Marketing Information Systems: Data, Research & Decision-Making

TAKEAWAYS:

• A Marketing Information System (MIS) is a structured process that turns raw data (internal + external) into actionable insights for marketing strategies.
• The 5 components of MIS (internal reports, marketing intelligence, marketing research, marketing analytics, and marketing decision support) work together like a feedback loop to refine decisions.
• Marketing research follows a 6-step process (problem definition → data collection → analysis → reporting → decision → implementation) and uses tools like surveys, experiments, and data mining.
• Internal data (sales records, customer databases) and external data (competitor analysis, economic trends) are both critical—Nepal’s NTC uses both to set mobile tariffs.
• CRM analytics (e.g., eSewa’s purchase history) helps personalize offers, while predictive modeling (e.g., Daraz’s demand forecasting) reduces stockouts.
• Ethical concerns (privacy, bias) and costs (time, budget) must balance the benefits of data-driven marketing.

1. What is a Marketing Information System (MIS)?

A Marketing Information System (MIS) is a structured framework that collects, stores, analyzes, and distributes marketing data to help managers make informed decisions. Unlike generic business intelligence, MIS focuses exclusively on marketing-related data—customer behavior, market trends, competitor actions, and internal performance.

Why is MIS Important?

• Reduces guesswork: Replaces intuition with data-backed strategies.
• Improves targeting: Helps segment markets precisely (e.g., Khalti’s targeting of rural vs. urban users).
• Enhances efficiency: Identifies waste (e.g., NTC’s unused spectrum allocation).
• Drives innovation: Spots gaps (e.g., Pathao’s surge pricing during festivals).

How MIS Works: The Feedback Loop

flowchart TD
    A["Marketing Problem"] --> B["Data Collection\n(Internal/External)"]
    B --> C["Data Storage\n(Databases, CRM)"]
    C --> D["Data Analysis\n(Statistics, AI)"]
    D --> E["Reporting\n(Dashboards, Insights)"]
    E --> F["Decision Making\n(Campaigns, Pricing)"]
    F --> G["Implementation\n(Execution)"]
    G -->|"Feedback"| A

Key Idea: MIS is not static—it’s a continuous cycle where each decision generates new data.

2. The 5 Components of MIS

MIS integrates five key elements, each playing a unique role. Visualize them as layers of a pyramid:

mindmap
  root((Marketing Information System))
    Internal Reports
    Marketing Intelligence
    Marketing Research
    Marketing Analytics
    Decision Support

A. Internal Reports System

Definition: Uses existing company data (sales, inventory, customer service records) to track performance. Example:

• Nabil Bank tracks loan default rates by region to adjust interest rates.
• Daraz monitors order fulfillment times to optimize logistics.

Visual: Internal Data Sources

(Imagine a dashboard showing monthly sales of smartphones in Kathmandu vs. Pokhara.)

B. Marketing Intelligence System

Definition: Gathers external data (competitors, economic trends, social media) continuously (not just for research projects). Sources:

• Competitor analysis (e.g., Ncell vs. NTC tariff wars).
• Economic indicators (e.g., Nepal Rastra Bank’s inflation data affecting demand for durables).
• Social listening (e.g., WhatsApp groups discussing product complaints).

Real-World Tie-In:

• Himalayan Java uses Google Trends to predict coffee demand spikes before Dashain.
• Toyota Kirloskar Nepal tracks global oil prices to adjust vehicle pricing.

C. Marketing Research System

Definition: Structured, project-based data collection to answer specific questions (e.g., "Why are eSewa users switching to Khalti?"). Steps (with Nepal example):

1. Define the problem: "Why is Pathao’s rider retention low in Chitwan?"
2. Develop research design: Survey riders vs. drivers.
3. Collect data: Online questionnaires + interviews.
4. Analyze data: Find that lack of insurance is the top concern.
5. Report findings: Present to Pathao’s HR team.
6. Take action: Introduce mandatory insurance for riders.

Tools Used:

• Surveys (Google Forms, SurveyMonkey).
• Experiments (A/B testing Daraz’s checkout page).
• Data mining (Analyzing eSewa’s transaction logs).

