Product Data Analyst

Turn product data into better decisions, experiences, and measurable customer impact.

Smartly is an AI-powered advertising platform uniting creative and media workflows. We are looking for a Product Data Analyst who helps product teams understand how customers use the platform, choose investments, and measure whether launches create value.

About the role

Reporting to the Manager, Data Platform, you will be a hands-on contributor in the Data Analytics squad, partnering closely with Product, Design, Engineering, and Data Engineering across the company.

This role combines analytics engineering with product analytics. You will write code, build production-grade models and trusted data products, then use them to answer product questions and shape decisions for teams across Smartly. Dashboards are a delivery surface, not the goal.

Smartly moves fast. You will turn ambiguity into a plan, deliver in valuable increments, make sound technical choices, surface blockers early, and drive work through adoption. This is not a coordination or reporting-production role.

What you will own

• Build production-quality dbt models in BigQuery, from source-aligned layers to reusable product marts and metric foundations.
• Design tables deliberately, defining grain, keys, joins, data types, null handling, history, and performance before implementation.
• Write clean, maintainable code using version control, pull requests, automated tests, documentation, and peer review.
• Turn recurring questions and one-off reports into durable models, shared definitions, and self-service data products.
• Partner with Product Managers and Engineers on instrumentation, event schemas, data contracts, validation, and quality monitoring.
• Own domain data quality by tracing discrepancies, lineage, and join coverage; fix root causes before users find them and expose reliability.
• Analyse journeys, funnels, cohorts, retention, adoption, experiments, and commercial outcomes to recommend next steps.
• Build decision-ready dashboards when appropriate, keeping trusted modelling and metric logic beneath the visualisation.
• Drive work end to end: clarify outcomes, scope pragmatically, ship iteratively, communicate directly, and ensure adoption.
• Use AI tools to increase speed while protecting confidential data, reviewing generated code, and verifying conclusions.

What success looks like

• Within three months, you understand the product and architecture, contribute reviewed code regularly, and ship a trusted model or improvement.
• Within six months, you own a product area’s analytics foundations: tested, documented models actively used in decisions and trusted by product teams.
• You replace recurring manual analysis and reporting with reusable models, reliable metrics, and effective self-service.