9.5 Root Cause & Decision Analysis Techniques
Key Takeaways
- Root Cause Analysis diagnoses foundational systemic breakdowns rather than treating superficial symptoms, using Fishbone (Ishikawa) Diagrams and the 5 Whys technique.
- Fishbone diagrams categorize causal factors using standard industry taxonomies: 6Ms (Manufacturing), 8Ps (Services), and 4Ss (Office/Administrative).
- Decision Trees evaluate complex multi-stage choices under risk by calculating Expected Monetary Value (EMV) at Chance Nodes and evaluating choices at Decision Nodes.
- Decision Tables and Matrices formalize multi-variable conditional business rules into exhaustive condition and action grids, eliminating logical gaps and contradictions.
- SWOT Analysis evaluates internal Strengths/Weaknesses against external Opportunities/Threats, while Risk Analysis quantifies uncertainty using Probability x Impact matrices.
9.5 Root Cause & Decision Analysis Techniques
Quick Summary: Business analysts must resolve complex organizational problems and evaluate high-stakes choices under uncertainty. BABOK® Guide v3 details diagnostic techniques (Fishbone / Ishikawa Diagrams, 5 Whys) alongside structured decision models (Decision Trees with EMV, Decision Tables, SWOT Analysis, Risk Matrices) to uncover true root causes and optimize strategic decisions.
1. Root Cause Analysis: Fishbone (Ishikawa) & 5 Whys
Root Cause Analysis systematically identifies the foundational, underlying reasons for an undesirable business outcome, defect, or process failure, preventing teams from merely treating surface-level symptoms.
+-----------------------------------------------------------------------------------+
| Fishbone (Ishikawa) Diagram Architecture |
+-----------------------------------------------------------------------------------+
| Category Branch 1 Category Branch 2 Category Branch 3 |
| \ \ \ |
| \── Primary Cause \── Primary Cause \── Primary Cause |
| \ \── Sub-cause \ \── Sub-cause \ |
| ═════════════════════════════════════════════════════════════════> [PROBLEM HEAD]|
| / / / (Defect Statement)
| /── Primary Cause /── Primary Cause / |
| / / / |
| Category Branch 4 Category Branch 5 Category Branch 6 |
+-----------------------------------------------------------------------------------+
The Standard Fishbone Category Taxonomies
| Industry Domain | Taxonomy Framework | Constituent Category Ribs |
|---|---|---|
| Manufacturing / Operations | 6Ms | Machine (equipment, tech), Method (procedures, SOPs), Material (raw inputs, data), Measurement (metrics, inspection), Manpower / Mindpower (personnel, skills), Milieu / Mother Nature (environment). |
| Service Organizations | 8Ps | Product (service offering), Price (pricing model), Place (channel, location), Promotion (marketing), People (staff, training), Process (workflow, handoffs), Physical Evidence (documentation, UI), Productivity & Quality (efficiency). |
| Office & Administrative | 4Ss | Surroundings (work environment), Suppliers (vendors, external data), Systems (software, hardware), Skills (competency, training). |
The 5 Whys Technique
An iterative questioning technique that drills down through successive layers of causation by repeatedly asking "Why?" (typically 5 times) until the systemic process or policy defect is revealed.
Problem: A major online banking outage occurred during Friday peak hours.
├── Why 1? The primary transaction database crashed from memory exhaustion.
│ └── Why 2? An unindexed batch reconciliation query consumed 100% of RAM.
│ └── Why 3? A junior developer deployed the query directly to production.
│ └── Why 4? The automated CI/CD deployment pipeline bypassed query testing.
│ └── Why 5 (ROOT CAUSE)? There was no mandatory peer code review policy
│ or automated query linting gate for database changes.
2. Decision Trees and Expected Monetary Value (EMV)
A Decision Tree is a graphical, predictive decision-modeling tool that evaluates sequential choices, uncertain outcomes, and financial payoffs under conditions of risk.
