Job Description

Uber partners with hotels and airlines -- including Marriott, Delta, Air France, and more -- to reward customers with hotel and airline loyalty points when they use Uber. These are marquee deals for Uber, but the economics of the deal rely on a strong understanding of deal value and incrementality. This role will drive the experiments and measurement for all of our loyalty deals. What Youll Do: Design and run experiments to measure incrementality of loyalty deals Run regressions to estimate impact of the deals using methods like two-stage least squares or instrumental variables Work with stakeholders to understand deal economics and provide recommendation on how to optimize deal structure Refine ambiguous questions and generate new hypotheses about the product through a deep understanding of the data, our customers, and our business Design experiments and interpret the results to draw detailed and impactful conclusions. Define how our teams measure success, by developing Key Performance Indicators and other users/business metrics, in close partnership with Product and other subject areas such as engineering, operations and marketing Collaborate with applied scientists and engineers to build and improve on the availability, integrity, accuracy, and reliability of data logging and data pipelines. Develop data-driven business insights and work with cross-functional partners to find opportunities and recommend prioritisation of product, growth, and optimisation initiatives. What Youll Need: Undergraduate and/or graduate degree in Math, Economics, Statistics, Engineering, Computer Science, or other quantitative fields. 10+ years experience as a Scientist Excellent understanding of statistical principles backed by an academic foundation Advanced SQL expertise Experience with either Python or R for data analysis Significant experience in setting up and evaluation of complex experiments Proven track record to wrangle large datasets, extract insights from data, and summarise learnings/takeaways. Experience with Excel and some dashboarding/data visualisation (ie Tableau, Mixpanel, Looker, or similar)