Digital Thinking and Innovation – Innovation through Analytics

Introduction

• Businesses today collect massive volumes of data; making sense of it requires systematic analytics.
• Lecture frames analytics as the bridge between raw data and actionable information.
• Core theme: Innovation is increasingly driven by the insights that analytics can unlock.

Learning Outcomes (Course-Level Targets)

• Explain the need for data analytics.
• Identify major analytics tools.
• Describe the skill set of a data analyst.
• Define Big Data Analytics.
• Distinguish the four types of big-data analytics.
• Cite real-life, cross-industry examples of analytics in action.

Data vs. Information

• Data: raw, unorganized facts.
• Information: data that have been processed, structured or contextualized so that they become meaningful.
• Example progression: list of dates → booking system → full departure/arrival schedule.

Data Analytics: Definition & Workflow

• Science of analyzing raw data to draw conclusions, boost productivity and support business gain.
• Applies to both enterprise-scale and individual-level data streams.
• Typical pipeline:
  • Data extraction & collection
  • Data cleaning
  • Categorization & storage
  • Algorithmic / statistical analysis (many steps automated)
  • Visualization & reporting
  • Decision / action
• Output includes behavioural patterns, anomaly detection, trend forecasts, etc.

Illustrative Use-Cases of Data Analytics (Non-“Big Data” Scale)

1. Fraud Management Reports in banking—identify unauthorized access, hacking, suspicious transactions.
2. Live-Tracking Dashboards in ride-hailing (Meru, Ola, Uber, Mega)—vehicle tracking, customer requests, real-time revenues, emergency alerts.
3. Sales & Target Reports—revenue tracking and future-goal setting across sectors.
4. Google Analytics—site visits, user geolocation, device profiling.

Why Do Organisations Need Data Analytics?

1. Gather Hidden Insights – discover non-obvious patterns aligned with business requirements.
2. Generate Reports – communicate insights so relevant teams can act.
3. Perform Market Analysis – benchmark strengths/weaknesses against competitors.
4. Improve Business Requirements – refine products & customer experience via evidence.

Tools Landscape for Data Analytics

• R – statistical computing & data modelling.
• Python – rich ML & visualisation libraries (Scikit-learn, TensorFlow, Pandas, Matplotlib, Keras).
• Tableau Public – free, connects to numerous sources, web-ready dashboards.
• QlikView – in-memory processing, associative analytics; compresses data to $10\%$ of original size.
• SAS – mature environment for data manipulation & analytics.
• OpenRefine (GoogleRefine) – specialist in cleaning and transforming messy data, plus web scraping.
• Microsoft Excel – ubiquitous for internal client data; supports pivot-table previews.
• RapidMiner – integrated platform for predictive analytics, mining text, ML; supports multiple DBs (Access, SQL Server, Oracle, Teradata…)
• KNIME – open-source, modular pipelines that blend analysis, reporting, and integration.
• Apache Spark – large-scale processing engine; runs on Hadoop clusters $100\times$ faster in memory, $10\times$ faster on disk; foundation for data pipelines & ML model dev.

The Data Analyst Role

• Professional who translates numbers into understandable narratives and strategic guidance.
• Core skill set:
  • Statistics & probability
  • Data collection methodologies
  • Data cleaning & wrangling
  • Exploratory & confirmatory data analysis
  • Hidden-insight discovery
  • Data visualisation & storytelling
  • Report generation
  • Foundational machine learning

Big Data Analytics: Concepts

• Big Data – very large, fast, and/or varied machine-generated datasets, often unstructured, exceeding traditional RDBMS capacity.
• Data Analytics – examination of any data (structured or unstructured).
• Big Data Analytics – complex examination of large & varied sets to surface actionable knowledge for informed decisions.

Four Types of Big-Data Analytics

Gartner Model Examples:

• Automotive:
  • Descriptive – “How many cars did we sell last year?”
  • Diagnostic – “Why did we sell only x cars?”
  • Predictive – “If we run $x$ advertising programs, how many cars can we sell?”
  • Prescriptive – “What must we do to reach sales target $x$?”
• Healthcare (HBP):
  • Descriptive – number of hypertension diagnoses last year.
  • Diagnostic – why specific patients developed HBP.
  • Predictive – stroke likelihood for Mr. Jones.
  • Prescriptive – optimal medication plan to prevent stroke.

Real-Life Big Data Analytics Success Stories

• Customer Acquisition & Retention – Coca-Cola (2015)
  • Built a digital-led loyalty program; data strategy credited for strong retention.
• Targeted Advertising – Netflix
  • Recommender engine leverages search/watch history to personalise next-watch suggestions and content investment decisions.
• Risk Management – UOB Bank
  • Big-data system cut Value-at-Risk computation from $18\text{ hours}$ to few minutes, enabling near-real-time risk analysis.
• Innovation & Product Development – Amazon Fresh / Whole Foods
  • Data-driven logistics reveal grocery-buying patterns and optimise supplier collaboration.
• Supply-Chain Optimisation – PepsiCo
  • Retailers share warehouse & POS inventories; PepsiCo reconciles and forecasts production/shipment volumes, ensuring shelf availability.

Practical, Ethical & Strategic Considerations

• Data privacy & consent—especially with customer behaviour tracking (e.g., Netflix, Coca-Cola loyalty programs).
• Algorithmic bias—ML models used in prescriptive analytics must be audited to avoid unfair outcomes (e.g., loan approvals, hiring).
• Real-time vs. batch trade-offs—Spark & in-memory tech (e.g., $100\times$ speedup) shift organisations toward continuous decision loops.
• Skills shortage—demand for multi-disciplinary analysts combining statistics, domain knowledge and storytelling.

Connections to Foundational Principles & Future Lectures

• Builds on earlier modules covering Digital Transformation and sets stage for upcoming topics like Innovation through IoE (Internet of Everything).
• Analytics acts as the evidential backbone that justifies, measures and refines digital-age innovation projects.

Key Takeaways

• Data alone holds little value until processed into information and insight.
• Analytics spans descriptive summaries to AI-driven prescriptions, each adding incremental business value.
• A diverse tool ecosystem—open-source & commercial—supports every stage from cleaning to in-memory big-data crunching.
• Real-world successes demonstrate competitive advantage in customer retention, risk reduction, product innovation, and supply-chain efficiency.