Org Structures, Culture & KM: How Firms Store & Share Knowledge

Core Concepts: Organizations as Knowledge Systems

1. Organizations as Knowledge Repositories

Every organization is a knowledge system—a living network where:

• Explicit knowledge (documented, codified) lives in manuals, databases, and reports.
• Tacit knowledge (skills, experience, intuition) resides in employees’ minds.

Why it matters: If knowledge isn’t captured or shared, it’s lost when employees leave (e.g., a Daraz delivery manager quitting takes undocumented route optimization tricks with them).

2. Organizational Structures & Knowledge Management

Structures determine how knowledge moves. Three key types:

Worked Example: Kathmandu Traffic Routes

• Problem: Traffic jams waste 3 hours/day in Kathmandu (Nepal’s cost: ~$1B/year).
• KM Fix: The Kathmandu Metropolitan City used explicit knowledge (GPS data) + tacit knowledge (local drivers’ shortcuts) to redesign routes. Result: 20% faster travel in Thapathali.
• Structure Used: Matrix (traffic engineers + local drivers’ input).

3. Organizational Culture & Knowledge Sharing

Culture is the unwritten rules that decide if employees share knowledge. Two extremes:

Real-World Example: Google’s "20% Time" Policy

• Idea: Engineers spend 20% of time on passion projects (e.g., Gmail, Google Maps).
• KM Impact:
  • Tacit → Explicit: Failed projects are documented for others.
  • Culture: "Psychological safety" encourages risk-taking.
• Result: 50% of Google’s innovations come from this policy.

Nepali Counterpart: NEPSE’s Knowledge Black Hole

• Problem: Stockbrokers hoard trading tips; no central database.
• Consequence: Retail investors lose millions to insider deals.
• Fix Needed: A knowledge-sharing platform (like NEPSE’s failed "Investor Education Portal").

In the Real World

1. eSewa’s Payment Fraud Prevention

   • KM Idea: Explicit rules (fraud detection algorithms) + tacit expertise (customer service agents’ scam patterns).
   • How: Agents log scam attempts in a shared database. AI flags similar transactions in real time.
   • Result: 30% drop in fraud since 2022.

2. Pathao’s Driver Knowledge Base

   • KM Idea: Community-driven tacit knowledge (drivers share "no-go zones" via app feedback).
   • How: Pathao’s algorithm combines driver reports with accident data to reroute rides.
   • Impact: 15% fewer delays in Lalitpur.

3. Nabil Bank’s Loan Approval Delays

   • KM Problem: Branch managers hoard approval criteria (tacit knowledge).
   • Fix: Bank introduced a centralized loan policy manual (explicit) + mentorship program (tacit).
   • Outcome: Loan processing time cut from 10 to 3 days.

4. Knowledge Management in Different Organizational Layers

Not all layers share knowledge equally. Here’s how KM works at each level:

Case Study: Toyota’s "Andon Cord"

• KM Idea: Explicit rule (stop the line if a problem arises) + tacit trust (workers feel safe reporting issues).
• Result: Toyota’s assembly lines have zero defects in 99% of cases.
• Nepali Parallel: Nepal Rastra Bank’s "Customer Feedback System" fails because frontline staff fear retaliation for reporting issues.

5. Knowledge Management in Nepali Contexts

Challenge 1: Brain Drain

• Problem: Skilled IT professionals (e.g., from Nepal’s software hubs in Kathmandu/Lalitpur) migrate to India/Singapore, taking tacit knowledge with them.
• KM Solution: Knowledge repositories (e.g., Nepal Engineering College’s digital library) + mentorship programs.

Challenge 2: Rural-Urban Knowledge Gap

• Problem: Farmers in Kavrepalanchok lack access to weather forecasts or pest-control knowledge.
• KM Fix: Agriculture Development Bank’s SMS-based knowledge sharing (e.g., "Send ‘KHAMAR’ to 1234 to get rice disease tips").

Challenge 3: Political Interference in KM

• Problem: At NTC, political appointees override technical knowledge (e.g., approving substandard infrastructure).
• KM Lesson: Explicit documentation (e.g., engineering reports) can protect against political bias.

Exam Tip

How This Unit is Tested (TU/PU/NEB Patterns)

1. Definitions & Comparisons (30%)

   • Expect questions like:
     • "Differentiate between functional and divisional structures in KM."
     • "How does Google’s 20% time policy convert tacit knowledge to explicit?"
   • Answer Tip: Use tables (like the one above) for comparisons.

2. Case Analysis (40%)

   • Format: "Analyze how NTC’s functional structure hinders knowledge sharing."
   • Your Approach:
     1. Identify the structure (functional).
     2. Link to KM weaknesses (silos, slow cross-dept flow).
     3. Suggest fixes (e.g., "matrix teams for infrastructure projects").
   • Real-World Tie: Always relate to Nepali examples (e.g., NTC, NEPSE, Daraz).

3. Problem-Solving (30%)

   • Example Question: "A bank in Nepal faces slow loan approvals. Design a KM strategy using organizational culture and structure."
   • Your Answer:
     • Structure: Shift from functional → matrix (loan officers + risk analysts).
     • Culture: Introduce "knowledge-sharing Fridays" (reward employees for documenting processes).
     • Tech: Use explicit tools (e.g., loan approval software with audit trails).

Model Answer Snippet for Case Study (10 Marks)

Question: "How can Pathao improve its knowledge management to reduce rider delays in Kathmandu?"

Answer:

1. Structure: Adopt a hybrid matrix structure—combine regional teams (divisional) with cross-functional "delay reduction squads" (matrix).
2. Tacit → Explicit:
   • Driver feedback loop: Riders log delays via app; AI clusters common issues (e.g., "traffic at Thapathali 5–7 PM").
   • Explicit database: Share insights with all drivers (e.g., "Avoid Ring Road after 6 PM").
3. Culture:
   • Incentivize sharing: Top 10 drivers with most useful feedback get discounts.
   • Psychological safety: Anonymous reporting for unsafe routes.
4. Tech:
   • Real-time dashboards: Show live traffic + delay hotspots (like Waze but Pathao-specific).
5. Real-World Proof:
   • Google Maps uses similar KM: combines explicit data (traffic cameras) + tacit knowledge (user-reported accidents).

Visual for Exam:

flowchart TD
  A["Driver Reports Delay"] --> B["AI Clusters Issues"]
  B --> C["Explicit Database"]
  C --> D["Shared with All Drivers"]
  D --> E["Reduced Delays"]
  F["Anonymous Feedback"] --> D
  G["Incentives"] --> F