How I strategized an AI-powered learning ecosystem for Trailhead to simplify choices and accelerate skill growth
How I strategized an AI-powered learning ecosystem for Trailhead to simplify choices and accelerate skill growth
How I strategized an AI-powered learning ecosystem for Trailhead to simplify choices and accelerate skill growth
Learning Management System | AI Strategy Case Study
Learning Management System | AI Strategy Case Study
Learning Management System | AI Strategy Case Study
Improved search experience and learning flow for 3M+ learners, boosting discoverability by 60% and reducing cognitive overload across core tasks.
Improved search experience and learning flow for 3M+ learners, boosting discoverability by 60% and reducing cognitive overload across core tasks.
Improved search experience and learning flow for 3M+ learners, boosting discoverability by 60% and reducing cognitive overload across core tasks.
My Role
My Role
My Role
Product Designer
Product Designer
Product Designer
Nature of Project
Nature of
Project
Nature of Project
AI-Enhanced Learning Experience & Search Strategy Initiative
AI-Enhanced Learning Experience & Search Strategy Initiative
AI-Enhanced Learning Experience & Search Strategy Initiative
Timeline
Timeline
Timeline
40 Weeks
(Aug 2024 - May 2025)
40 Weeks
(Aug 2024 - May 2025)
40 Weeks
(Aug 2024 - May 2025)
Platform
Platform
Platform
Web
Web
Web
Responsibility
Responsibility
Responsibility
Research + Product Strategy + {AI Design} + UX Architecture + Devs Hands Off
Research + Product Strategy + {AI Design} + UX Architecture + Devs Hands Off
Research + Product Strategy + {AI Design} + UX Architecture + Devs Hands Off
Discovery
Discovery
Discovery
Identified core problem areas through interviews, competitor analysis, journey mapping, and task analysis.
Defined Jobs to Be Done, scoped the work with PM, and aligned on success metrics and testing time.
Identified core problem areas through interviews, competitor analysis, journey mapping, and task analysis.
Defined Jobs to Be Done, scoped the work with PM, and aligned on success metrics and testing time.
Identified core problem areas through interviews, competitor analysis, journey mapping, and task analysis.
Defined Jobs to Be Done, scoped the work with PM, and aligned on success metrics and testing time.
Define
Define
Define
Prioritized AI opportunities grounded in learner needs and platform goals.
Shaped the AI-powered learning ecosystem vision with clear experience principles.
Prioritized AI opportunities grounded in learner needs and platform goals.
Shaped the AI-powered learning ecosystem vision with clear experience principles.
Prioritized AI opportunities grounded in learner needs and platform goals.
Shaped the AI-powered learning ecosystem vision with clear experience principles.
Development
Development
Development
Built prototypes and conceptual models demonstrating how AI supports learner decisions.
Validated designs through moderated user testing and refined patterns for clarity.
Presented the concept to the Trailhead team and aligned on KPIs for future impact measurement.
Built prototypes and conceptual models demonstrating how AI supports learner decisions.
Validated designs through moderated user testing and refined patterns for clarity.
Presented the concept to the Trailhead team and aligned on KPIs for future impact measurement.
Built prototypes and conceptual models demonstrating how AI supports learner decisions.
Validated designs through moderated user testing and refined patterns for clarity.
Presented the concept to the Trailhead team and aligned on KPIs for future impact measurement.
Business Context
Business Context
Business Context
What is Trailhead?
What is Trailhead?
What is Trailhead?
{ Salesforce’s version of an online university }
{ Salesforce’s version of an online university }
{ Salesforce’s version of an online university }
Trailhead is Salesforce’s learning management system that enables skill development, career readiness, and continuous learning through modular, gamified, and role-based content.
Learners explore topics, complete hands-on challenges, earn badges, and grow toward certifications - all within a single, guided ecosystem.
Trailhead is Salesforce’s learning management system that enables skill development, career readiness, and continuous learning through modular, gamified, and role-based content.
Learners explore topics, complete hands-on challenges, earn badges, and grow toward certifications - all within a single, guided ecosystem.
