Uber Senior Strategy & Operations Manager Interview Experience
Uber · Senior · BizOps & Strategy
Lyft rejected me after the SQL take-home because I built this whole Python pipeline to pull from their locked spreadsheet and they were basically like, this could have been done with simpler query logic. Then at Uber, my roommate stopped me from over-indexing on PCA charts and told me, nobody knows what PCA is, Mark.
Interview process
I got reached out to because Uber Eats was ramping up and hiring a lot at this level, which definitely changed the feel of the process. The loop was pretty clean: recruiter screen, an Excel take-home, a hiring manager interview with a behavioral plus a live mini case, and then a final leadership panel where I had to present a full case. What stood out to me was how practical all of it was. Nothing felt random or academic. They wanted to see whether I could work through the kind of marketplace and operator problems they actually deal with, and in the final they really cared about whether I could communicate the so what to senior leaders.
Interview rounds · 4
- 1
Recruiter screen
BehavioralI got cold-reached by a recruiter because Uber Eats was ramping up hard, and after the usual resume and motivation questions he spent a big chunk of the call basically coaching me on what was coming next and how to prep for it.
Q1. Can you walk me through your resume and background?
How they answeredI walked through my background in analytics and strategy/ops type work and why that translated well to a role that sits between data, operators, and leadership. I framed myself as someone who can investigate what is going on in the business and then craft the story and recommendation off of it.
Q2. Why would you be interested in working at Uber?
How they answeredI said I was interested in the scope and pace, especially on the Eats side where the work is very operational and data-driven. The role felt like a fit because it mixes analytical problem solving with actually influencing decisions on things like growth, targeting, and spend.
- 2
Take-home assignment
AnalyticalData AnalysisThe Excel take-home was very practical and honestly pretty straightforward for me, mostly pivot-table-heavy with some SUMIFS and COUNTIFS, but you had to be careful with how the data was structured or you could get the wrong answer fast.
Q1. Using the order and merchant data, what is the average number of orders per eater overall, per member, and per non-member for the week of April 11 to April 17?
How they answeredThe tricky part was not just counting orders. I had to make sure I was getting the right eater counts in the time window, then back into which of those eaters were members versus non-members before calculating averages. I used lower and upper date bounds plus member-status logic. It was straightforward once I set the logic up correctly, but it was easy to mix up eater count with order count.
Q2. Using the merchant and order data, which cuisine types and restaurants are performing best?
How they answeredI handled most of it with pivot tables and some calculated fields, but I first checked whether merchant_id mapped one-to-one to cuisine type. That mattered a lot. Because it was one-to-one, I could use much simpler COUNTIF-style logic. If it had been one-to-many, I would have needed a more complex formula setup to account for merchants spanning multiple cuisine categories.
Q3. Please complete the full Excel analysis and submit your work product.
How they answeredI organized every question into its own tab to make it easy to review and added visuals even though they did not ask for them. My hindsight is that you can always make the workbook more communicable and cleaner. The analysis tool itself can get busy, so anything that ranks results clearly or makes the answer more readable helps a lot.
- 3
Phone screen
BehavioralAnalyticalExecutionThe hiring manager round was half behavioral and half live marketplace case, and the vibe was actually pretty laid-back even though he was clearly checking whether I really knew my own work and whether I could think in Uber's three-sided marketplace terms.
Q1. Tell me about an analytical project you worked on that had impact.
How they answeredI used a project where I could show real scale and quantified impact in roughly the $10M to $25M range. He let me talk through the analytical details for a bit, but he kept pushing on the so what. He wanted to know what changed in the business, where it stood now, and whether I could tie the work to an actual outcome instead of just a clean analysis. That felt very on-brand for them.
Follow-up questions- So what happened after that?
- Where does it stand now?
Q2. Restaurants are underperforming in a certain region. What is going on, and how would you look at it?
How they answeredI started with the KPIs I would diagnose first, then when he steered me toward vendor throughput I said radius was one lever I would test. From there he pushed me into guardrails versus hero metrics. I said I would watch whether sales and take improved, but also guardrails like late deliveries and throughput so we do not create a short-term lift that hurts reorder behavior and goodwill. He was very quick with the yes-and-then-what style follow-ups.
Follow-up questions- It looks like vendor throughput is the issue. What would you do?
- What metrics would you watch to make sure changing that does not hurt the business?
- 4
Final / onsite round
PresentationAnalyticalProduct StrategyEstimationData AnalysisExecutionThe final was an hour-long leadership panel where I presented a case to senior Eats leaders and got grilled live on my assumptions, segment choices, and pricing logic, so it felt a lot more like defending a real business recommendation than giving a polished monologue.
Q1. Uber Eats entered 'U City' two years ago, a competitor entered four years ago, and we still have only about 18% market share. Diagnose what is going on and what you would do to improve growth.
How they answeredI backsolved the market from the demographic data they gave me by making segment-level assumptions on order frequency, then worked back into TAM, SAM, and current penetration. I focused on price, merchant supply, and partnerships because the data set was vendor-heavy and I did not want to build a whole case around user behavior I could not observe. I also segmented merchants with clustering and highlighted targeted initiatives by segment instead of giving one blanket answer.
Follow-up questions- How did you size the opportunity from the demographic information we gave you?
- Why are you focusing on price, merchant supply, and partnerships?
Q2. You are recommending lower marketplace fees for EMTs. Wouldn't that cannibalize other businesses?
How they answeredI said that was a fair challenge, but in the data there was already variability in marketplace fees, and EMTs were still higher on average even after a targeted reduction. My view was that some cannibalization is unavoidable any time pricing moves, but the proposal was not broad-based discounting. It was a more targeted fee adjustment where the aggregate fee level still remained above smaller businesses, so I did not think it would cannibalize them to the same degree.
Tips from the candidate
I would prep for this by really knowing your own background cold and not sounding over-rehearsed. I got feedback in mocks that I was too scripted and too frameworky, and that is exactly the wrong energy here. For the case, focus on the business takeaway first and keep the super technical stuff in the appendix unless they ask. Also know how to talk about guardrail metrics in a three-sided marketplace, because they care a lot about whether your fix helps one side while quietly hurting another.
Company culture
My read was that Uber Eats is hiring pretty aggressively right now when the business is doing well, and that matters because volume hiring changes the process a bit. The recruiter was unusually helpful and basically told me how to think about the interviews, which felt like they wanted to move good candidates through fast. Even though Uber's reputation is very go-go-go, the actual interviewers were younger, low-key, and practical, not performatively intense. The intense part came from the content, not the personalities. They seem very focused on real business judgment, clear communication to leadership, and whether you can make an operational recommendation without losing sight of marketplace tradeoffs.