DCX Product Analytics Analyst; mwd

Position: DCX Product Analytics Analyst (mwd)
Location: Zürich

Company

Marsh

Description

The Product Analytics Analyst will partner with brokerage client executives and product teams to deliver broker‑facing insights that improve renewal placement outcomes for clients. You will run hands‑on analysis, build and validate models, and prototype dashboards that inform broker decisions during renewals. You will translate broker workflows into analytics requirements, produce actionable deliverables, and work closely with engineers to ope rationalise models and measurement.

Primary

Responsibilities

• Engage with brokers and product stakeholders to capture renewal pain points and translate them into clear analytics problems and success criteria.
• Prepare clean and feature‑engineer data from existing systems and processes (policies, claims, quotes, broker notes, rating history) and third‑party sources.
• Build, evaluate and validate models that produce required insights using appropriate techniques.
• Produce reproducible analysis notebooks and model artefacts (feature definitions, training pipelines, validation results).
• Prototype broker‑facing dashboards and report mock‑ups that translate model outputs into actionable broker guidance.
• Work with engineering/ML‑ops to define data contracts, API specifications and acceptance criteria for operational models; support handover and testing.
• Document model assumptions, data lineage and run bias/fairness checks in line with governance.

Expected Outputs and Deliverables

• Models that provide insights to support the renewal discussion.
• Prototype dashboards and wireframes for broker workbench (MVP/iterative versions).
• Feature and data dictionaries, ETL specification notes and examples of SQL queries.
• Playbooks and short how‑to guides for brokers to act on model‑driven recommendations.

Collaboration & Stakeholder Interactions (Day‑to‑Day)

• Brokers – run discovery sessions, pilot dashboards, gather feedback, iterate content and format; occasionally join broker renewal calls to observe workflows.
• Engineering – write clear acceptance criteria, support UAT, review deployment steps and monitor production behaviour.
• BI/UX – partner on dashboard design, data visualisations and ensuring insights are interpretable for non‑technical users.
• Risk/Governance – provide documentation and respond to model governance queries; follow privacy and data access policies.

Required Tools, Technologies and Technical Proficiencies (Levels)

• Python – Intermediate to Advanced (pandas, scikit‑learn, XGBoost/Light
  
  GBM; testable scripting and notebooks).
• Statistical Modelling – Intermediate (classification/regression, feature engineering, cross‑validation, calibration).
• BI & Visualization – Intermediate (Looker/ Tableau/ Power BI: prototype dashboards and deliverable‑ready visualisations).
• Data Warehousing – Familiar to Intermediate (Snowflake / Big Query / Redshift; understand schemas, partitioning).
• ETL / Transformation – Familiar (dbt desirable; ability to author and review SQL‑based transformations).
• MLOps Exposure – Familiar (experience packaging models, basic CI/CD, model monitoring concepts).

Necessary Skills,

Education and Experience Technical Skills

• Python scripting & data science libraries.
• Data visualisation experience.
• Core statistical understanding.
• Familiarity with cloud data warehouses and ETL patterns.
• Exposure to MLOps concepts (versioning, monitoring) and Git.

Business & Interpersonal Skills

• Strong stakeholder management and communication; ability to translate technical results into actionable broker guidance.
• Product‑mindful – ability to scope MVPs and prioritise features for adoption.
• Commercial awareness of insurance renewal dynamics and placement outcomes.

Education

• Required – Bachelors degree in a quantitative or analytical discipline (e.g., Statistics, Mathematics, Computer Science, Economics, Engineering) OR equivalent practical experience.
• Preferred – Masters degree in a quantitative field.

Experience

• Typical – 26 years in analytics/data science roles with demonstrable hands‑on modelling experience.
• Desirable – 13 years exposure to insurance/financial services or broker workflows; experience preparing models for production environments.

Marsh McLennan is committed to creating a…