← Back

Data Scientist – Attribution

Berlin

Role Description

The data science team’s mission at SIGNA Sports United is to provide data-driven solutions to improve the efficiency and decision-making across our group of companies in different areas including pricing, marketing, and demand forecasting.
As Data Scientist for marketing attribution modeling, you will partner with marketing managers and use logs and telemetry from our websites to efficiently allocate marketing resources. The position requires a sound knowledge of statistics to properly apply empirical models for marketing attribution and communicate the results to non-technical audiences across our group of companies.

- Collaborate with stakeholders to refine a vague problem statement and formulate the right question.
- Design, implement, and carry out the statistical analysis to provide insights for decision making related to marketing.
- Summarize and communicate to technical and non-technical audiences the results from the data analysis and the underlying assumptions.
- Write reproducible data analysis using large databases from our business units across Europe.
- Build data pipelines

Requirements

Basic Qualifications
- M.Sc in a quantitative field (applied mathematics, statistics, econometrics, computer science, machine learning, data science or equivalent) or related experience
- Sound knowledge of statistics, and in particular knowledge of probabilistic modeling (Hidden Markov Models, Monte Carlo methods, etc.)
- Knowledge of at least one modern open-source programming/scripting language for data science such as Python or R.
- Ability to communicate results clearly to both colleagues and peers on the team as well as less technically versed audiences
- Independent, driven and looking to make an impact using data-driven solutions

Preferred Qualifications
- 3+ years of professional experience working in data science in e-retail and marketing attribution
- Skilled with specific libraries for data analysis and visualizations including jupyter, pandas, numpy, scikit-learn, dplyr, parallel, etc.
- Experience applying and estimating deep neural networks (e.g. LSTM, GRU, etc.)