Uber Eats Growth Strategy

Context

Uber Eats was in hyper-growth mode in Australia and New Zealand, competing with Menulog and DoorDash. The strategy team was responsible for identifying growth opportunities across the three-sided marketplace: restaurants, consumers, and couriers.

Problem

Growth was slowing in mature cities while newer markets were still finding their footing. The team needed to identify which levers had the most impact on GMV, and sequence them for maximum compounding effect.

Approach

1. 1Built a marketplace diagnostic model to decompose GMV by supply density, consumer demand, and delivery reliability — identifying where each city was underperforming.
2. 2Ran supply expansion sprints in restaurants-light suburbs to improve cuisine coverage and reduce choice fatigue for consumers.
3. 3Designed A/B experiments to test pricing nudges, cart suggestions, and re-engagement messaging.
4. 4Created an ops playbook for launching new suburbs that could be replicated across markets.

What I shipped

• ✓Marketplace diagnostic model used by city teams to identify growth priorities
• ✓Restaurant acquisition playbook for supply-constrained suburbs
• ✓Experiment framework for consumer-side conversion tests with clear hypothesis and success metrics
• ✓Suburb launch playbook adopted by 3 additional ANZ markets

Impact

• ·Supply expansion in targeted suburbs improved estimated consumer choice metrics.
• ·Conversion experiments provided learnings that shaped the regional experiment roadmap.
• ·Ops playbooks reduced suburb launch time by standardising the activation checklist.

Learnings

• Marketplace thinking requires holding three customer types simultaneously. Optimising one side often hurts another — the diagnostic model forced us to think systemically.
• Ops playbooks compound. The first suburb launch was painful. The fifth was easy. Documentation and standardisation are leverage.

One key tradeoff

Metric definitions