Data Scientist (Contract), Manchester

Are you passionate about uncovering the hidden drivers behind business performance? We're looking for an experienced Data Scientist Contractor to help us identify what makes the highest-performing products in our portfolio stand out and translate those insights into actionable business decisions.

You'll work on a high-impact project focused on understanding the characteristics of successful products across a large and diverse portfolio. Building on existing analysis, you'll validate current assumptions, apply advanced machine learning techniques, and develop a robust framework that explains the key drivers of product performance. This is a highly visible piece of work that will influence strategic decision-making across the business and provide direct exposure to senior leadership stakeholders.

• Validating and enhancing existing analytical models and hypotheses.
• Applying unsupervised learning techniques to segment and cluster product data.
• Building supervised learning models to identify and predict high-performing products.
• Using model explainability techniques such as SHAP to uncover and communicate key performance drivers.
• Turning complex analytical findings into clear, compelling narratives for both technical and non-technical audiences.
• Developing scalable, programmatic solutions using data housed within Snowflake.
• Creating engaging stakeholder-facing outputs, ideally leveraging Streamlit.

• Proven commercial experience delivering Data Science projects from problem definition through to insight generation.
• Strong expertise in both supervised and unsupervised machine learning.
• Experience with model interpretability techniques, particularly SHAP.
• Excellent stakeholder communication and storytelling skills.
• Strong Python and SQL capabilities.
• Experience working with Snowflake and ideally Streamlit.
• Ability to challenge assumptions, validate existing work, and build trusted recommendations.

We are shortlisting w/c 1st June with the ideal start within the next 2 weeks. If you enjoy turning complex data into clear business value and want to help define what ''great'' looks like across a large product portfolio, we'd love to hear from you.