E-commerce Checkout Conversion Crisis

=== ROOT CAUSE ANALYSIS FRAMEWORK ===

STEP 1: CLARIFYING QUESTIONS (Ask These First)
------------------------------------------------

1. SCOPE: Is the 25% drop affecting all order values equally, or is it concentrated in specific price ranges within the $100-$5000 band?
   - Why ask: High-value orders may have different payment processing or fraud detection thresholds that could be triggered

2. SUDDEN vs. GRADUAL: Did the drop happen instantly at a specific time yesterday or build throughout the day?
   - Why ask: Instant = likely deployment, external service failure, or fraud system change; Gradual = payment provider degradation or competitive event

3. FUNNEL STEP: Which specific step of checkout saw the abandonment? (Cart → Checkout → Payment Entry → Payment Processing → Confirmation)
   - Why ask: Pinpointing exact step narrows root cause dramatically—payment processing vs. payment entry have different causes

4. PLATFORM BREAKDOWN: What's the split of the drop between mobile (iOS/Android) and web? Are both affected equally?
   - Why ask: Platform-specific issues are common; helps identify if it's technical vs. systemic

5. PAYMENT METHODS: Are all payment methods affected, or specific ones? (Credit card, digital wallets, buy-now-pay-later)
   - Why ask: Payment provider or gateway issues often affect specific payment types; high-value orders may use different methods

6. GEOGRAPHY: Is this global or concentrated in specific regions/countries?
   - Why ask: Regional payment provider issues, fraud rules, or currency exchange problems could be localized

7. DEPLOYMENTS: You mentioned no known deployments—can we confirm zero code changes, configuration updates, or third-party service updates?
   - Why ask: "No known deployments" often means untracked changes—need to verify definitively


STEP 2: MECE HYPOTHESIS FRAMEWORK
----------------------------------

CATEGORY A: INTERNAL - PAYMENT SYSTEM & FRAUD (HIGHEST PRIORITY)

1. Payment gateway API failure or elevated error rates for high-value transactions
   Impact: CRITICAL | Likelihood: HIGH | Data: Payment API success rates by transaction value, error codes, response times

2. Fraud detection system became more aggressive (flagging more high-value orders)
   Impact: CRITICAL | Likelihood: HIGH | Data: Fraud system decline rates, risk scores distribution, false positive rates

3. 3D Secure authentication failing or timing out for high-value transactions
   Impact: CRITICAL | Likelihood: MEDIUM | Data: 3DS completion rates, timeout errors, segmented by order value

4. Payment gateway timeout issues specific to high-value transaction processing
   Impact: HIGH | Likelihood: MEDIUM | Data: Payment latency by order value, timeout error logs

5. Payment method restrictions changed (card limits, wallet caps for high-value)
   Impact: HIGH | Likelihood: MEDIUM | Data: Payment method availability logs, declined transaction reasons


CATEGORY B: INTERNAL - CHECKOUT UX/TECHNICAL

6. Recent untracked deployment broke checkout flow (config change, CDN update, A/B test)
   Impact: CRITICAL | Likelihood: HIGH | Data: Deployment logs (including config, feature flags, CDN), version control commits

7. JavaScript errors preventing checkout submission
   Impact: CRITICAL | Likelihood: MEDIUM | Data: Client-side error logs, Sentry/Bugsnag reports by browser/device

8. Mobile app or web checkout form validation failing (especially for high-value orders with extra fields)
   Impact: HIGH | Likelihood: MEDIUM | Data: Form submission success rates, field-level interaction tracking

9. Page load performance degraded (slow checkout loads causing abandonment)
   Impact: HIGH | Likelihood: MEDIUM | Data: P95 page load times, session abandonment by load time

10. SSL/TLS certificate or security issue preventing secure checkout
    Impact: CRITICAL | Likelihood: LOW | Data: SSL handshake errors, browser security warnings


CATEGORY C: EXTERNAL - PAYMENT PROVIDERS & SERVICES

11. Payment processor experiencing outage or degradation (Stripe, PayPal, etc.)
    Impact: CRITICAL | Likelihood: MEDIUM | Data: Payment processor status pages, external monitoring, success rates by provider

12. Third-party fraud service (Sift, Riskified) changed rules or is down
    Impact: CRITICAL | Likelihood: MEDIUM | Data: Fraud service API health, rule change logs, response times

13. Bank authorization decline rates increased (card issuer-side changes)
    Impact: HIGH | Likelihood: MEDIUM | Data: Soft declines vs. hard declines, decline reason codes from banks

14. Currency exchange rate service failure (for international orders)
    Impact: MEDIUM | Likelihood: LOW | Data: Currency conversion API errors, international order breakdown


CATEGORY D: BUSINESS LOGIC & PRICING

15. Pricing calculation bug showing incorrect totals for high-value orders
    Impact: HIGH | Likelihood: MEDIUM | Data: Pricing logs, customer support tickets about incorrect amounts

