Business Intelligence: Definitions, Systems, and Strategic Value

What is Business Intelligence (BI)?

Business Intelligence (BI) refers to the processes, technologies, and tools used to transform raw data into meaningful insights that drive better decision-making. It bridges the gap between data and actionable knowledge, enabling organizations to optimize performance, identify trends, and respond to market changes proactively.

Key Definitions

• BI as a Process: A cyclical workflow of collecting, analyzing, and disseminating data to support strategic goals.
• BI as a Technology: Software and hardware tools (e.g., SQL, Tableau, Power BI) that process and visualize data.
• BI as a Culture: An organizational mindset that values data-driven decision-making over intuition.

---

The BI Framework: Components and Architecture

BI systems integrate multiple layers to deliver insights. Below is a breakdown of its core components:

1. Data Sources

Raw data originates from:

• Internal sources: Transactional databases (e.g., sales records, customer databases).
• External sources: Market trends, competitor data, social media, government statistics.

2. Data Storage and ETL

• ETL (Extract, Transform, Load): Cleans and structures raw data for analysis.
  • Extract: Pull data from sources (e.g., Daraz’s order history).
  • Transform: Standardize formats (e.g., converting currency to NPR).
  • Load: Store in a data warehouse (centralized repository).

3. BI Tools and Analytics

Tools like Power BI, Tableau, or Qlik enable:

• Reporting: Static summaries (e.g., monthly sales reports).
• Ad-hoc analysis: Exploratory queries (e.g., "Why did Pathao’s deliveries drop in Kathmandu?").
• Predictive modeling: Forecasting (e.g., NEPSE predicting stock trends).

4. Dashboards and Visualization

Interactive dashboards (e.g., Google Data Studio) present insights via:

• Charts (line, bar, pie).
• Heatmaps (e.g., traffic congestion in Kathmandu).
• Key Performance Indicators (KPIs) like customer retention rate or operational efficiency.

---

Types of Business Intelligence

BI is categorized based on its analytical focus:

---

In the Real World

1. eSewa’s Fraud Detection

   • Idea Used: Descriptive + Predictive BI
   • How: eSewa analyzes transaction patterns to flag unusual activities (e.g., sudden large payments). Machine learning models predict fraud risks in real-time, reducing financial losses.

2. Nabil Bank’s Loan Approval

   • Idea Used: Prescriptive BI
   • How: The bank uses BI to assess a customer’s creditworthiness by analyzing income, repayment history, and market trends. The system then recommends loan terms (interest rate, tenure) automatically.

3. Pathao’s Driver Routing

   • Idea Used: Optimization (Prescriptive BI)
   • How: Pathao’s algorithm dynamically adjusts driver routes based on traffic (NTC data), demand hotspots, and fuel efficiency to minimize delivery time and costs.

---

Worked Example: Kathmandu Traffic Congestion Analysis

Scenario: The Kathmandu Metropolitan City (KMC) wants to reduce traffic jams using BI.

Step-by-Step Trace

1. Data Collection:

   • Sources: NTC’s traffic cameras, GPS data from Pathao/Daraz delivery vehicles, and police reports.
   • Metrics: Vehicle count, average speed, accident frequency.

2. Data Processing:

   • Clean data (remove duplicates, handle missing values).
   • Aggregate by time (peak hours: 8–10 AM, 5–7 PM).

3. Analysis:

   • Descriptive: Identify congested routes (e.g., Thapathali to Kantipath).
   • Diagnostic: Correlate congestion with events (e.g., school hours, festivals).
   • Predictive: Use historical data to forecast high-traffic days (e.g., Dashain, Tihar).

4. Prescriptive Action:

   • Recommend solutions:
     • Dynamic traffic light timing (integrated with NTC’s system).
     • Promote carpooling via apps (like Khalti’s ride-sharing pilot).
     • Expand metro routes (based on demand hotspots).

Visualization:

---

Advantages and Challenges of BI

Advantages

• Data-Driven Decisions: Reduces guesswork (e.g., NEPSE traders use BI for stock picks).
• Competitive Edge: Companies like Daraz use BI to personalize recommendations.
• Efficiency: Automates reporting (e.g., banks generate monthly statements instantly).
• Risk Management: Predicts market downturns (e.g., Global Investment House in Nepal).

Challenges

---

Case Study: Daraz Nepal’s BI-Driven Growth

Background: Daraz, an Alibaba subsidiary, dominates Nepal’s e-commerce market. BI helps it:

1. Personalize Recommendations:
   • Uses collaborative filtering (like Amazon) to suggest products based on past purchases.
   • Example: If a user buys a phone, Daraz recommends cases, screen guards, and accessories.

2. Demand Forecasting:

   • Predicts stock for festivals (e.g., Dashain, Tihar) to avoid shortages.
   • Reduces overstocking by 20% using time-series analysis.

3. Supply Chain Optimization:

   • BI models analyze delivery routes (like Pathao) to cut costs.
   • Partners with NTC to optimize last-mile delivery during monsoons.

Visual:

---

Exam Tip

1. Definitions Matter: Always define BI in the context of process, technology, and culture. Examiners often ask for a 3-part definition.
2. Link Theory to Practice: For questions on BI types (descriptive/predictive), give a Nepali example (e.g., "NEPSE uses predictive BI for stock trends").
3. Diagrams = Easy Marks: Draw the BI framework (data sources → warehouse → tools → dashboards) in exams. Label each component clearly.
4. Case Study Focus: Expect questions on how companies like Daraz or Nabil Bank use BI. Prepare a short bullet-point analysis (e.g., "Daraz uses BI for X, Y, Z").
5. Advantages/Disadvantages: Memorize 2 pros and 2 cons of BI (e.g., "High cost but reduces errors").

---

Key Takeaway: BI is not just about tools—it’s about turning data into strategy. Whether it’s eSewa detecting fraud or Daraz optimizing deliveries, the goal is the same: smarter decisions, faster.