Contextual Product Recommendations
Contextual Product Recommendations
Contextual Product Recommendations
Designing a Contextual Recommendation Strategy to
Drive Relevance and Product Adoption
Designing a Contextual Recommendation Strategy to
Drive Relevance and Product Adoption
Designing a Contextual Recommendation Strategy to Drive Relevance and
Product Adoption
Role
Lead UX / Product Designer
Discover: understanding the problem
Discover: understanding the problem
Discover: understanding the problem
Product recommendations were displayed without context, resulting in low engagement and limited impact.
• Recommendations were not informed by user behavior or intent
• Placement felt random and disconnected from user actions
• Account openings driven by recommendations were underperforming
Why this mattered:
Customers were open to recommendations, but only when they felt relevant and timely. Generic promotions were easy to ignore and did not create meaningful value.
Product recommendations were displayed without context, resulting in low engagement and limited impact.
• Recommendations were not informed by user behavior or intent
• Placement felt random and disconnected from user actions
• Account openings driven by recommendations were underperforming
Why this mattered:
Customers were open to recommendations, but only when they felt relevant
and timely. Generic promotions were easy to ignore and did not create
meaningful value.
Product recommendations were displayed without context, resulting in low engagement and limited impact.
• Recommendations were not informed
by user behavior or intent
• Placement felt random and
disconnected from user actions
• Account openings driven by
recommendations were
underperforming
Why this mattered:
Customers were open to recommendations, but only when they felt relevant and timely. Generic promotions were easy to ignore and did not create meaningful value.
Synthesized existing research and behavioral insights
Synthesized existing research and behavioral insights
Users expressed frustration with irrelevant or excessive
promotional content
Users expressed frustration with irrelevant or excessive
promotional content
Users expressed frustration with irrelevant or excessive
promotional content
Users expected recommendations to be intuitive, relevant, and aligned with their current goals
Users expected recommendations to be intuitive, relevant, and aligned with their current goals
Users expected recommendations to be intuitive, relevant, and aligned with their current goals
Define: framing the problem
Define: framing the problem
Define: framing the problem
The opportunity was to shift from random recommendations to a contextual, user-centered approach.
Core Problem:
How might we present product recommendations in a way that feels relevant, timely, and valuable to each individual customer?
Key Insight
Customers don’t want more recommendations—they want better recommendations.
• Personalization increases engagement
• Timing and context drive relevance
• Trust is critical—recommendations must feel helpful, not promotional
Why this mattered:
This reframed the problem from increasing visibility to improving quality and intent alignment.
The opportunity was to shift from random recommendations to a contextual, user-centered approach.
Core Problem:
How might we present product recommendations in a way that feels relevant, timely, and valuable to each individual customer?
Key Insight
Customers don’t want more recommendations—they want better recommendations.
• Personalization increases engagement
• Timing and context drive relevance
• Trust is critical—recommendations
must feel helpful, not promotional
Why this mattered:
This reframed the problem from increasing visibility to improving quality and intent alignment.
Opportunity Areas Across the Experience
Opportunity Areas Across the Experience
Mapped potential areas across the product where contextual recommendations could provide value based on user behavior and intent.
Mapped potential areas across the product where contextual recommendations could provide value based on user behavior and intent.
Mapped potential areas across the product where contextual recommendations could provide value based on user behavior and intent.
Develop: exploring solutions
Develop: exploring solutions
Develop: exploring solutions
Research Synthesis
• Synthesized behavioral and usability research
• Identified patterns in:
• Customer expectations
• Response to personalization
• Engagement with existing recommendations
Why this approach:
Grounding the strategy in existing research allowed us to move quickly while ensuring decisions were rooted in real user behavior.
Opportunity Framing
Defined “contextual” as aligning recommendations to user actions and intent.
Examples:
• After paying a mortgage → surface refinance options
• Within transactions → suggest relevant financial products
• In account flows → introduce complementary services
Research Synthesis
• Synthesized behavioral and usability
research
• Identified patterns in:
• Customer expectations
• Response to personalization
• Engagement with existing
recommendations
Why this approach:
Grounding the strategy in existing research allowed us to move quickly while ensuring decisions were rooted in real user behavior.
