, and transforming
data; designing and building scalable semantic
models; and implementing advanced analytics in
reports. They identify data and reporting
requirements, turn raw data into insights, and
manage Power BI assets (reports, dashboards,
workspaces, models) including security. They work
with data engineers to locate data sources and
improve data collection processes, and with database
administrators for data access.
DATA ENGINEER –
BUILDS AND MAINTAINS SYSTEMS THAT COLLECT, STORE, AND ORGANIZE
DATA SO OTHERS (LIKE ANALYSTS OR SCIENTISTS) CAN USE IT EASILY.

Focus on building and managing data infrastructure: They provision data platforms, ensure
data flow, and integrate data sources.

Provisions and sets up data platform technologies (on-premises and cloud), managing and
securing the flow of structured and unstructured data from multiple sources (relational,
nonrelational databases, data streams, file stores). They ensure secure and seamless
integration across data platforms, ingest, egress, and transform data, collaborate with
stakeholders on data requirements, and design/implement solutions. Their work goes beyond
database management and focuses on data wrangling to enable business intelligence and
data science projects. They support data analysts in accessing data and optimizing semantic
models.
DATA SCIENTIST –
USES ADVANCED MATH, STATISTICS, AND PROGRAMMING TO BUILD MODELS THAT CAN PREDICT OR
AUTOMATE DECISIONS.

Performs advanced analytics to extract value

from data, ranging from descriptive
(exploratory data analysis) to predictive
(machine learning for anomaly/pattern
Focus on advanced analytics and extracting detection and forecasting). Some work in
value from data: Their work ranges from deep learning with customized algorithms. A
descriptive to predictive analytics and often significant portion of their work involves data
involves data wrangling and feature wrangling and feature engineering, which
engineering. can be accelerated by data engineers. They
determine questions needing answers,
devise hypotheses/experiments, and often
rely on data analysts for visualization and
reporting.
DATABASE ADMINISTRATORS-
MAKES SURE THE DATABASES (WHERE
DATA IS STORED) ARE SAFE, ORGANIZED,
AND WORK PROPERLY.
• Focus on the operational aspects of data platforms:
They ensure availability, performance, security, and
backup/recovery of database solutions.
• Implements and manages the operational aspects of
cloud-native and hybrid data platform solutions 1
(Azure data services, SQL Server). Responsible for
overall availability, consistent performance, and
optimizations of database solutions. They work with
stakeholders to implement policies, tools, and
processes for data backup and recovery. They
monitor and manage database health and hardware,
and manage data security by controlling user access
and privileges. Their role differs from data engineers
who focus on data wrangling.
ROLES IN DATA

• Collaboration is crucial between data roles: Data analysts rely on data

engineers for data access and optimized models, while data scientists benefit
from data engineers' data wrangling skills.
• Role transitions are possible: Individuals in roles like database administration
and business intelligence can move into data engineering by acquiring new
skills. One person can also be responsible for multiple roles such as being a
business analyst as well as a data analyst.
ROLES IN DATA

Role Focus Simple Example

Business needs & Suggesting ways to boost
Business Analyst
improvements sales
Data Analyst Studying data Finding best-selling products
Collecting and storing sales
Data Engineer Building data systems
data
Data Scientist Advanced data modeling Predicting customer behavior
Keeping databases safe and
Database Administrator Managing databases
fast
TASKS OF A DATA
ANALYST
• A data analyst plays a critical role in helping
organizations make informed decisions by
uncovering insights from data.
• The work of a data analyst spans five core
areas: Prepare, Model, Visualize, Analyze,
and Manage.
• These tasks ensure that data is accurate,
usable, insightful, and securely distributed.
PREPARE

• “Garbage in, garbage out”—Good analysis starts with good

data.
• Data Preparation includes:
• Profiling, cleaning, and transforming raw data.
• Ensuring data integrity: correcting inaccuracies, filling gaps, formatting consistency.
• Understanding data sources and their performance implications.
• Addressing privacy and security concerns, such as anonymizing personal data.
• 📝 Key Insight: Most time is spent here. Without proper preparation, even the
best models and visualizations will fail.
MODEL

• “Structure your data for performance and understanding.”

• Data Modeling defines how tables relate, and adds logic (metrics, calculations).
• A good semantic model:
• Boosts report accuracy and performance.
• Enables faster and clearer insights.
• Simplifies future updates.
• 💡 Example: In Power BI, modeling influences report load time and
responsiveness.
VISUALIZE

• “Tell the story—make insights visible and understandable.”

• Effective visualization:
• Solves business problems.
• Uses appropriate visuals to build a narrative.
• Reduces noise by focusing only on relevant data.
• Supports accessibility through inclusive design (color, font, interactivity).
• 🛠 Tools: Power BI’s AI visuals and features (e.g., Q&A, Quick Insights) enhance
storytelling without coding.
 ANALYSE

• “Interpret the data to uncover insights and trends.”

• Use Power BI's analytical tools to:
• Identify patterns and anomalies.
• Forecast trends and predict outcomes.
• Communicate insights in a user-friendly way.
• AI integrations (e.g., Azure ML) deepen analysis without needing a data
scientist.
MANAGE

• “Deliver the right insights, to the right people, securely.”

