How organisations turn data into insight and better decisions, and what it takes to do that well and ethically.
DescriptiveWhat happened?
PredictiveWhat will happen?
PrescriptiveWhat should we do?
Business Intelligence & Data AnalyticsSchool of Business Management, UUM
Dr. Khairol Anuar IshakSession A261 · Sep Sem 2026/2027
Navigate→Space next · M menu · F full screen
Roadmap
Today's plan: 70 minutes
Six short sections. Each ends with something for you to do: sort, classify, vote or answer.
00–07
Warm-up
Our class, described with data.
7 min
07–17
1.1 Decisions
Three levels, five steps.
10 min
17–22
1.2 Analytics defined
Definition + quick check.
5 min
22–40
1.3 D · P · P
Descriptive, predictive, prescriptive + game.
18 min
40–50
1.4 Big data
4 Vs, MapReduce, cloud, AI.
10 min
50–65
1.5–1.6 Practice & ethics
Industries, law, ethics dilemma.
15 min
65–70
Exit ticket
5 questions.
5 min
By the end of this lecture
You will be able to…
1
LO 1-1 · Classify decisions
Tell strategic, tactical and operational decisions apart.
2
LO 1-2 · Walk the decision process
Describe the five steps from problem to choice.
3
LO 1-3 · Name the analytics type
Spot examples of descriptive, predictive and prescriptive analytics.
4
LO 1-4 · Apply it to business
Describe how analytics supports decisions in finance, HR, marketing, health care, supply chains and more.
Chapter map: 1.1 Decision making · 1.2 Business analytics defined · 1.3 Types of analytics · 1.4 Big data, cloud & AI · 1.5 Analytics in practice · 1.6 Legal & ethical issues
Warm-up · descriptive analytics in action
Who are we? You answered, the data describes
70
Students who responded
7.72 /10
Mean AI-readiness score
1.29
Standard deviation (spread)
2
International students
Gender mix
49 F21 M
Self-reported in the enrolment survey.
AI-readiness: mean and range
02.557.510
Lowest 4.75 · highest 10.0
This slide describes what has already happened: counts, averages, spread and a chart. That is descriptive analytics, the first of the three types you'll meet today.
Introduction
Why analytics has taken off
Three developments together pushed analytics into almost every business function:
🗄️
1 · More data
Organisations now track and store huge volumes of data: transactions, sensors, apps, social media and more.
🧠
2 · Better methods
New methods for pulling knowledge out of data: statistics, data mining, machine learning and optimisation.
⚡
3 · Computing power
Cheap, fast processing and cloud platforms make heavy analysis possible on demand.
Think: which of these three do you use every day without noticing? (Hint: every tap in Shopee, Grab or TikTok is data.)
1.1 · Decision making
Managers make decisions at three levels
Click a level of the pyramid.
Your turn · LO 1-1
Strategic, tactical or operational?
1.1 · Decision making · LO 1-2
Five steps from problem to decision
Case: a Kedah coffee chain is deciding where to open its next outlet. Click each step.
Common ways people decide
Tradition“We have always done it this way.”
IntuitionGut feeling and experience.
Rules of thumbSimple shortcuts, e.g. “near a school = busy”.
📊 Relevant dataFacts and analysis: the focus of this course.
1.2 · Definition
What is business analytics?
Business analytics is the scientific process of transforming data into insight for making better decisions.
Decisions based on data and facts are usually seen as more objective than tradition, intuition or rules of thumb. Analytics tools help us to:
💡Create insight from raw data
📈Forecast more accurately, so plans are better
⚖️Quantify risk and uncertainty
🎯Find better alternatives through analysis and optimisation
Quick check · 3 questions
Quick check #1: decisions & definition
1.3 · A categorisation of analytical methods · LO 1-3
Three types of analytics: one question each
Descriptive
What happened?
Queries & reports
Descriptive statistics
Charts & dashboards
Clustering, association rules
🛒 “Sales at the Sintok outlet fell 12% last month.”
Predictive
What will happen?
Regression & time series
Supervised learning
Simulation
🔮 “Next month we'll sell about 4,200 cups.”
Prescriptive
What should we do?
Rule-based models
Optimisation
Simulation optimisation
Decision analysis
✅ “Order 38 kg of beans and add one barista on Fridays.”
Hindsight→Foresight→Action· more complexity, more value
1.3 · Type 1 of 3
Descriptive analytics: what happened?
Techniques that describe the past. It's where almost every analysis starts, and it's what Chapter 2 (Excel) and your dashboards do.
Unsupervised learning finds patterns without a target to predict: • Cluster analysis: groups similar customers (similarity) • Association rules: “people who buy X also buy Y” (correlation)
Data queryAsk a database for records with certain characteristics, e.g. all orders above RM100 in September.