Opportunity Framing
Defined “contextual” as aligning recommendations to user actions and intent.
Examples:
• After paying a mortgage → surface
refinance options
• Within transactions → suggest
relevant financial products
• In account flows → introduce
complementary services
Exploring Placement & Contextual Triggers
Exploring Placement & Contextual Triggers
Explored multiple placement strategies for contextual recommendations, evaluating how timing and proximity to user actions influenced relevance, visibility, and overall user experience.
Explored multiple placement strategies for contextual recommendations, evaluating how timing and proximity to user actions influenced relevance, visibility, and overall user experience.
Explored multiple placement strategies for contextual recommendations, evaluating how timing and proximity to user actions influenced relevance, visibility, and overall user experience.
Why this mattered:
Aligning recommendations with user behavior reduced friction and made them feel like helpful insights rather than interruptions.
Experience Design & Exploration
• Explored how recommendations could appear across:
• Dashboard
• Transaction flows
• Confirmation pages
• New account experiences
• Created concepts defining:
• Visual design
• Messaging tone
• Interaction patterns
Why this was important:
Consistency across surfaces ensured the experience felt intentional and cohesive rather than scattered.
Validation & Iteration
• Partnered with behavioral researchers to test concepts
• Validated that:
• Personalization increased engagement
• Messaging needed to feel tailored and insightful
• Visual design should balance visibility with subtlety
Key insight:
Recommendations should feel like financial guidance, not advertising.
Guardrails & Guidelines
Defined a framework to balance business goals with user experience.
Key guardrails:
• Frequency limits to prevent overexposure
• Ability for users to dismiss recommendations
• Context relevance requirements
• Visual standards to avoid “ad-like” presentation
Why this mattered:
Without guardrails, recommendations risked overwhelming users
and degrading trust.
Validation & Iteration
• Partnered with behavioral researchers
to test concepts
• Validated that:
• Personalization increased
engagement
• Messaging needed to feel tailored and
insightful
• Visual design should balance visibility
with subtlety
Key insight:
Recommendations should feel like financial guidance, not advertising.
Guardrails & Guidelines
Defined a framework to balance business goals with user experience.
Key guardrails:
• Frequency limits to prevent
overexposure
• Ability for users to dismiss
recommendations
• Context relevance requirements
• Visual standards to avoid “ad-like”
presentation
Why this mattered:
Without guardrails, recommendations risked overwhelming users
and degrading trust.
Translating Context into Actionable Recommendations
Translating Context into Actionable Recommendations
Developed concepts showing how contextual insights could translate into actionable product recommendations, connecting user behavior to relevant financial opportunities.
Developed concepts showing how contextual insights could translate into actionable product recommendations, connecting user behavior to relevant financial opportunities.
Deliver: refining & aligning the solution
Deliver: refining & aligning the solution
Deliver: refining & aligning the solution
Strategic Alignment
• Created a strategic presentation for Product and Design leadership
• Communicated:
• The opportunity
• The framework
• The expected impact
Why this approach:
Clear storytelling and structured thinking helped align stakeholders and drive adoption of the strategy.
Cross-Team Collaboration
• Worked with multiple teams to refine visual and interaction patterns
• Helped translate strategy into actionable implementation guidance
Why this mattered:
Ensured the strategy could scale across teams and product areas.
Key Decisions
Shift from Random → Contextual Delivery
• Grounded recommendations in user behavior to increase relevance.
Treat Recommendations as Insights, Not Ads
• Designed experiences to feel helpful and personalized.
Prioritize Timing Over Volume
• Focused on delivering fewer, higher-quality recommendations.
Establish Guardrails to Protect UX
• Balanced business goals with user trust and experience quality.
Design for System-Level Consistency
• Ensured recommendations worked across the entire product ecosystem.
Strategic Alignment
• Created a strategic presentation for Product and Design leadership
• Communicated:
• The opportunity
• The framework
• The expected impact
Why this approach:
Clear storytelling and structured thinking helped align stakeholders and drive adoption of the strategy.