D. Marketing Analytics System

Definition: Uses statistical tools and AI to predict trends and measure campaign effectiveness. Key Techniques:

Case Study: Daraz’s Demand Forecasting Daraz uses machine learning to predict demand for Diwali gifts in Nepal. In 2022, their model forecasted a 30% increase in LED lights sales in Kathmandu, allowing them to stock 20% more inventory and avoid shortages.

E. Marketing Decision Support System (DSS)

Definition: Interactive tools (e.g., Excel solvers, AI chatbots) that help managers simulate scenarios before deciding. Example:

• NTC’s DSS models how changing mobile data prices affects rural vs. urban adoption.
• Nabil Bank’s DSS predicts default risks for loan applicants using credit scores.

Visual: DSS in Action

flowchart LR
    A["Manager Input:\n'What if we lower prices by 10%?'] --> B["DSS Simulates:\n- Competitor reaction\n- Sales volume change\n- Profit impact"]
    B --> C["Output:\n'Revenue drops 5%, but market share gains 15%'"]
    C --> D["Decision:\n'Proceed with discount in Pokhara only'"]

3. Marketing Research Process: Step-by-Step

Every research project follows a 6-step cycle. Let’s trace how eSewa might research customer satisfaction:

flowchart TD
    A["1. Define Problem:\n'Why are users rating eSewa 2.5/5?'"] --> B["2. Develop Approach:\n- Survey 10,000 users\n- Focus groups in 3 cities"]
    B --> C["3. Collect Data:\n- Online survey\n- App store reviews"]
    C --> D["4. Analyze Data:\n- 60% complain about slow transactions\n- 30% want more offline agents"]
    D --> E["5. Present Findings:\n'Report to eSewa’s UX team'"]
    E --> F["6. Take Action:\n- Hire 500 offline agents\n- Speed up payment processing"]

Key Research Methods

Worked Example: Pathao’s Rider Retention Problem: Riders in Janakpur quit at twice the rate of Kathmandu. Research Steps:

1. Survey 500 riders → Find low earnings and lack of safety are top issues.
2. Interview managers → Discover no performance bonuses in Janakpur.
3. Analyze data → Correlate low earnings with high quit rates.
4. Recommendation: Introduce performance-based incentives in Janakpur. Result: Quit rate drops by 40% in 6 months.

4. Data Sources: Internal vs. External

Real-World Example: NTC’s Data Strategy NTC uses:

• Internal data: Call drop rates by tower.
• External data: Nepal Government’s population growth reports.
• Secondary data: GSMA Intelligence reports on global 5G trends. Outcome: Targeted 4G expansion in rural areas where demand was underestimated.

5. Tools and Technologies in MIS

Case Study: Khalti’s Fraud Detection Khalti uses AI-powered MIS to:

1. Flag unusual transactions (e.g., sudden large payments).
2. Cross-reference with user behavior (e.g., "This user never transfers > Rs. 50,000").
3. Send alerts to fraud teams in real-time. Result: 30% reduction in fraudulent transactions in 2023.

6. Challenges and Ethical Considerations

A. Common Challenges

1. Data Overload: Too much data can paralyze decision-making (e.g., NTC’s terabytes of call logs).
2. Data Quality Issues: Incomplete or biased data (e.g., rural users underrepresented in surveys).
3. Cost and Time: High-quality research is expensive (e.g., Daraz’s demand forecasting costs millions).
4. Technological Barriers: Small businesses lack advanced analytics tools.

B. Ethical Concerns

Real-World Ethical Dilemma: Pathao’s Surge Pricing

• Issue: During lockdowns, Pathao increased fares by 300%.
• Ethical Question: Is this exploitative or fair (since supply dropped)?
• MIS Role: Pathao’s system automatically adjusted prices based on demand, but lacked transparency about how data was collected.

7. Marketing Information Systems in Action: Case Study

Company: Nabil Bank (Nepal)

Challenge: High loan default rates in far-western Nepal. MIS Solution:

1. Internal Data: Analyzed 3 years of loan repayment data.
2. External Data: Combined with Nepal Rastra Bank’s rural income reports.
3. Analytics: Identified lack of collateral and low awareness as key issues.
4. Decision: Launched "Kisan Credit Card" with collateral-free loans for farmers.
5. Result: Default rates dropped by 25% in target regions.