16. Tax calculation errors (especially for cross-border high-value orders)
    Impact: MEDIUM | Likelihood: MEDIUM | Data: Tax calculation API logs, tax-related checkout abandonment

17. Shipping calculation errors (no shipping options for high-value items)
    Impact: MEDIUM | Likelihood: MEDIUM | Data: Shipping API responses, "no shipping available" error rates

18. Coupon or discount code system malfunction
    Impact: MEDIUM | Likelihood: LOW | Data: Coupon application logs, discount calculation errors


CATEGORY E: EXTERNAL - COMPETITIVE & MARKET

19. Major competitor launched aggressive promotion targeting high-value purchases
    Impact: MEDIUM | Likelihood: LOW | Data: Competitive intelligence, social media monitoring, app store rankings

20. Negative PR or social media event affecting brand trust
    Impact: MEDIUM | Likelihood: LOW | Data: Social media sentiment, news monitoring, brand search trends


CATEGORY F: DATA QUALITY & MEASUREMENT (CHECK EARLY!)

21. Checkout conversion tracking instrumentation broken
    Impact: CRITICAL | Likelihood: MEDIUM | Data: Sanity check actual revenue vs. tracked conversions, event volume consistency

22. Analytics pipeline failure undercounting conversions
    Impact: CRITICAL | Likelihood: LOW | Data: Data freshness checks, pipeline logs, compare multiple tracking sources

23. Checkout event definition changed (now excluding certain completion states)
    Impact: HIGH | Likelihood: LOW | Data: Event schema changes, conversion event volume trends


STEP 3: PRIORITIZATION MATRIX
------------------------------

CHECK IMMEDIATELY (High Impact × High Likelihood):
→ Payment gateway API failure for high-value transactions (#1)
→ Fraud detection system became more aggressive (#2)
→ Recent untracked deployment (#6)
→ Data tracking instrumentation broken (#21) - ALWAYS verify data quality first
→ JavaScript errors preventing checkout (#7)

Rationale: These have both the highest probability and would cause exactly the symptom described (sudden 25% drop across mobile and web). Payment and fraud systems are the most common culprits for high-value checkout issues.

CHECK SECOND (High Impact × Medium Likelihood OR Quick to Verify):
→ Payment processor experiencing outage (#11) - Check status pages immediately
→ Third-party fraud service changes (#12)
→ 3D Secure authentication failing (#3)
→ Mobile/web form validation failing (#8)
→ Bank decline rates increased (#13)

Rationale: These are plausible and relatively fast to verify through logs and external monitoring.

CHECK THIRD (Lower Impact or Lower Likelihood but worth investigating):
→ Pricing calculation bug (#15)
→ Page load performance degraded (#9)
→ Payment method restrictions changed (#5)
→ SSL/TLS issues (#10)
→ All remaining hypotheses in order of context-specific likelihood


STEP 4: STRATEGIC DATA REQUESTS
--------------------------------

1. DATA QUALITY VERIFICATION (rule out false alarm):
   "First, let's verify this is real:
    - Compare tracked checkout conversions vs. actual completed revenue for yesterday vs. prior week
    - Check analytics event volumes for consistency
    - Confirm conversion tracking events are firing at expected rates
    - Cross-reference with payment processor transaction counts"

2. SEGMENTATION (narrow scope fast):
   "Show me checkout conversion rate for yesterday vs. last 7 days, segmented by:
    - Platform: iOS vs. Android vs. Web (desktop and mobile web separately)
    - Order value bands: $100-500, $500-1000, $1000-2500, $2500-5000
    - Payment method: Credit card, debit card, PayPal, Apple Pay, etc.
    - Geography: Top 10 countries
    - New vs. returning customers
    - Hour-by-hour breakdown for yesterday"

3. FUNNEL BREAKDOWN (pinpoint failure step):
   "Provide funnel completion rates for each step:
    - Add to cart → Checkout initiation
    - Checkout initiation → Payment info entered
    - Payment info entered → Payment submitted
    - Payment submitted → Payment processing success
    - Payment processing success → Order confirmation
    Show for yesterday vs. prior week, with drop-off rates at each stage"

4. TECHNICAL HEALTH (rule out infrastructure):
   "Can you provide for yesterday vs. prior week:
    - Payment API success rates by provider and order value
    - Payment API error codes and their frequency
    - Payment API response times (P50, P95, P99)
    - 3D Secure completion rates
    - JavaScript error rates (client-side)
    - Page load times for checkout pages (P95)
    - SSL handshake failures"

5. FRAUD & SECURITY SYSTEMS (critical for high-value):
   "Pull fraud system data:
    - Fraud system decline rates by order value
    - Risk score distribution for attempted orders
    - False positive rate trends
    - Fraud service API health and response times
    - Any fraud rule changes or model updates
    - Bank soft decline vs. hard decline rates"