+-----------------------------------------------------------------------------------+
| Decision Tree Node Notation |
+-----------------------------------------------------------------------------------+
| 1. DECISION NODE (Square - ▢): |
| Represents a choice controlled directly by the decision-maker. |
| |
| 2. CHANCE / EVENT NODE (Circle - ◯): |
| Represents an uncertain event; branches must show probabilities summing to 1.0|
| |
| 3. TERMINAL / END NODE (Triangle - ▷): |
| Represents the final financial payoff or net outcome of a branch pathway. |
+-----------------------------------------------------------------------------------+
Expected Monetary Value (EMV) Calculation
Decision Node [Build Custom vs. Buy SaaS]
│
├── BRANCH 1: Buy SaaS (Upfront Cost = -$200K)
│ └── ◯ Chance Node: Market Adoption
│ ├── [High Demand (P = 0.70)] ──> Payoff = +$800,000
│ └── [Low Demand (P = 0.30)] ──> Payoff = +$200,000
│ EMV = (0.70 * $800K) + (0.30 * $200K) = $560K + $60K = $620,000
│ Net Value = $620K - $200K = $420,000
│
└── BRANCH 2: Build Custom (Upfront Cost = -$500K)
└── ◯ Chance Node: Market Adoption
├── [High Demand (P = 0.70)] ──> Payoff = +$1,200,000
└── [Low Demand (P = 0.30)] ──> Payoff = +$100,000
EMV = (0.70 * $1.2M) + (0.30 * $100K) = $840K + $30K = $870,000
Net Value = $870K - $500K = $370,000
*DECISION: Select Branch 1 (Buy SaaS) because Net EMV ($420K) > Branch 2 ($370K).*
3. Decision Tables and Matrices
Decision Tables formalize complex, multi-variable conditional business rules into a structured tabular grid, guaranteeing that all possible combinations of conditions are evaluated without logical gaps or contradictions.
Decision Table Anatomy (Four Quadrants)
+-----------------------------------------------------------------------------------+
| Decision Table Architecture |
+-----------------------------------------------------------------------------------+
| CONDITION STUB (Inputs) │ CONDITION ALTERNATIVES / RULES (R1 - R4) |
| • Credit Score >= 720? │ Y Y N N |
| • Debt-to-Income <= 36%? │ Y N Y N |
| ────────────────────────────────────┼──────────────────────────────────────────── |
| ACTION STUB (Outputs / Decisions) │ ACTION ENTRIES |
| • Auto-Approve Prime Loan Rate │ X - - - |
| • Refer to Senior Underwriter │ - X X - |
| • Decline Application │ - - - X |
+-----------------------------------------------------------------------------------+
[!NOTE] Rule Completeness Formula: A decision table with $N$ binary (Yes/No) conditions has exactly $2^N$ possible rule combinations. For example, 3 conditions yield $2^3 = 8$ rules. If a table contains fewer rules without intentional collapsing, an unhandled logic defect exists.
4. SWOT Analysis
A SWOT Analysis is a foundational strategic evaluation framework that examines the internal and external factors influencing an organization or proposed change initiative.
┌─────────────────────────────────────────────────────────────────────────────┐
│ SWOT 2x2 Matrix │
├─────────────────────────────────────────────────────────────────────────────┤
│ INTERNAL FACTORS │
│ (Controllable by the Enterprise) │
│ │
│ STRENGTHS (S) WEAKNESSES (W) │
│ • Proprietary patented technology • Legacy mainframe technical deb│
│ • Highly skilled engineering talent • High operational cost base │
│ • Robust balance sheet / cash reserves • Fragmented customer data silos│
│ ─────────────────────────────────────────────────────────────────────── │
│ EXTERNAL FACTORS │
│ (Uncontrollable Environmental Forces) │
│ │
│ OPPORTUNITIES (O) THREATS (T) │
│ • Emerging open banking regulations • Aggressive venture-backed apps│
│ • Growing millennial market segment • Stricter data privacy fines │
│ • Rapid growth of mobile payments • Rising interest rate headwinds│
└─────────────────────────────────────────────────────────────────────────────┘
5. Risk Analysis & Risk Matrix
Risk Analysis evaluates uncertainties that could negatively impact project objectives, quantifying their Probability (Likelihood) and Impact (Consequence) to determine overall Risk Exposure.
