Staff Business Data Analyst - Quality Management System

The Staff Business Data Analyst is a senior individual contributor responsible for the analytical engine of defect reduction across the Expert Network. You own the measurement system that decides what gets fixed: you design defect detection and quantification approaches, build and analyze defect funnels, isolate root causes in large operational datasets, size the customer impact of candidate fixes, and validate — statistically, not anecdotally — that improvement initiatives actually moved resolution, quality, and customer friction outcomes. This is an analytics role with a quality mission. You will apply descriptive, diagnostic, and causal analysis to one of the richest operational datasets anywhere — AI-scored conversation data covering ~100% of customer interactions — and translate it into a prioritized, defensible improvement roadmap. The Expert Network operates as a scaled contact center environment spanning multiple BPO vendor partner sites, and this role requires real fluency in how contact center operations run: queues, routing, transfers, handle patterns, and the operating rhythms that shape customer outcomes. Structured improvement methods (Lean, Six Sigma, Kaizen, root cause analysis) are tools in your kit for driving the fixes your analysis identifies; the analysis itself is the craft. You will operate as a single-threaded owner for specific defect domains and hold cross-functional partners — service delivery, training, partner management, product, and platform teams — accountable for their share of the quality outcome using evidence they cannot argue with. Responsibilities 1. Analytics Strategy & Defect Measurement Design the measurement approach for assigned defect domains: define the defect, the opportunity base, and the DPMO (Defects Per Million Opportunities) instrumentation so every defect rate ladders directly to a customer outcome metric Build and maintain defect funnels — quantifying defect volume, severity, and contribution to outcome degradation — and evolve the defect taxonomy as the operating environment changes (new categories, refined severity weights, expanded coverage) Prioritize the defect backlog analytically: rank initiatives by modeled customer impact, and defend the ranking with data when it is contested Contribute to the continuous improvement of the quality measurement architecture itself — ensuring we measure the right things, validating that our measures stay true, and pressure-testing quality markers against outcome data 2. Analysis, Root Cause & Outcome Validation Analyze resolution, quality, sentiment, and customer friction signals across large operational datasets (SQL and Python/R against conversation-level and journey-level data) to identify defects and isolate root causes — moving from symptom to verified cause, not plausible narrative Design the leading-indicator instrumentation beneath each outcome metric and monitor it for signal, drift, and regression Validate initiative impact with appropriate rigor: pre/post analysis with controls, cohort comparison, or causal methods where the stakes demand them — and say clearly when the needle did not move and why Partner with analytics and data science teams to evaluate, adopt, and improve AI-native quality measurement (conversation scoring, resolution scoring, customer distress detection) — including validating AI-scored metrics against human-labeled ground truth Design and operate human in the loop (HITL) quality processes: calibration sessions, human review sampling, and ground truth labeling workflows that keep AI scoring aligned to human judgment, with clear accuracy gates before any measure goes live 3. From Insight to Improvement — Driving the Fix Own the operating loop for assigned defect domains: defect detected → root cause verified → initiative launched → outcome validated → loop closed Deploy structured improvement methods — root cause analysis, Kaizen and rapid improvement events, value stream mapping, standard work — as the de