PII Warning
Contextual safety education and protection of users in messaging
Year
2024
Role
Product Designer
Platform
Web, Mobile
Team
PM, engineers, designer
Work done
Designed real-time warning systems that detect when users share personally identifiable information (PII) in messaging conversations, providing contextual education and intervention to protect user privacy and safety.
Context
In marketplace messaging, users often share sensitive personal information like phone numbers, addresses, or financial details, making them vulnerable to fraud and privacy breaches. The PII Warning feature was designed to proactively identify these situations and provide immediate, contextual education about the risks. The challenge was to create warnings that felt helpful rather than intrusive, while effectively reducing risky behavior without hampering legitimate communication between buyers and sellers.
Real-time PII detection when a user is about to share personal identifiable information
Process
- Data analysis - Identifying patterns of PII sharing in user conversations
- Risk assessment - Understanding fraud vectors and privacy vulnerabilities
- UX writing - Crafting clear, non-alarming warning messages
- Interaction design - Designing contextual interventions and educational flows
- A/B testing - Optimizing warning timing and messaging for effectiveness
When a user receives a message containing PII, we also warn them
Key Features
- Real-time PII detection in text messages and images
- Contextual warnings with specific risk explanations
- Alternative communication suggestions
- Linked educational content about data privacy
- Escalation paths for reporting suspicious requests
On web and desktop devices
Outcome
- Significant reduction in PII sharing in high-risk conversations
- Decreased fraud reports related to information harvesting
- Improved user awareness of privacy risks
- Positive user feedback on protective messaging
- Enhanced platform trust and safety reputation