ReportThe output of a query, often with statistics and charts.
RuleIf forecast > stock, reorder the gap + 10% buffer
→
DecisionReorder 330 cups' worth
Current stock covers 500 cups. Move the slider and watch the recommendation change.
Your turn · LO 1-3
Descriptive, predictive or prescriptive?
1.3 · Coverage of business analytics
The road ahead: where each topic fits
The textbook (Camm et al., Business Analytics, 5e) covers all three types. Filter to see which chapters build which skill.
Chapter 2 of our course (Excel formulas) sits firmly in descriptive territory: the foundation for everything after.
1.4 · Big data
Big data: the four Vs
Big data is any data set too large or complex for standard processing and typical desktop software. IBM describes it with four Vs. Click a card to flip it.
V
Volume
How much?
Volume
The sheer amount of data generated and stored.
🛒 Every transaction at every outlet, every day, for years.
V
Velocity
How fast?
Velocity
The speed at which new data arrives and must be processed.
🚗 Grab tracks thousands of drivers' GPS positions every few seconds.
V
Variety
How many kinds?
Variety
Many formats: tables, text, images, audio, video, sensor logs.
💬 Customer reviews, photos and ratings in one app.
V
Veracity
How trustworthy?
Veracity
The quality and reliability of the data: errors, bias, missing values.
🤔 Fake reviews, typos, or a form where someone entered their age as 200.
1.4 · Big data technologies
Hadoop & MapReduce: divide and conquer
Hadoop is open-source software that stores and processes big data across many computers. MapReduce is its programming model: map splits the work, reduce combines the results. Try it by counting words in customer reviews:
Input (3 computers)
💻1 “great coffee great price”
💻2 “slow service great coffee”
💻3 “great price slow wifi”
→
① Map
→
② Shuffle (group)
→
③ Reduce
Each computer works on its own piece at the same time.
1.4 · The cloud
Cloud computing & data security
☁️ Cloud computing
Using data and software on servers outside the organisation, over the internet. Think Google Drive, Microsoft 365, AWS.
Makes storing and processing massive data feasible and cost-effective
Pay for what you use; scale up in minutes
You already use it: your UUM Google and Microsoft accounts
🔒 Data security
Protecting stored data from destructive forces and unauthorised users, such as hackers.
Confidential data in the cloud must be protected
Access control, encryption, backups
A breach damages customers, trust and the firm's reputation
Discuss (1 min): Would you store your company's customer list in a free cloud app? What would you check first?
1.4 · Artificial intelligence
Artificial intelligence & the data scientist
AI uses big data and computers to make decisions that used to need human intelligence.
Facial recognition at security checkpoints
Self-driving vehicles
Generative AI assistants like Gemini, ChatGPT, Copilot and Claude, which most of you already use
Big data created strong demand for data scientists: analysts trained in computer science and statistics who can process and analyse massive data. Qualified people are in short supply, which is an opportunity for you.
Computer science
Statistics
Business knowledge
Data scientist
Business-savvy analysts, the BBA advantage, sit where all three overlap.
Quick check · 3 questions
Quick check #2: types & big data
1.5 · Business analytics in practice · LO 1-4
Analytics across the organisation
1.6 · Ethics
Ethics: data is about people
Customers trade their data for benefits (discounts, convenience). They should understand that trade-off.
The terms are set in an agreement between customer and company, and the company must honour it.
Firms must protect data from data breaches, meaning unauthorised use of data. This is a major concern for every company.
Models built on data must be used fairly and responsibly.
🤔 Dilemma: you decide
You run analytics for a café loyalty app. Marketing wants to sell members' purchase histories to an insurance company that will use it to price health policies. It's legal under the sign-up terms, which few people read.
1.6 · Legal issues
The law: consent, purpose and control
🇪🇺 GDPR (European Union, since May 2018)
One of the strictest privacy laws in the world. It requires that:
requests for consent are easy to understand and access
the intended use of data is stated, and consent is easy to withdraw
individuals can get a copy of their data and demand it be erased
🇲🇾 PDPA 2010 (Malaysia)
Malaysia's Personal Data Protection Act governs how businesses handle personal data in commercial transactions.
Amended in 2024, with new duties phased in during 2025
Includes data-breach notification, appointing a Data Protection Officer, and a right to data portability
Your responsibility: analytics professionals must understand the laws on collecting, storing and using personal data wherever they work.
1.6 · Professional ethics
INFORMS ethics guidelines: how analysts should behave
Click a word to see what it means in practice.
Source: INFORMS (Institute for Operations Research and the Management Sciences) Ethics Guidelines, as summarised in Camm et al., Business Analytics, 5e.