Expedia, Inc. Data Scientist Interview Guide

Our candidates report that Expedia’s Data Scientist interviews, especially on the advertising side, reward people who can connect analytics to business trade-offs. The standout case about choosing between a more relevant hotel and a higher-bidding hotel shows that they care about relevance versus revenue, not just whether you can name the right metric. We’ve also seen questions around banner ad strategy, best-performing advertisers, and increasing search ads, which suggests the team wants candidates who can reason in marketplace terms and explain why one decision helps the product more than another.

A recurring theme is that Expedia leans heavily on practical communication and product judgment. Multiple candidates described behavioral questions about failure, recovery, and how to present technical findings to non-technical stakeholders, and those answers seemed to matter as much as the technical ones. In the more technical conversations, SQL depth came through clearly — subqueries, window functions like LAG, and careful query construction — while Python coding was minimal. That pattern tells us they are screening for analysts who can work independently with data, not just people who can solve abstract coding puzzles.

We’ve also noticed that the machine learning discussion stays conceptual rather than overly theoretical, with topics like bagging vs. boosting, XGBoost vs. Random Forest, and bias-variance tradeoffs appearing alongside CV project deep-dives. The non-obvious differentiator here is domain fluency in ads and experimentation: candidates who can talk through auction logic, dashboard interpretation, and experiment validity tend to sound much more credible. In short, Expedia seems to value someone who can move comfortably between SQL, product thinking, and ad-tech intuition without losing the business context.