• Involves managing Power BI assets like dashboards, reports, and workspaces.
• Ensure:
• Proper access control.
• Sharing and collaboration.
• Reduced data silos through shared models.
• Certified datasets to build trust in insights.
• The management of Power BI assets helps reduce the duplication of efforts and
helps ensure security of the data.
TASKS OF A DATA ANALYST
Task Description Example
Remove duplicate entries,
Clean, transform, and
fill missing values, and
1. Prepare organize raw data for
standardize date formats in
analysis.
Excel data.
Connect sales and customer
Define relationships
tables in Power BI and
2. Model between data tables and
calculate revenue per
create calculations/metrics.
customer segment.
Build a bar chart showing
Create charts and
monthly sales trends, with
3. Visualize dashboards to tell data
filters for regions and
stories.
product types.
Identify that sales dipped in
Interpret data to find
Q2 due to a supply chain
4. Analyze patterns, trends, and
delay, using trend lines in a
actionable insights.
report.
Share a dashboard via a
Oversee report distribution,
Power BI app, control user
5. Manage security, and data asset
access, and certify the
maintenance.
GET STARTED BUILDING WITH POWER BI

Microsoft Power BI is a complete reporting solution that offers data preparation,

data visualization, distribution, and management through development tools and
an online platform.
Power BI can scale from simple reports using a single data source to reports
requiring complex data modeling and consistent themes. Use Power BI to create
visually stunning, interactive reports to serve as the analytics and decision
engine behind group projects, divisions, or entire organizations.
Power BI is an essential tool to data analysts and their organization; however, all
data professionals' benefit from understanding how Power BI works to explore
and present data insights within organizations.
 THREE Power BI Desktop
(desktop application)
PRIMARY
COMPONENTS
OF POWER BI Power BI service (online
platform)

Power BI Mobile (cross-

platform mobile app)
POWER BI DESKTOP

• A free desktop application serving as the primary development tool for data
analysts and report creators.
• Used for:
• Connecting to various data sources.
• Transforming and modeling data to prepare it for analysis.
• Creating visualizations and interactive reports.
• Available for free download (Windows Store or online).
POWER BI SERVICE

• An online platform accessed via a web browser (app.powerbi.com) using a

school or work account (or a free trial/developer account).
• Functions as a hub for:
• Organizing and managing created reports and other Power BI items.
• Distributing reports to end-users.
• Creating high-level dashboards that provide an overview and link to detailed reports.
• Developing apps to bundle related reports for simplified sharing.
• Easily group related reports to users in a simple format
POWER BI MOBILE

• A cross-platform mobile application designed for consuming and viewing reports

on mobile devices.
• Reports are displayed in a mobile-optimized format, which can be configured in
Power BI Desktop.
THE COMMON POWER BI WORKFLOW

1. Connect to data with Power BI Desktop.

2. Transform and model data with Power BI Desktop.
3. Create visualizations and reports with Power BI Desktop.
4. Publish report to Power BI service.
5. Distribute and manage reports in the Power BI service.
BUILDING
BLOCKS OF
POWER BI

1. Semantic models
2. Visualizations
CREATE A SEMANTIC MODEL

• Definition: A semantic model in Power BI encompasses all connected data, the

transformations applied to it, the relationships established between different
datasets, and any calculations performed. It serves as the foundation for
building reports.
• Steps to Create a Semantic Model:
1. Connect to Data: Bring in data from various sources as needed.
2. Transform Data: Clean, shape, and prepare the data according to your
requirements.
3. Create Relationships and Calculations: Define how different tables relate to
each other and add calculated columns or measures to enrich the model.
POWER BI DESKTOP –
VISUALIZATIONS AND REPORTS
• Visualizations (visuals) are added to the report canvas by dragging and
dropping data fields.
• Power BI often suggests a visual type, which can be easily changed.
• Reports are built by combining multiple visuals on one or more pages.
• Best Practice: Keep report pages simple with related data for easy
understanding.
• Power BI's low-code nature allows for intuitive "drag and drop" functionality.
• Visuals are interactive; selecting data points in one visual can dynamically
update others.
• Reports can include drill-through capabilities and filters for deeper exploration.
• Once complete, reports are published to the Power BI service.
POWER BI SERVICE –
DASHBOARDS
• Dashboards are created in the Power BI service after a report is published.
• They are single-page layouts composed of tiles.
• Tiles are created by "pinning" visuals from existing reports.
• Dashboard tiles are not interactive; clicking a tile redirects the user to the
underlying report.
• Purpose: Dashboards provide a high-level overview of key information, acting
like a vehicle dashboard.
• Users can then navigate to reports for detailed analysis.
THESE ARE FUNDAMENTAL TO HOW YOU
OPERATE WITHIN THE POWER BI SERVICE.
WHEN YOU PUBLISH ANY REPORT YOU CREATE,
YOU MUST CHOOSE A WORKSPACE TO HOUSE
IT.