Cross-Team Collaboration
• Worked with multiple teams to refine visual and interaction patterns
• Helped translate strategy into actionable implementation guidance
Why this mattered:
Ensured the strategy could scale across teams and product areas.
Key Decisions
Shift from Random → Contextual Delivery
• Grounded recommendations in user
behavior to increase relevance.
Treat Recommendations as Insights,
Not Ads
• Designed experiences to feel helpful
and personalized.
• Prioritize Timing Over Volume
• Focused on delivering fewer, higher
quality recommendations.
• Establish Guardrails to Protect UX
• Balanced business goals with user
trust and experience quality.
• Design for System-Level Consistency
• Ensured recommendations worked
across the entire product ecosystem.
Designing Guardrails for Contextual Recommendations
Designing Guardrails for Contextual Recommendations
Established a framework to ensure recommendations are relevant, timely, and non-intrusive—balancing business goals with user trust and experience quality.
Established a framework to ensure recommendations are relevant, timely, and non-intrusive—balancing business goals with user trust and experience quality.
Outcome
Outcome
Outcome
The strategy was well received by Product and Design leadership and influenced
the product roadmap.
• Contributed to the development of features using customer data to trigger recommendations
• Shifted the organization toward a more personalized, data-informed approach
• Established a scalable framework for recommendation design
Impact:
This work transformed product recommendations from generic promotions into a contextual, user-centered experience that balances business goals with
customer value.
The strategy was well received by Product and Design leadership and influenced
the product roadmap.
• Contributed to the development of
features using customer data to
trigger recommendations
• Shifted the organization toward a more
personalized, data-informed approach
• Established a scalable framework for
recommendation design
Impact:
This work transformed product recommendations from generic promotions into a contextual, user-centered experience that balances business goals with
customer value.
Personalized Recommendations Within the Dashboard
Personalized Recommendations Within the Dashboard
Client has a savings account with Truist.
We would recommend:
They have a lot of transactions so a Truist One Checking account would better suit their needs as a spending account.
Client has a savings account with Truist.
We would recommend:
They have a lot of transactions so a Truist One Checking account would better suit their needs as a spending account.
Client has a savings account with Truist.
We would recommend:
They have a lot of transactions so a Truist One Checking account would better suit their needs as a spending account.
Client has a high balance linked loan.
We would recommend:
Encourage them to compare our LightStream interest rate with their current one to get the best rate.
Client has a high balance linked loan.
We would recommend:
Encourage them to compare our LightStream interest rate with their current one to get the best rate.
Client has a high balance linked loan.
We would recommend:
Encourage them to compare our LightStream interest rate with their current one to get the best rate.
Surface relevant product recommendations directly within the account experience, using behavioral signals such as transaction patterns and account balances.
Surface relevant product recommendations directly within the account experience, using behavioral signals such as transaction patterns and account balances.
Key Learnings
Key Learnings
Key Learnings
What I’d do differently:
Involve data and engineering earlier to align on personalization capabilities
Define measurement metrics earlier to track success
What went well:
Strong synthesis of research into a clear strategy
Effective balance between business goals and user experience
Creation of scalable guidelines adopted across teams
What I’d do differently:
Involve data and engineering earlier to align on personalization capabilities
Define measurement metrics earlier to track success
What went well:
Strong synthesis of research into a clear strategy
Effective balance between business goals and user experience
Creation of scalable guidelines adopted across teams
Contextual Recommendations Triggered by User Actions
Contextual Recommendations Triggered by User Actions
Deliver timely recommendations at key moments—such as after completing a transaction—ensuring suggestions feel relevant and actionable rather than intrusive.
Deliver timely recommendations at key moments—such as after completing a transaction—ensuring suggestions feel relevant and actionable rather than intrusive.
Other Case Studies
Other Case Studies
Other Case Studies
Improving Findability of Statements & Tax Documents in a Consumer Financial Platform
Improving Findability of Statements & Tax Documents in a Consumer Financial Platform
Defining and Launching an AI Assistant
to Modernize Digital Banking
Defining and Launching an AI Assistant
to Modernize Digital Banking