Lessons from scaling Uber and Opendoor | Brian Tolkin (Head of Product at Opendoor, ex-Uber)

Episode Summary

Executive Summary: Brian Tolkien, head of product and design at Opendoor, explains how deep ops experience made him a stronger product leader: it built customer closeness, operational intuition, and respect for the interplay between product and operations. He shares lessons from Uber and Opendoor on scaling, experimentation, reviews, jobs-to-be-done, and staying calm under pressure.

Main Topics: Product + operations as a complementary system (Priority: 5/5): Brian argues that ops and product should work in harmony: ops provides ground truth and speed, while product/engineering provides scale and automation. He uses Uber and Opendoor to show why both functions are necessary in ops-heavy businesses. How ops experience improves product leadership (Priority: 5/5): Starting in ops gave Brian a deep understanding of how the business works, how customers behave, and what actually moves metrics. That foundation helps him identify the right technology to build later. Scaling from manual processes to software (Priority: 5/5): He describes how early operational work at Uber evolved from one-on-one onboarding and human-in-the-loop surge pricing into automated systems, showing the transition from local iteration to technology leverage. Product reviews and leadership style (Priority: 4/5): Brian explains how he structures product reviews to be useful rather than adversarial: small groups, clear goals, context-rich docs, and a probing leadership style that avoids turning reviews into a firing squad. Jobs to be done at Opendoor (Priority: 4/5): He uses jobs-to-be-done as a practical framework to force empathy and focus on the customer’s real context, especially in a business with infrequent, high-stakes transactions like home buying and selling. Experimentation under low volume constraints (Priority: 4/5): Brian discusses how to run A/B tests and make decisions when sample sizes are too small for clean statistical significance, including power analysis, longer experiments, alternative methods, and judgment. Competition, stress, and resilience (Priority: 4/5): He shares stories about Zillow, early Uber chaos, and launching in China to illustrate staying competition-aware but customer-focused, and remaining calm under pressure to avoid spreading stress to teams.

Key Arguments: Operations experience creates product intuition because it exposes leaders to the real mechanics of the business, customer pain points, and the daily signals that move outcomes. Product and ops are not in competition; they should function like a twin-turbine jet plane, with each side adding different strengths and enabling better outcomes together. In early-stage ops-heavy businesses, the highest-leverage engineering work should focus on the core economic and operational bottlenecks, not on peripheral improvements. As the business scales, manual operations should be progressively replaced by technology, freeing ops teams to tackle the next challenge instead of disappearing. Product reviews should primarily help teams make the product better, not feel like performance evaluations or criticism sessions. Jobs-to-be-done is useful because it forces teams to think from the customer’s context and understand the broader problem, not just the feature request. In low-sample environments, experimentation still matters, but teams must be honest about power, runtime, and when intuition plus supporting evidence is the better decision process. Staying calm under pressure improves outcomes because reflected stress tightens teams and reduces clarity rather than improving execution. Competition should be treated as a reality, but not the center of strategy; the best defense is doing the job-to-be-done better than anyone else. Vertical integration is essential in businesses like Opendoor because pricing, risk, capital markets, product, and operations all have to work together.

Data Points: Uber employee number: 100 - Brian joined Uber as employee 100 before UberX or UberPool existed. Duration at Uber: nearly 5 years - Brian spent almost five years at Uber before moving to Opendoor. Product review size guideline: under 10 people - He prefers keeping live product review conversations relatively small for quality discussion. Product review cadence at Opendoor: 2 slots per week - Teams can sign up for two product review slots each week. Real estate sale frequency: once every 7 years - Brian cites this as the average frequency of home buying/selling, making the product less intuitive for builders. Confidence threshold in experiments: 80% - He mentions sometimes accepting 80% confidence instead of 95% when trade-offs justify it. China launch sleep: 30 minutes - He slept about 30 minutes during the Uber Pool launch in Chengdu. Uber Pool launch prep time: 6:00 a.m. - The team aimed to launch Uber Pool in Chengdu in time for the morning rush hour. Product review audience cap: 10 - He says the best product review conversations happen when attendance stays relatively small. Early driver onboarding: 90 minutes to 2 hours - Uber Black drivers were initially onboarded individually in person for this length of time. Early manual surge windows: 7 p.m. to 3 a.m. - GMs manually controlled surge pricing activation windows in the early days. Uber launch onboarding volume: 1,000 a week - He describes the point at which onboarding volume became too high for manual processes. Uber interview date: July 4 - Brian accidentally scheduled his Uber onsite interview on July 4th.

Pivotal Quotes: "“I think Uber always had this mentality, and Opendoor does too, of kind of like a twin turbine jet plane, where you can fly the plane on one engine for a little bit if you need to, but it's operating most efficiently and effectively if both are working together.”" — Brian Tolkien: Explaining how product and operations should complement each other rather than compete. "“When you reflect the stress onto your teams, everybody tenses up. It counterintuitively doesn't produce better outcomes.”" — Brian Tolkien: Describing why calm leadership matters during intense execution moments. "“The team brings the depth and you bring the breadth.”" — Brian Tolkien: His view of senior leaders probing teams while respecting their expertise and context.

Implications: For product leaders, ops experience is a strategic advantage in complex businesses. The best teams build strong feedback loops, automate the right bottlenecks, and stay customer-focused while using data, experimentation, and judgment together.