Staff Data Scientist - Paid Marketing Optimization

Responsibilities

• Design our Creative Analytics framework (creative performance modeling) - Own, design and develop our Creative Analytics Framework. Drive its findings into decisions with the creative and media-buying teams, and automate detection, reporting and action with MarTech. Longer term, build models that separate what a creative actually causes from what the algorithm’s delivery decides for it, so we can predict the incremental drivers of creative efficiency.
• Cohort maturity prediction (conversion-lag prediction) - Minimize steering delay: predict how many of today’s clicks will convert, and connect today’s spend to conversions that haven’t happened yet.
• Bidding signal optimization (value-based bidding) - Work with our channel managers and Applied AI team to improve the predicted conversion value we send to Google and Meta, iterating through which funnel events we optimise on and what predictions we attach to them. Predict too early and the signal is noisy, too late and the algorithm doesn’t get enough data to learn from.
• Audience optimization (uplift modeling) - Create customer uplift models to identify incremental value and build audiences that improve marketing efficiency. Tell the networks who to look for and who to skip, and feed that back as prospecting seeds, remarketing audiences and exclusion lists.
• Work closely with the Data Engineering and Applied AI team on scalable data pipelines, the evolution of our ETL processes, and the adoption of new tools and frameworks.

Requirements

• Python, SQL and a distributed framework like Spark; we use Databricks.
• Experience with marketing or attribution data, and a working understanding of how bidding platforms consume the signal you send them. You don't need to have run campaigns, but you should find how they behave genuinely interesting.
• Advanced modelling. Propensity models to value earlier funnel events and refine targeting, and uplift models to identify incremental value.
• Causal inference and experiment design. Separating what a channel drives from what it takes credit for, and designing the test that settles it when the platform controls delivery.
• The instinct to turn a business hypothesis into something testable, and the honesty to say when the data can't answer it.
• Comfort working across teams you don't control. This role runs on close collaboration with Applied AI, marketing measurement, MarTech and tracking, and CRM.
• You might come from marketing DS, ads ML, recommender systems or any applied ML role where your model’s output fed another algorithm.

Benefits

• An open, collaborative, dynamic and diverse culture;
• A generous monthly allowance for lessons on Preply.com, Learning & Development budget and time off for your self-development;
• A competitive financial package with equity, leave allowance and health insurance;
• Not in Barcelona? We offer an attractive relocation package to join us in our Preply Barcelona Hub
• Access to free mental health support platforms;
• Access to Gympass-partnered wellness and gym centers throughout Spain to promote and support well-being and physical health;
• The opportunity to unlock the potential of learners and tutors through language learning and teaching in 175 countries (and counting!).