Trailhead is Salesforce’s learning management system that enables skill development, career readiness, and continuous learning through modular, gamified, and role-based content.
Learners explore topics, complete hands-on challenges, earn badges, and grow toward certifications - all within a single, guided ecosystem.
Trailhead = Content + Guidance + Skill Mastery
Trailhead = Content + Guidance + Skill Mastery
Trailhead = Content + Guidance + Skill Mastery
3M +
Learners
3M +
Learners
3M +
Learners
1K +
Modules
1K +
Modules
1K +
Modules
~1500
Badges Available
~1500
Badges Available
~1500
Badges Available
Why redesign with AI intergration was needed?
Why redesign with AI intergration was needed?
Why redesign with AI intergration was needed?
(From both business & user point of view)
(From both business & user point of view)
(From both business & user point of view)
Experience Elements
Experience Elements
Experience Elements
Before
Before
Before
After
After
After
Who Benefits?
Who Benefits?
Who Benefits?
Search
Search
Search
Generic, keyword-based results; too many irrelevant options
Generic, keyword-based results; too many irrelevant options
Generic, keyword-based results; too many irrelevant options
Intent-aware, conversational search with meaningfully narrowed results
Intent-aware, conversational search with meaningfully narrowed results
Intent-aware, conversational search with meaningfully narrowed results
Business + User
Business + User
Business + User
Discovery of learning path
Discovery of learning path
Discovery of learning path
Hundreds of modules & trails → choice overload and decision fatigue
Hundreds of modules & trails → choice overload and decision fatigue
Hundreds of modules & trails → choice overload and decision fatigue
Intent-aware, conversational search with meaningfully narrowed results
Intent-aware, conversational search with meaningfully narrowed results
Intent-aware, conversational search with meaningfully narrowed results
Users
Users
Users
Recommendations
Recommendations
Recommendations
Static, one-size-fits-all & often ignored
Static, one-size-fits-all & often ignored
Static, one-size-fits-all & often ignored
Personalized, context-aware recommendations that adapt over time
Personalized, context-aware recommendations that adapt over time
Personalized, context-aware recommendations that adapt over time
Business + User
Business + User
Users
Progress Tracking
Progress Tracking
Progress Tracking
Fragmented, unclear “what’s next”
Fragmented, unclear “what’s next”
Fragmented, unclear “what’s next”
Unified dashboard + guided next steps
Unified dashboard + guided next steps
Unified dashboard + guided next steps
Both (slightly more User)
Both (slightly more User)
Users
Design Opportunity: Applying AI to Learning
Design Opportunity: Applying AI to Learning
Design Opportunity: Applying AI to Learning
AI integration enables Trailhead to proactively guide learners instead of leaving them to navigate thousands of modules on their own.
AI integration enables Trailhead to proactively guide learners instead of leaving them to navigate thousands of modules on their own.
AI integration enables Trailhead to proactively guide learners instead of leaving them to navigate thousands of modules on their own.
Search flow and Progress Tracking flow
Search flow and Progress Tracking flow
Search flow and Progress Tracking flow
Initial Brief
Initial Brief
Initial Brief
"How might we enhance the core learning experience on Trailhead?"
"How might we enhance the
core learning experience on Trailhead?"
"How might we enhance the core learning experience on Trailhead?"
Research, Benchmarking & Problem Framing
Research, Benchmarking & Problem Framing
Research, Benchmarking & Problem Framing
Research objective
Research objective
Research objective
To uncover where learners get stuck across the learning journey and identify high-impact moments where AI could meaningfully support
learning decisions without removing autonomy.
To uncover where learners get stuck across the learning journey and
identify high-impact moments where AI could meaningfully support
learning decisions without removing autonomy.
To uncover where learners get stuck across the learning journey and identify high-impact moments where AI could
meaningfully support learning decisions without removing autonomy.
