Designing for Trust in AI

Fostering Trust in AI: Driving Revenue, Loyalty, and Adoption

Role: UX Manager

Note: Visuals and Artifacts from this project are proprietary and cannot be shared.

Situation:

The company's push for AI tools faced resistance as users hesitated to accept AI tools, services, and recommendations, resulting in low adoption rates for new features.

The Goal:

Increase the adoption of AI tools and services by overcoming user hesitancy through building trust.

The Approach:

The product team hypothesized that user trust is the key to bridging the gap between beneficial AI services and user hesitancy. To validate this hypothesis we conducted the following steps:

• Defined Trust in a manner which was actionable to our product teams.

• Explored what Trust means to our users in the context of our product.

• Developed a means to measure and track user Trust.

• Created guidelines and principles for enhancing user Trust through product experiences.

• Audited key user journeys to identify opportunities for enhancing trust across the product.

• Validated the Trust approach through a pilot product launch to confirm our hypothesis.

How it was done:

I led a cross-functional team of UX Designers, Researchers, and Content Strategists through a comprehensive process including; Discovery, Definition, Ideation, Prototyping, and Validation.

Research Methods:

• Qualitative Research: Conducted 75 in-depth interviews with users to comprehend their perceptions of trust.

• Thematic Analysis: Identified a framework to measure user trust and tangible product-level attributes defining trust from the user's perspective.

• Quantitative Research: Conducted a global survey of 1200 users to validate the Trust framework, establish a baseline score, and determine which attributes most influence the perception of Trust.

Guidelines and Principles:

Developed Trust Principles for product teams and accompanying Design and Content Guidelines to ensure product teams build trust-enhancing features.

Validation:

Piloted the Trust Principle with a feature redesign and launch to demonstrate that the Trust Framework is actionable and positively impacts feature adoption and revenue.

Challenges Overcome:

• Addressed and mitigated user hesitancy toward AI, resulting in improved user experiences and higher feature adoption rates.

• Developed a standardized definition of Trust for measuring and improving trust across all feature and product teams in the platform.

• Successfully linked Trust to meaningful KPIs and business outcomes, including increased revenue and reduced churn rates.

Impact:

• Increased Revenue: Users with higher trust levels spent 4x more than those with lower levels of trust.

• Reduced Churn: Users with higher trust demonstrated significantly lower churn rates over a 90-day period.

• Organizational Buy-In: Established Trust as a critical organizational metric, securing additional resources for scaling the impact to additional product teams.