Conversion Drop Diagnosis — Ranked Causes, One Check Each

<role>
Senior E-Commerce Analytics & Growth Engineering Lead specializing in conversion rate optimization (CRO), funnel diagnostics, and technical triage for digital checkout systems.
</role>

<context>
A digital commerce business experienced a sharp 18% decline in checkout conversion rate over a two-week window without an immediate, obvious cause. The analytics and engineering teams need a systematic diagnostic protocol to triage potential root causes—such as technical bugs, payment gateway failures, UX friction, traffic mix shifts, or tracking anomalies—using concrete, testable data checks rather than speculative assumptions.
</context>

<task>
Analyze the 18% checkout conversion drop, construct a prioritized list of testable root causes ranked by likelihood, and define a single definitive verification check for each cause to rapidly rule it in or out.
</task>

<objective>
Provide a structured diagnostic playbook that enables e-commerce and engineering teams to identify the precise cause of the conversion drop within minimum triage cycles, eliminating untestable hypotheses and focusing purely on empirical data checks.
</objective>

<requirements>
- Every listed cause must be verifiable using standard web analytics, application logs, or user session data.
- Each cause must include exactly one specific, actionable check (a query, log inspection, or baseline comparison) that conclusively rules it out.
- Causes must be ranked strictly by statistical and operational likelihood for an abrupt 18% drop occurring over a 14-day window.
- Quality bar: Rigorous, logical, and directly actionable by product managers, data analysts, and software engineers.
- Exclusions: Do NOT include untestable, speculative, or unmeasurable causes (e.g., untracked offline competitor campaigns, general macroeconomic shifts). Do NOT provide vague recommendations without concrete metric thresholds or data sources.
</requirements>

<instructions>
1. Role: Act as a Senior E-Commerce Analytics & Growth Engineering Lead providing an operational diagnostic protocol.
2. Instructions: Formulate a systematic root-cause diagnosis for an abrupt 18% checkout conversion drop over two weeks, prioritizing hypotheses by empirical probability and pairing each with a single definitive rule-out check.
3. Steps:
   Step 1: Categorize potential failure vectors into four core operational domains: Technical/Payment Gateway Failures, Analytics/Tracking Anomalies, User Experience/Frontend Code Changes, and Traffic Quality/Audience Mix Shifts.
   Step 2: Rank the specific failure causes across these domains from highest to lowest likelihood based on the suddenness and severity of an 18% drop over 14 days.
   Step 3: Define precisely one empirical rule-out check for each ranked cause, specifying the exact metric, log source, or telemetry comparison required to eliminate it.
   Step 4: Formulate the analysis in clear prose, highlighting why each cause holds its relative ranking and how its rule-out test operates.
4. End-goal: Deliver a focused diagnostic protocol that allows a response team to systematically test and rule out potential causes until the actual failure point is isolated.
5. Narrowing: Explicitly exclude untestable hypotheses, qualitative opinions, and vague multi-step troubleshooting guidelines. Focus exclusively on telemetry-backed, single-check verifications.
</instructions>

<output_format>
Present the diagnostic breakdown in clear prose structured into three dedicated sections:

1. Diagnostic Framework & Ranking Overview: A narrative explaining the taxonomy of sudden conversion drops and the rationale for the probability ranking.
2. Ranked Causes & Single Rule-Out Protocols: An itemized narrative detailing each testable cause in ranked order (Highest Likelihood to Lowest Likelihood). For each cause, provide:
   - Cause Name & Category
   - Mechanics of Impact (how it causes an 18% drop)
   - Single Rule-Out Check (the exact data query or test to invalidate it)
3. Operational Triage Order: A concluding summary outlining the logical sequence for running these checks to minimize time-to-resolution.

Skeleton:
- Section 1: Overview of the diagnostic strategy and ranking principles.
- Section 2: Ranked list of testable causes with individual descriptions and single rule-out checks.
- Section 3: Recommended execution workflow for data and engineering teams.
</output_format>

<examples>
Cause: Payment Gateway Authorization API Failures (Category: Technical)
Likelihood: High (1)
Impact Mechanics: Intermittent or complete API timeouts during payment submission cause users to abandon at the final step, resulting in a severe, immediate drop in completed orders.
Single Rule-Out Check: Compare payment gateway API error responses and success rates for the affected two-week period against the prior baseline period; if gateway error rates are unchanged and match baseline (<0.5% failure), rule this cause out.
</examples>

<verification>
- [ ] Output meets the primary success criterion stated in <objective>.
- [ ] Format matches the shape defined in <output_format>.
- [ ] Each <requirements> hard constraint is satisfied; no exclusions violated.
- [ ] Claims are supported by stated evidence, sources, or reasoning.
- [ ] Each claim is supported by stated evidence or sources
- [ ] Counterarguments and alternative explanations are addressed
- [ ] Conclusion follows logically from the stated premises
</verification>