BY DEFAULT, EACH OF YOU HAS ACCESS TO

SOMETHING CALLED MY WORKSPACE. THIS IS
YOUR PERSONAL SANDBOX – PERFECT FOR
TESTING THINGS OUT AND EXPERIMENTING.
HOWEVER, AND THIS IS IMPORTANT, WHEN
YOU WANT TO SHARE YOUR BRILLIANT WORK
WITH OTHERS, YOU SHOULD ALWAYS CREATE
AND USE A SHARED WORKSPACE.

THINK OF SHARED WORKSPACES AS

COLLABORATIVE FOLDERS WHERE MULTIPLE
PEOPLE CAN ACCESS AND WORK WITH
REPORTS AND DASHBOARDS.
NOW, WHAT IF YOU HAVEN'T ACTUALLY BUILT
A REPORT YET? NO PROBLEM AT ALL! POWER
BI PROVIDES SEVERAL SAMPLE REPORTS
THAT YOU CAN PLAY AROUND WITH. THESE
ARE A FANTASTIC WAY TO GET A FEEL FOR
HOW REPORTS ARE STRUCTURED AND
WHAT'S POSSIBLE. WHEN YOU ACCESS
THESE SAMPLES, THEY'LL LOAD INTO YOUR
MY WORKSPACE, SO YOU CAN EXPLORE
THEM PRIVATELY WITHOUT AFFECTING
ANYONE ELSE. YOU CAN FIND THESE SAMPLE
REPORTS IN THE LEARN SECTION OF THE
NAVIGATION PANE WITHIN THE POWER BI
SERVICE. I HIGHLY ENCOURAGE YOU TO TAKE
A LOOK AT THESE – IT'S A GREAT WAY TO
LEARN BY EXAMPLE.
OKAY, YOU'VE BUILT A REPORT IN A SHARED
WORKSPACE AND NOW YOU WANT TO SHARE IT WITH
A WIDER AUDIENCE. THIS IS WHERE APPS COME IN.
WITHIN A WORKSPACE, YOU CAN CREATE AN APP,
WHICH ESSENTIALLY PROVIDES A SIMPLIFIED AND
FOCUSED INTERFACE FOR THE PEOPLE WHO WILL BE
CONSUMING YOUR REPORTS AND DASHBOARDS.
WHEN YOU CONFIGURE AN APP, YOU GET TO DECIDE
EXACTLY WHAT CONTENT FROM THE CURRENT
WORKSPACE TO INCLUDE. YOU ALSO CHOOSE WHO
YOUR AUDIENCE WILL BE. THIS IS A REALLY
POWERFUL FEATURE FOR CONTROLLING WHAT
INFORMATION DIFFERENT GROUPS WITHIN YOUR
ORGANIZATION SEE. HERE'S A CRUCIAL POINT: ONCE
YOU CREATE AN APP, YOU NEED TO UPDATE THE APP
AFTER EVERY CHANGE YOU MAKE TO THE ITEMS
WITHIN THE WORKSPACE THAT ARE INCLUDED IN THE
APP. THIS MIGHT SOUND LIKE AN EXTRA STEP, BUT
IT'S ACTUALLY A GREAT WAY TO MAINTAIN CONTROL
OVER THE VERSION OF THE CONTENT YOUR
AUDIENCE SEES. YOU MIGHT MAKE SEVERAL TWEAKS
AND IMPROVEMENTS TO A REPORT, BUT YOU ONLY
WANT YOUR USERS TO SEE THE POLISHED, FINAL
VERSION. UPDATING THE APP ALLOWS YOU TO DO
TOUR AND USE THE POWER BI SERVICE

• Apps are generally the best way to share content within an organization. While
you could grant users direct access to the workspace itself, this might give
them access to more content than you intend. Similarly, sharing individual
reports or dashboards can become cumbersome if you're making frequent
changes that you don't want consumers to see immediately. Apps provide a
much more streamlined and controlled sharing experience.
• Think of these as pre-packaged
apps that are designed for specific
data sources or services. Instead
of building an app from scratch,
you can find an existing template
app that suits your needs and then
simply connect it to your data.
Template apps can be a real time-
saver and a fantastic way to
quickly share insights with minimal
effort. For example, if your
organization uses a tool like
GitHub, there might be a GitHub
template app available. You can
connect it to your organization's
GitHub data, and suddenly you
have pre-built reports and
dashboards showing key metrics
like top contributors and pull
requests – without having to build
any of that yourself!
• To access template apps,
look for the Apps icon in
the left navigation pane,
then click Get apps, and
finally select Template
apps.
• Data is rarely static, so Power BI
provides ways to refresh the
underlying data that your reports
and dashboards are built upon.
In the Power BI service, you can
configure scheduled refreshes of
your semantic models (which are
essentially your datasets). This
means you can set up a
schedule for Power BI to
automatically connect to your
data source and pull in the latest
information – daily, weekly, or at
other intervals. In addition to
scheduled refreshes, you also
have the option to perform on-
demand refreshes if you need to
update the data immediately.
GET DATA IN POWER BI
• Tutorial