Want to quantify real-world cross-price-elasticities? Estimate competitive intensity? Make millions of counterfactual predictions? And use all this to actively drive quantifiable, significant impact? Then come join our team! As Zalando’s algorithmic pricing team, we price 500k products across 20+ countries every day. To do so, we bring together deep-learning-based forecasting methods with rigorous econometric demand modeling and off-policy learning.
Pricing is a key lever to steer Zalando’s €15+bn in GMV, to improve our platform economics, and to better understand our and partners' competitive positioning in the dynamic fashion marketplace. Better pricing has significant economic impact, but through managing our seasonal inventory it also reduces waste and our environmental footprint. Here, you can make a real difference at the intersection of state-of-the-art econometrics and causal machine learning.
In this role, you will be a central thought leader in our understanding of demand systems: how does price affect demand? How do value-anchoring effects around discounts come into play? How can we estimate cross-price substitution in a computationally tractable way? Can we measure the role consumer consideration plays? What about competitors' prices? Can we model forward- and backward-looking consumer behavior? How can these insights inform our pricing algorithms, and what might they tell us about related topics such as assortment, marketing, and product ranking?
You will learn about our pricing systems and utilize our full dataset for high-exposure projects. You will be part of a cross-functional team, responsible for selecting the scientific approaches as well as building the systems for pricing science. Your colleagues will include PhD economists, former professors, and machine learning experts. You can grow your career through hands-on implementation of the latest ideas in the literature, and hard-and-fast A/B-tested evidence of your contributions. You’ll get to work and collaborate alongside senior Zalando economists (including Justin Rao and Greg Crawford). You will have access to Zalando’s comprehensive data sources, computational infrastructure, and software tooling.
Help us build a system that uses off-policy experimentation, automated causal learning, demand modeling, and continuous measurement to increase market efficiency and reduce waste.
WHERE YOUR EXPERTISE IS NEEDED
- Help us reach the frontier of demand estimation: coming from our existing and validated own-price elasticities, we want to estimate more complete models of demand that include substitution, consideration, display, and competition.
- We have terabytes worth of observational and experimental data on purchase histories, clickstreams, and product similarity scores, amongst others. You will use these sources as needed, and propose efficient experimentation strategies to generate interventional data where necessary.
- Influence neighboring teams in algorithmic pricing and beyond to allow your breakthroughs in demand estimation to impact Zalando’s pricing and platform algorithms end-to-end.
- As an evangelist of econometric methods, you will collaborate with economists and machine learning experts to build the most data-efficient, impact-oriented pricing solutions possible.
WHAT WE’RE LOOKING FOR
- Excellent educational background (PhD, or M.Sc. with relevant professional experience) in economics or marketing, ideally specialized in structural empirical industrial organization or quantitative marketing.
- Strong background in demand estimation. Ideal candidates have experience building and estimating custom models with large-scale, real-world datasets, as well as making the tradeoffs associated with alternative methodological approaches given the available data.
- Experience driving product impact: You are able to maintain a medium-term research roadmap while regularly delivering incremental improvements to our systems. You also speak fluent “business” and can work with senior stakeholders.
- Scientific pragmatism: good judgment for the right level of rigor to deliver results. You know when in-depth modeling pays off, and when pasting aggregated data into a spreadsheet will do the trick.
- The ability to implement your models in Python or R, and to wrangle terabytes of data using SQL and/or spark on large, complex datasets.
We encourage you to apply even if you do not meet every single qualification: some strong candidates may lack parts of the listed qualifications. Research shows that underrepresented groups are more prone to experiencing imposter syndrome and doubting the strength of their candidacy, so we urge you not to exclude yourself prematurely and to submit an application if you're interested in this work.