Visual: Nabil Bank’s MIS Framework

mindmap
  root((Nabil Bank’s MIS for Loan Defaults))
    Internal Data
      Loan repayment history
      Customer demographics
    External Data
      Nepal Rastra Bank reports
      Weather data (crop failure risks)
    Analytics
      Cluster analysis: High-risk regions
      Regression: Income vs. repayment correlation
    Action
      Kisan Credit Card program
      Farmer education campaigns

8. Digital and Modern Marketing Integration

Modern MIS leverages digital tools to create real-time insights:

• AI Chatbots (e.g., eSewa’s virtual assistant) analyze user queries to improve service.
• IoT Sensors (e.g., Daraz’s warehouse temperature logs) track perishable goods.
• Blockchain (e.g., Nepal Food Basket’s supply chain transparency) ensures data integrity.

Example: YouTube’s MIS YouTube uses MIS to:

1. Track watch time (internal data).
2. Analyze trending topics (external: Google Trends).
3. Recommend videos (analytics: collaborative filtering).
4. Adjust ad prices (DSS: auction models).

In the Real World

1. eSewa’s Purchase History Analytics

   • Idea Used: CRM analytics + predictive modeling
   • How: eSewa tracks user transaction patterns (e.g., "Users who buy insurance also upgrade phones"). They then personalize offers like:
     • "You bought a SIM card—here’s a 10% discount on a new phone!"
   • Impact: 20% increase in upsells (e.g., mobile users buying data bundles).

2. Daraz’s Demand Forecasting for Diwali

   • Idea Used: Time-series forecasting + machine learning
   • How: Daraz’s MIS predicts surges in demand for:
     • LED lights (up 300% in October).
     • Gifts (personalized recommendations based on past purchases).
   • Result: Reduced stockouts by 40% and increased sales by 15%.

3. NTC’s Network Optimization

   • Idea Used: Geospatial analytics + internal call data
   • How: NTC’s MIS identifies:
     • Areas with high call drops (e.g., Dharan, Biratnagar).
     • Times of peak usage (e.g., 7–9 PM during exams).
   • Action: Added towers in Dharan and optimized spectrum allocation.
   • Impact: Call drop rate improved by 35% in 2023.

Exam Tip

How to Score Full Marks

1. Define Clearly: Always start with precise definitions (e.g., "A Marketing Information System is a structured process that...").
2. Use Diagrams: Draw flowcharts (e.g., MIS components) or tables (e.g., internal vs. external data) to visualize answers.
3. Link to Nepal: Every example must be Nepali (e.g., eSewa, NTC, Daraz) to show local relevance.
4. Step-by-Step for Research: If asked about marketing research, list all 6 steps with a real example.
5. Critique Limitations: For MIS challenges, mention ethics, cost, and data quality—examiners love balanced answers.
6. Case Studies: For 10+ marks, use Nabil Bank, Daraz, or eSewa as examples—they’re highly examinable.

Common Mistakes to Avoid:

• ❌ Vague answers (e.g., "MIS is important" → Why? Explain with data).
• ❌ Ignoring Nepal context (e.g., using Coca-Cola instead of Root Beer).
• ❌ Skipping diagrams (even a simple flowchart can add 2+ marks).

Sample High-Scoring Answer Structure:

*"A Marketing Information System (MIS) is a structured framework that integrates internal reports, marketing intelligence, research, analytics, and decision support to provide actionable insights. For example, eSewa’s MIS uses CRM data to track user behavior and predict churn, reducing customer loss by 15%. The 5 components work as follows:

1. Internal reports (e.g., eSewa’s transaction logs).
2. Marketing intelligence (e.g., tracking Khalti’s new features).
3. Marketing research (e.g., surveying users on payment delays).
4. Analytics (e.g., predicting Diwali sales spikes).
5. Decision support (e.g., A/B testing discount offers). Challenges include data privacy (e.g., eSewa’s user data leaks) and high costs (e.g., Daraz’s AI tools)."*