6. RECENT CHANGES (correlate with timing):
   "Audit all changes in past 48 hours:
    - Any code deployments (frontend or backend)
    - Feature flag changes
    - Configuration updates (payment, fraud, pricing)
    - A/B test starts/stops
    - CDN or infrastructure changes
    - Third-party service updates (payment processors, fraud services)
    - Database schema or pricing table changes"

7. EXTERNAL CONTEXT (payment providers):
   "Check external systems:
    - Payment processor status pages (Stripe, PayPal, etc.)
    - Fraud service provider status (Sift, Riskified, etc.)
    - Any reported outages or incidents
    - Bank network issues or maintenance windows"


STEP 5: INTERVIEW ANSWER STRUCTURE (Use This Flow)
---------------------------------------------------

1. CLARIFY (2-3 minutes):
   "Before diving into hypotheses, let me clarify a few critical details about the drop..."
   [Ask top 5-6 clarifying questions from Step 1]
   
   Example: "First, can you tell me if the 25% drop is evenly distributed across the $100-$5000 range, or concentrated in higher-value orders? Second, what time yesterday did this start—was it instant or gradual? Third, which specific step in the checkout funnel saw the abandonment?"

2. FRAMEWORK (1 minute):
   "I'll organize potential root causes using a MECE framework with six categories: Payment & Fraud Systems, Technical/UX, External Services, Business Logic, Competitive Factors, and Data Quality. This ensures we systematically cover all possibilities without overlap."

3. HYPOTHESES (5-7 minutes):
   "Let me walk through the most likely causes, starting with highest probability..."
   
   [Present top 12-15 hypotheses organized by category]
   
   Example structure:
   "PAYMENT & FRAUD: Given this affects high-value orders specifically, I'd first investigate:
   - Payment gateway failures for large transactions—these often have different processing paths
   - Fraud system becoming more aggressive—fraud models often treat high-value orders differently
   - 3D Secure issues—many high-value transactions require additional authentication
   
   TECHNICAL: Next, technical issues that could affect both mobile and web:
   - Untracked deployment or config change—you mentioned no 'known' deployments, but we should verify
   - JavaScript errors breaking checkout submission
   - Form validation failures for high-value orders which may have extra required fields
   
   EXTERNAL: Then external dependencies:
   - Payment processor experiencing issues—need to check their status
   - Third-party fraud service degradation or rule changes..."
   
   [For each hypothesis: state it, explain why it fits the symptoms, specify data to check]

4. PRIORITIZATION (2-3 minutes):
   "Based on impact, likelihood, and ease of verification, here's how I'd investigate:
   
   FIRST: I'd start with a quick data quality check to rule out a tracking bug—we need to confirm this is a real revenue drop, not a measurement issue.
   
   SECOND: Immediately check payment gateway API health and fraud system decline rates. These are the most common causes of sudden high-value checkout drops and are fast to verify through logs and dashboards.
   
   THIRD: Audit for any untracked changes—config updates, feature flags, A/B tests—that might have been deployed despite the 'no deployments' statement.
   
   The logic: Payment systems and fraud rules are the highest probability causes for high-value order failures, and they can be verified quickly. Then we work through technical issues that affect both mobile and web."

5. DATA REQUESTS (2-3 minutes):
   "To test these hypotheses, I need specific data in this order..."
   [Present your strategic data requests from Step 4]
   
   Example: "First, give me checkout conversion segmented by order value, platform, and payment method for yesterday vs. the prior week—this will tell us exactly where the problem is concentrated. Second, pull payment API success rates and error codes by order value. Third, show me the checkout funnel drop-off at each step..."

6. NEXT STEPS (1-2 minutes):
   "Once I have this data, I'd:
   1. Isolate the root cause by systematically eliminating hypotheses
   2. If it's a payment or fraud system issue, work with those teams to roll back changes or adjust rules
   3. If it's a code issue, identify the deployment and roll back or hotfix
   4. Monitor recovery in real-time and validate fix effectiveness
   5. Conduct a retrospective to understand how this wasn't caught and improve monitoring"


COMMON MISTAKES TO AVOID:
-------------------------
❌ Not asking which checkout funnel step failed—this is the #1 question for conversion issues
❌ Ignoring fraud detection systems—they're a top cause of high-value order failures but often overlooked
❌ Taking "no deployments" at face value—config changes, feature flags, and CDN updates often aren't tracked
❌ Not segmenting by order value—high-value orders have different payment processing and fraud rules
❌ Forgetting to verify data quality—always sanity check actual revenue vs. tracked conversions
❌ Not checking payment processor status pages—external service issues are common
❌ Missing 3D Secure authentication—required for many high-value transactions and frequently breaks
❌ Not requesting error codes—"payment API failure" is too vague; need specific error codes to diagnose
❌ Ignoring soft declines vs. hard declines—bank authorization issues show up differently
❌ Not considering that mobile and web may use different payment SDKs—simultaneous failure suggests upstream issue
❌ Jumping to competitive factors too quickly—technical issues are far more likely for sudden 25% drops
❌ Not asking about hour-by-hour breakdown—tells you exactly when it started and helps correlate with changes