Senior Business Analyst

• 4+ years in analytics or BI, with strong SQL (BigQuery or similar).
• Hands-on experience running an ETL/ELT platform. Keboola is ideal; dbt, Fivetran, Airflow or similar also count. You can build, schedule and debug a pipeline without waiting on engineering.
• Experience with customer event data: a CDP (Bloomreach/Exponea or equivalent), app/web tracking SDKs, UTM and attribution logic.
• A track record of solving cross-team data problems without formal authority. Bring examples of problems you chased to closure.
• Clear writing: you can explain a data gap, and what it costs, to a non-technical owner.
• AI tools are part of how you work every day.

• Nice to have: Tableau; experience with email/push channel data.

Groupon connects 42 million customers with local experiences - restaurants, events, wellness and travel - and the million-plus merchants who deliver them. Our mission is to get people offline and into the real world at prices that make it possible. We're an AI-native company in the middle of a platform transformation, moving from a deals marketplace to an experience discovery platform that works for customers and merchants at the same time.

Every Marketing decision at Groupon rests on customer journey data: what users do on web and app, how that's tracked, and how it flows through Keboola into Bloomreach (our CDP) and BigQuery. Right now, that flow has gaps. About 3% of traffic can't be attributed to a source, and nobody has a full end-to-end view of where the journey data breaks.

You're the person who fixes that. First you map the flow and find the breaks. Then you drive each fix to the finish, working with Engineering, CRM and Product, until the numbers are right. It suits someone who spots a problem, finds the person who can fix it, and follows up until it's done.

• Your work shows up in business decisions. You own the data Marketing uses to decide, not a report nobody reads.
• You build, not just request. You run the Keboola pipelines yourself: build them, schedule them, fix them.
• The mandate is clear. You know exactly what success looks like at 6 months (see below).
• The team works AI-first. You use AI in your day-to-day work, and you document data products so AI agents can use them as well as people.
• There's room to grow. You may take end-to-end ownership of the data for a whole business line, covering its models, pipelines and reporting.

Your first 6 months: the User Journey data flow

• Map the journey tracking end to end: web/app events, CDP cookie and identity, email/push send and click IDs, UTM and traffic-source classification, orders. For each step, show what's tracked, where it lands, who owns it and where the gaps are.
• Rank the gaps by business impact and drive them to closure with Engineering, CRM and Product. You raise the ticket, follow up, and confirm the fix in the data.
• Build monitoring that catches breaks before stakeholders do: data-quality checks and alerts on the journey datamarts and the pipelines behind them.
• The Keboola pipelines behind Marketing data: extractors, transformations and orchestrations. You keep the email/push and journey datamarts accurate and on schedule.
• The tracking and attribution backlog: unknown traffic source, event attribute mapping, affiliate/referrer logic and push notification journey gaps.
• CRM and lifecycle reporting: email/push dashboards, RFM segmentation and managed-channels views.
• Documentation of data products in the data catalog, so other people and AI agents can use them.

• You count a problem as solved when the fix shows in the data, not when the ticket is raised.
• You get things done through people who don't report to you, and you follow up without being chased.
• You'd rather find a break yourself than hear about it from a stakeholder.
• You build things the next person, or AI agent, can use without a walkthrough.

Groupon runs on five principles. Here's what they look like in this role:

• Extreme Ownership: a gap is closed when you've confirmed the fix, not when you've handed it off.
• Speed Over Comfort: you fix the failing job yourself instead of joining a queue.
• Impact Obsessed: you rank gaps by business cost, not by how easy they are to fix.
• Simplify to Scale: one journey map with a named owner for each step.
• Disciplined: automated checks, not manual spot-checks.

• The journey data-flow map is published, with a named owner for every step.
• Unknown traffic source is below 1% (about 3% as of Jul '26).
• The top 5 journey tracking gaps are closed or have a committed fix date.
• Automated checks are running on the journey datamarts and their pipelines.