Signals from Learner Conversations - User Interviews + Sentiment Analysis
Signals from Learner Conversations -
User Interviews + Sentiment Analysis
Signals from Learner Conversations - User Interviews + Sentiment Analysis
Sentiment Analysis
Sentiment Analysis
Sentiment Analysis
125+ reviews analyzed
125+ reviews analyzed
125+ reviews analyzed
Media Scanning Sources: Quora, YouTube comments, Reddit
Media Scanning Sources: Quora, YouTube comments, Reddit
Media Scanning Sources: Quora, YouTube comments, Reddit
Usability Testing and user interviews
Usability Testing and user interviews
Usability Testing and user interviews
Signals from the Learning Market- Competitor Analysis + Industry Trend
Signals from the Learning Market-
Competitor Analysis + Industry Trend
Signals from the Learning Market- Competitor Analysis + Industry Trend
Pattenrs observed across other platforms
Pattenrs observed across other platforms
Pattenrs observed across other platforms
Platforms Analyzed : LinkedIn Learning, Microsoft Learn, Google Skill Boost, Udemy, Code Academy
Platforms Analyzed : LinkedIn Learning, Microsoft Learn, Google Skill
Boost, Udemy, Code Academy
Platforms Analyzed : LinkedIn Learning, Microsoft Learn, Google Skill Boost, Udemy, Code Academy
What works elsewhere
What works elsewhere
What works elsewhere
Autocomplete reduces hesitation within the first 2-3
characters
Autocomplete reduces hesitation within the first 2-3
characters
Autocomplete reduces hesitation within the
first 2-3 characters
Flexible phrasing supports exploratory and goal-based
learning
Flexible phrasing supports exploratory and goal-based
learning
Flexible phrasing supports exploratory and
goal-based learning
Popular searches, tags, and paths build early confidence
Context-aware AI agents support learners within their
learning flow
Study plans reinforce motivation and consistent learning
habits
Popular searches, tags, and paths build early confidence
Popular searches, tags, and paths build early
confidence
Context-aware AI agents support learners within their
learning flow
Context-aware AI agents support learners within
their learning flow
Study plans reinforce motivation and consistent learning
habits
Study plans reinforce motivation and consistent
learning habits
Where Trailhead falls short
Where Trailhead falls short
Search relies on exact keywords
Search relies on exact keywords
Search relies on exact keywords
No intent or goal understanding
Learners must “know what to ask” before learning begins
Progress and recommendations feel fragmented
Agentforce provides generic Trailhead information, not
learner-context support
No intent or goal understanding
Learners must “know what to ask” before learning
begins
Progress and recommendations feel fragmented
Agentforce provides generic Trailhead information,
not learner-context support
No intent or goal understanding
Learners must “know what to ask” before learning begins
Progress and recommendations feel fragmented
Agentforce provides generic Trailhead information, not
learner-context support
Opportunity Area
Opportunity Area
Opportunity Area
AI can evolve Trailhead from a content library into a context-aware learning companion by guiding intent, reinforcing habits, and
supporting learners at the moment of need.
AI can evolve Trailhead from a content library into a context-
aware learning companion by guiding intent, reinforcing
habits, and supporting learners at the moment of need.
AI can evolve Trailhead from a content library into a context-aware learning companion by guiding intent, reinforcing
habits, and supporting learners at the moment of need.
With this synthesis the brief was defined as…
With this synthesis the brief was defined as…
With this synthesis the brief was defined as…
Phase 1 : Discovery - Designing How Learning Begins
Goal: Reduce hesitation at the first moment of learning by adapting discovery to learner intent.
Lane 1: Search Modal
1.1. Search Placement - Making intent visible before results
Learners were not struggling because search was broken - they were struggling because the search meant something
different than it looked like.
Now that search placement matches intent, what do users see when they search?
1.2. Search Modal Results — Designing for different learner states Placement
Once intent was clear, now goal is to focus on how search results should adapt to who is searching - not just what they type.
