Back to blog
4 min read

[PM] Can we make Yelp better?

product managementyelpmobileux

For this post, I'm going to discuss a feature I'd like to be added on the Android version of Yelp. From what I understand, the Android team on Yelp is short-staffed so updates are rolled out somewhat slowly.

Disclosure: I'm not affiliated with Yelp.

BIG SIDE NOTE: This is not how a PM process works. Real product management involves extensive user research, data analysis, stakeholder alignment, and iterative testing. This is more of a thought exercise on potential improvements.

Current State of Yelp Android

Yelp's Android app serves its core purpose well - helping users discover and review local businesses. However, there are several areas where the user experience could be significantly improved.

Key Pain Points

1. Discovery Friction

Finding the right restaurant or business can be cumbersome, especially when you're looking for something specific but don't know exactly what you want.

2. Limited Filtering Options

The current filtering system is basic and doesn't account for nuanced preferences that users might have.

3. Social Features Underutilized

Yelp has a wealth of social data but doesn't leverage it effectively to improve recommendations.

4. Offline Functionality

Limited offline capabilities make the app less useful when you have poor connectivity.

Proposed Feature: Smart Recommendations

The Problem

Users often know what type of experience they want but struggle to find the right place. Current search requires knowing specific cuisine types or business names.

The Solution

Implement a "Smart Recommendations" feature that uses:

  • Time of day: Suggest coffee shops in the morning, bars in the evening
  • Weather data: Recommend indoor activities when it's raining
  • User history: Learn from past visits and ratings
  • Social signals: Factor in what friends have liked
  • Location context: Different suggestions for business districts vs. residential areas

Implementation Details

User Interface

  • Quick suggestion cards on the home screen
  • "Feeling lucky" button for spontaneous discovery
  • Mood-based categories: "Quick bite," "Date night," "Family friendly"

Backend Logic

  • Machine learning model trained on user behavior patterns
  • Real-time data integration (weather, traffic, events)
  • Collaborative filtering based on similar users

Success Metrics

  • Engagement: Time spent in app, number of searches
  • Conversion: From suggestion to business visit
  • User satisfaction: Ratings of recommended places
  • Retention: Return usage of the feature

Additional Improvements

Enhanced Social Features

  • Friend activity feed: See where friends have been recently
  • Group planning: Collaborative decision-making for group outings
  • Check-in sharing: Easy sharing to social media

Better Business Information

  • Real-time wait times: Integration with restaurant systems
  • Menu integration: Photos and pricing information
  • Availability indicators: Open/closed status with more granularity

Improved Search

  • Natural language processing: "Good sushi near me for under $30"
  • Visual search: Search by photos of food or ambiance
  • Voice search optimization: Better voice command recognition

Technical Considerations

Performance

  • Caching strategies for offline functionality
  • Progressive loading for better perceived performance
  • Battery optimization for location-based features

Privacy

  • Granular privacy controls for location and social data
  • Transparent data usage explanations
  • Opt-in recommendations rather than default tracking

Competitive Analysis

What Others Do Well

  • Google Maps: Excellent integration with other Google services
  • Foursquare: Strong recommendation algorithm
  • OpenTable: Seamless reservation integration

Yelp's Advantages

  • Review quality: More detailed, authentic reviews
  • Local business coverage: Comprehensive database
  • Community trust: Established user base and credibility

Implementation Roadmap

Phase 1: Data Foundation (Months 1-2)

  • Implement enhanced user behavior tracking
  • Build recommendation algorithm infrastructure
  • A/B test basic suggestion features

Phase 2: Core Features (Months 3-4)

  • Launch smart recommendations
  • Implement mood-based categories
  • Add social signal integration

Phase 3: Advanced Features (Months 5-6)

  • Natural language search
  • Real-time business data integration
  • Enhanced social features

Conclusion

Yelp has the foundation to become the definitive local discovery platform. By leveraging its rich data and focusing on intelligent recommendations, it could significantly improve user experience and engagement.

The key is balancing sophistication with simplicity - making the app smarter without making it more complex for users. The goal should be to help users discover great local experiences with minimal friction.