The 4 BABOK Negative Risk Response Strategies
| Strategy | Mechanism | Enterprise Scenario |
|---|---|---|
| Avoid | Eliminate the risk entirely by altering project scope, architecture, or approach. | Canceling a high-risk custom payment gateway in favor of an established certified vendor. |
| Mitigate | Implement specific actions to reduce the probability or impact of the risk. | Implementing redundant multi-region cloud servers and automated load balancers to prevent downtime. |
| Transfer | Shift financial or operational impact to a third party (insurance, warranties, contracts). | Purchasing commercial cyber-liability insurance and negotiating SLA penalty clauses with vendors. |
| Accept | Acknowledge the risk and take no proactive action, establishing a contingency reserve if it occurs. | Accepting minor UI rendering quirks on deprecated legacy browser versions. |
Enterprise Scenario: Hospital Emergency Department Triage Bottleneck
A regional hospital network experienced critical patient wait-time spikes in its emergency department (ED):
- Fishbone Analysis: Facilitated a session using the 8Ps Framework. Process: Triage nurses used duplicate paper intake forms; People: New nursing staff lacked triage protocol training; Physical Evidence: Room signage was confusing; Productivity: Lab blood tests averaged 90 minutes.
- 5 Whys: Drilled into lab delays: Why? Phlebotomy tubes sat in queues -> Why? Order transport was manual -> Why? Pneumatic tube station was broken -> Why? Maintenance contract lapsed -> Root Cause: Vendor SLA tracking was unassigned.
- Decision Tree: Evaluated replacing the pneumatic tube system ($120K cost, Net EMV = +$450K) vs. hiring 4 courier staff ($180K annual OpEx, Net EMV = +$280K), leading to a high-confidence recommendation to replace the mechanical system.
[!TIP] CCBA Exam Tip: On the exam, when calculating Expected Monetary Value (EMV) for Decision Trees, remember to evaluate backward from right to left (Rollback method). At each Chance Node (circle), multiply branch probabilities by payoffs. At each Decision Node (square), choose the branch with the highest net payoff.
[!WARNING] CCBA Exam Trap: Be careful not to confuse Internal with External factors in SWOT questions. A company's outdated IT infrastructure is a Weakness (Internal), NOT a Threat. A new competitor launching an AI platform is a Threat (External), NOT a Weakness.
A business analyst is facilitating a root cause analysis session to diagnose why customer loan onboarding applications in a retail banking branch experience excessive error rates. The analyst selects a Fishbone (Ishikawa) diagram. Because the banking branch operates in a pure customer service environment, which standard category taxonomy is MOST appropriate to structure the causal branches?
A business analyst is evaluating two competing system architecture options using a Decision Tree. Option A requires an upfront investment of $100,000 and leads to a chance node with two outcomes: a 60% probability of generating $500,000 in revenue, and a 40% probability of generating $200,000 in revenue. Option B requires an upfront investment of $250,000 and leads to a chance node with a 60% probability of generating $800,000 in revenue, and a 40% probability of generating $300,000 in revenue. Based on Net Expected Monetary Value (EMV minus upfront cost), which option should the business analyst recommend?
An enterprise insurance platform is designing an automated underwriting rules engine that evaluates applicant eligibility based on four independent binary (True/False) conditions: (1) Valid Driver's License, (2) Clean Driving Record, (3) Vehicle Age <= 10 Years, and (4) Primary Garage in Covered Zone. To guarantee that the rules engine has no missing logical combinations or unhandled exceptions, how many distinct rule columns must be evaluated in the Decision Table?