SI 425: Introduction to User Modeling and Uber Case Study Strategic Analysis

Course Logistics and Assignment 1

• Class Activity: Name Tags: Students are instructed to create personalized name tags using provided materials (passed out by Ilana). These are intended to help the instructor learn everyone's names and are the students’ personal property to keep for the semester.
• Group Work:
  • Students continue working with the groups formed in the previous class, although these are not permanent for the entire semester.
  • New students who were absent on Tuesday are grouped together.
• Assignment 1 Details:
  • Deadline: Due next Thursday by midnight through Canvas.
  • Scope: An individual assignment to kick off the semester and review various topics.
  • Focus: The assignment centers on Uber and user modeling, exploring pervasive questions applicable to all platforms.
  • Deliverable: Students must pick one idea for increasing Uber's active users. This idea can be their own, a peer's idea (inspired by class discussion), or an AI-generated idea.
• Class Deliverable: By the end of class, groups must submit three good ideas for increasing Uber's active user base via a provided Google Form.

Policy on AI Usage (SI 425)

• Consulting AI: It is not a violation of SI 425 policy to consult with AI tools like ChatGPT or Gemini for Assignment 1 brainstorming.
• Documenting Usage: If AI is used, students must document their usage, including the specific prompts asked and the responses received.
• Curating vs. Copying: Students are expected to curate AI-generated content rather than blindly copying it. Users must use ‘guardrails’ similar to those used in a professional job environment.
• In-Class Work: AI usage is permitted for assignments, but during in-class discussions, the focus is on peer-to-peer interaction.

Uber's Competitive Landscape

• Direct Rideshare Competitors: Lyft (United States), Didi (China).
• Robotaxi Services: Waymo.
• Traditional Transportation: Taxi cabs and private driving services.
• Car Rental Services: Zipcar, traditional rental companies.
  • Zipcar Experience: Described as having frictions such as specific parking locations and the requirement to return the vehicle to the original spot, costing users time and convenience.
  • Uber's Response: Global rental car options were integrated into the Uber app within the last year to compete directly with this market.
• Alternative and Public Transit:
  • Bikes: Capital Bike (DC) and other bike rideshare services.
  • Buses: Charter buses, public transit, and bus rentals.
  • Walking: Identified as a basic competitor for short distances ($10$ to $35$ minutes).
• Logistics and Delivery: DoorDash, Instacart (competing with Uber Eats).
  • Resource Maximization: Uber Eats allows Uber to match user needs with providers, using existing resources (drivers) to deliver products instead of just people.
• Private/Social Solutions: Personal car ownership and "designated driver" friends.
  • Car Ownership Trends: In some areas, ownership is projected to decline as younger generations may find it cheaper to use ridesharing for limited trips rather than paying for parking and maintenance in big cities.

Markets and Operational Challenges

• Geographic Reach: Currently dominant in urban and suburban areas.
• Underserved Areas: Rural areas are often neglected because demand is unpredictable for drivers.
• Banned Regions: Uber is banned in countries like Ireland to protect taxi drivers from "unfair competition."
• Regulatory History: Uber originally started as "UberCab" but had to drop "Cab" from its name due to legal litigation.
• International Failures (China/Didi): Uber exited China after a costly subsidy war with Didi. Fraudulent activity occurred where drivers used multiple phones to pretend to be riders to cash in on incentives.
• Driver Behavior: Drivers congregate at airports and major venues (concerts, games) because the demand is predictable and high.

Pricing Models and Market Segmentation

• Dynamic Pricing: A model that adjusts prices based on real-time market conditions—specifically time, supply (drivers), and demand (riders).
• Surge Pricing: High-demand spikes ($5$ PM or $9$ AM) can lead to significantly higher fares (e.g., a $2$-mile ride for $50$ dollars).
  • UX Change: Uber used to explicitly show a multiplier (e.g., "$3 ×$ Higher"). They found users were more likely to book if they simply saw the final high number rather than the explicit surge alert.
• Pre-Booking: Allows users to lock in a price up to $30$ days in advance. While the price is often higher than a $5$ AM booking, it provides "peace of mind."
• Market Segmentation (Price Discrimination): Uber offers different tiers to capture varied "willingness to pay":
  • UberX Share: For price-sensitive users willing to wait longer or walk to stops.
  • UberX: The standard service.
  • Uber Comfort: A balance between luxury and standard.
  • Uber Black: High-end luxury.
• Terminal-Specific Pricing: Data suggests Uber may offer different quotes depending on which airport terminal a user is at (e.g., McNamara Terminal/Delta vs. Evans Terminal/Spirit at Detroit Metro Airport), potentially based on the perceived willingness to pay associated with different airlines.

UX Innovations and Feature Proposals

• Existing/Emerging Features:
  • Safety: Features to request drivers of a specific gender identity.
  • Niche Services: Uber Pets (matching pet-friendly riders with drivers who aren't allergic).
  • Uber Carrier/Courier: Available in markets like New York for delivering items (like a chair) without the sender riding along.
  • Kiosks: Physical terminals at airports for users without functioning phones or apps.
• Proposed Innovations:
  • Uber Party Bus: On-demand buses for groups of $10$ to $12$ people, featuring music and lights, operating for late-night entertainment.
  • Amazon Integration: Deep-linking Uber drivers to Amazon package delivery routes to solve the "last-mile" delivery challenge.
  • Accessibility for Non-Digital Users: A "call-ahead" phone service with a simple interface for older adults or those without reliable cellular connectivity.
  • Payment Flexibility: Options for cash payments or alternative digital wallets for unbanked users.

Discussion: Ethics and Social Dynamics

• Tips: Originally, Uber had no tips. They were added to compete with Lyft.
  • Arguments for Tips: Incentive for high performance and better service.
  • Arguments against Tips: They allow the company to pay drivers a lower fixed rate, shifting the wage burden to the consumer.
• Ratings:
  • Both drivers and riders are rated.
  • Rating Inflation: The norm has evolved toward giving $5$ stars unless something goes catastrophically wrong, because a driver's livelihood (avoiding deactivation) depends on high scores.
• Strategic Ethics: Discussion of working with governments to establish cellular networks in exchange for anti-competition contracts, and the balance between corporate profit and the "greater good" for global connectivity.

Questions & Discussion

• Student Question: Is it in violation of SI $425$ policies to consult with ChatGPT or Gemini?
  • Professor Response: No, it is not a violation as long as you document your usage, curation, and the prompts used. You should use it as a professional tool with appropriate guardrails.
• Student Question: Why was Zipcar a bad experience?
  • Student/Professor Response: The friction of locating vehicles in specific spots and the requirement to return them to that same spot makes it less convenient than the door-to-door service of Uber.
• Student Question: Does Uber charge more at different airport terminals?
  • Professor Response: Experience and articles suggest they might. For instance, at Detroit Metro (DTW), pricing might differ between the North (Evans) and McNamara terminals based on location data and user profiles.
• Student (Team Viarooters): Mentioned bike rideshare services and private drivers as competitors.
• Student (Team 4 Question Marks): Mentioned taxi cabs as competitors.
• Student (Team Strawberry): Discussed walking as a transportation competitor for short durations like $10$ minutes.