Benchmarking Study and Opportunity
Benchmarking Study and Opportunity
Benchmarking Study and Opportunity
Final UI Designs
Final UI Designs
Phase 1 : Discovery - Designing How Learning Begins
Phase 1 : Discovery - Designing How Learning Begins
Goal: Reduce hesitation at the first moment of learning by adapting discovery to learner intent.
Goal: Reduce hesitation at the first moment of learning by adapting
discovery to learner intent.
Lane 1: Search Modal
Lane 1: Search Modal
1.1. Search Placement - Making intent visible before results
1.1. Search Placement-Making intent visible before results
Learners were not struggling because search was broken - they were struggling because the search meant something different than it
looked like.
Learners were not struggling because search was broken - they were
struggling because the search meant something different than it
looked like.
Now that search placement matches intent, what do users see when they search?
Now that search placement matches intent, what do users see when
they search?
1.2. Search Modal Results — Designing for different learner states Placement
1.2. Search Modal Results — Designing for different
learner states Placement
Once intent was clear, now goal is to focus on how search results should adapt to who is searching - not just what they type.
Once intent was clear, now goal is to focus on how search results should
adapt to who is searching - not just what they type.
Lane 2: Filter- Designing how learners narrow down decision
Help learners narrow hundreds of learning options into one confident next step without slowing them down
Final UI Designs
Final UI Designs
Lane 2: Filter- Designing how learners narrow down decision
Lane 2: Filter- Designing how learners narrow down
decision
Help learners narrow hundreds of learning options into one confident next step without slowing them down
Learners were not struggling because search was broken - they were
struggling because the search meant something different than it
looked like.
Final UI Designs
Final UI Designs
How AI plays role in Discovery Ecosystem?
How AI plays role in Discovery Ecosystem?
How AI plays role in Discovery Ecosystem?
Search and Filters become the controls… and AI becomes the learning brain that turns those inputs into meaningful discovery.
Search and Filters become the controls… and AI becomes the learning
brain that turns those inputs into meaningful discovery.
Search and Filters become the controls… and AI becomes the learning brain that turns those inputs into meaningful
discovery.
Impact
Impact
Impact
Phase 2: Progress Tracking - Designing How Learning Continues
Phase 2: Progress Tracking - Designing How Learning
Continues
Phase 2: Progress Tracking - Designing How Learning Continues
Goal: Reduce drop-off after discovery by making progress visible,
motivating, and self-directed.
Goal: Reduce drop-off after discovery by making progress visible, motivating, and self-directed.
Goal: Reduce drop-off after discovery by making progress visible, motivating, and self-directed.
Final UI Design
Final UI Design
Final UI Design
How AI plays role in Progress Tracking Ecosystem?
How AI plays role in Progress Tracking Ecosystem?
How AI plays role in Progress Tracking Ecosystem?
AI turns progress from a history log into a forward-looking learning guide.
AI turns progress from a history log into a forward-looking learning
guide.
AI turns progress from a history log into a forward-looking learning guide.
Rank “Jump Right In” and “Up Next” based on engagement risk
Rank “Jump Right In” and “Up Next” based on engagement risk
Adapt study plans to each learner’s real pace
Adapt study plans to each learner’s real pace
Impact
Impact
Impact
Phase 3: Course Handling - Designing how learning happens inside a module
Phase 3: Course Handling - Designing how learning
happens inside a module
Phase 3: Course Handling - Designing how learning happens inside a module
Goal: Reduce cognitive overload during learning while increasing retention
and action by turning content into something learners can interact with,
ask about, and remember.
Goal: Reduce cognitive overload during learning while increasing retention and action by turning content into something
learners can interact with, ask about, and remember.
Goal: Reduce cognitive overload during learning while increasing retention and action by turning content into something learners can interact with, ask about, and remember.
Final UI Design
Final UI Design
Final UI Design
Impact
Impact
Impact
Learner Feedback
Learner Feedback
Learner Feedback
What 26 Trailhead users said during usability testing-
What 26 Trailhead users said during usability testing-
What 26 Trailhead users said during usability testing-
Overall Results and Impact
Overall Results and Impact
Overall Results and Impact
The results went beyond expectations. By embedding AI into the learning flow, we didn’t just fix usability issues - we aligned the system
with how learners already study, allowing AI to adapt to their patterns and lower decision friction.
