Many people have experienced the frustration of waiting for someone who says they'll arrive in "15 minutes," only to show up much later. This common issue stems from optimism bias, poor traffic awareness, or social politeness, leading to inefficiencies in planning and minor social friction. While navigation apps already provide accurate travel time estimates, they don’t account for individuals' tendencies to misestimate their arrival. There’s a clear opportunity to personalize these estimates for better coordination.

How it Would Work

One approach could involve creating a system that automatically adjusts arrival estimates shared in messages or via live location-sharing apps. For instance:

• Detecting Estimates: The system could parse messages looking for arrival time phrases (like "be there in 20 minutes") or access shared live locations from apps like Google Maps.
• Personalized Adjustment: Using historical data, it could apply correction factors—if someone typically takes twice as long as they say, future estimates would be adjusted accordingly.
• Real-Time Refinement: Current traffic data could further refine these estimates.
• Passive Notifications: The recipient could get a subtle alert clarifying delays, like "Alex’s ‘15 minutes’ is likely 28 minutes due to traffic and their usual delay pattern."

Potential Applications

This could benefit social planners, professionals coordinating meetings, and service providers like therapists managing client arrivals. Messaging platforms might integrate it as a utility feature (increasing engagement), while map services could enhance accuracy through partnerships. A low-friction version—like a mobile app scanning messages for time estimates—could be initially tested before expanding via APIs.

Standing Out

Unlike basic GPS-based ETA sharing, this system could adapt to individual patterns, making it stickier over time. Privacy concerns could be addressed by processing data locally. Existing tools like Waze Carpool focus on ride-sharing rather than social planning, leaving room for a solution tailored to informal coordination.

Testing assumptions—like whether people’s ETA habits are consistent—and iterating on privacy-first integrations could help refine the approach. Over time, features like group delay predictions (e.g., "Sarah is always late when meeting Alex") could add further value.