The results went beyond expectations. By embedding AI into the
learning flow, we didn’t just fix usability issues - we aligned the system
with how learners already study, allowing AI to adapt to their patterns
and lower decision friction.
The results went beyond expectations. By embedding AI into the learning flow, we didn’t just fix usability issues -
we aligned the system with how learners already study, allowing AI to adapt to their patterns and lower decision friction.
My Learnings
My Learnings
My Learnings
What this project taught me about learning management systems
What this project taught me about learning management systems
What this project taught me about learning management systems
Focus Economics > Feature Density
Focus Economics > Feature Density
Unifying progress into a single surface
reduced scanning and restored flow.
Unifying progress into a single surface reduced scanning and
restored flow.
Unifying progress into a single surface
reduced scanning and restored flow.
AI needs context to be useful
AI needs context to be useful
By adding course-aware prompts, we
turned AI from a blank chat box into a
guided learning assistant.
By adding course-aware prompts, we turned AI from a blank
chat box into a guided learning assistant.
By adding course-aware prompts, we
turned AI from a blank chat box into a
guided learning assistant.
Planning converts intention into action
Planning converts intention into action
Motivation rises when learners can see
the finish line and the steps to reach it.
Motivation rises when learners can see the finish line and the
steps to reach it.
Motivation rises when learners can see
the finish line and the steps to reach it.
What this project taught me about building products
What this project taught me about building products
What this project taught me about building products
Involve & create together : Bringing stakeholders into testing sessions - not just readouts but it helped everyone see the same problems
and align faster.
Show, don’t tell : We didn’t debate abstractions rather we put real prototypes in front of users and watched what broke.
Involve & create together : Bringing stakeholders into testing sessions
- not just readouts but it helped everyone see the same problems
and align faster.
Show, don’t tell : We didn’t debate abstractions rather we put real
prototypes in front of users and watched what broke.
Involve & create together : Bringing stakeholders into testing sessions - not just readouts but it helped everyone see the
same problems and align faster.
Show, don’t tell : We didn’t debate abstractions rather we put real prototypes in front of users and watched what broke.
Future Scope
Future Scope
Future Scope
Something we would take up if we had more time or engineering bandwidth at hand-
Something we would take up if we had more time or engineering
bandwidth at hand
Something we would take up if we had more time or engineering bandwidth at hand-
Calendar Integration: AI would connect each learner’s study plan to their work calendar, making learning time visible and intentional.
AI-Generated Career Insights : Using progress, notes, and engagement patterns, AI could generate skill summaries, readiness indicators and
"What you’re becoming good at” snapshots
Calendar Integration: AI would connect each learner’s study plan to
their work calendar, making learning time visible and intentional.
AI-Generated Career Insights : Using progress, notes, and engagement
patterns, AI could generate skill summaries, readiness indicators and
"What you’re becoming good at” snapshots
Calendar Integration: AI would connect each learner’s study plan to their work calendar, making learning time visible and
intentional.
AI-Generated Career Insights : Using progress, notes, and engagement patterns, AI could generate skill summaries, readiness
indicators and "What you’re becoming good at” snapshots
Presenting the Learning Ecosystem
Presenting the Learning Ecosystem
Presenting the Learning Ecosystem
Presented AI-enhanced Trailhead LMS at Indiana University’s Capstone Showcase. Sharing the process with faculty, industry mentors, and
peers was a proud moment and a reminder that thoughtful design can spark real impact.
Presented AI-enhanced Trailhead LMS at Indiana University’s Capstone Showcase. Sharing the process with faculty, industry mentors, and
peers was a proud moment and a reminder that thoughtful design can spark real impact.
Presented AI-enhanced Trailhead LMS at Indiana University’s Capstone Showcase. Sharing the process with faculty, industry mentors, and peers was a proud moment and a reminder that thoughtful design can spark real impact.
