In-context data insight
‍Provide instructors with data insights about learner and content directly within their workflow, to support intervention and deliver adaptive paths
Platform
Brightspace, a cloud-based learning management system (LMS) designed for both educational institutions and corporate training.

My Role
Led end-to-end design:  defined solution with stakeholders, research, delivered design, defined prompt strategy, and AI–human oversight workflows.
Team
1 Product manager
1 Dev manager
1 ML expert
5 Developers
1 Principal designer (report to)
Timeline
Feb 2025 to current
🗄 The Background
Instructors struggles to use data on our platform
Over the years, we have a LOT of past research, client feedback, idea request, and past initiatives around instructors' data usage.
‍
I gathered all the info and synthesized the findings so that I know what problems are worth solving
Scattered and fragmented data tools
‍
Instructors struggle to find and use the data tools available, as they are spread across different parts of the platform without a unified entry point.
Low data literacy

‍Many instructors find it difficult to interpret student data, often lacking the background or confidence to translate it into actionable decisions.
Limited time and bandwidth
‍
‍Instructors are busy and need timely, relevant insights—not raw data—to make informed teaching decisions without added cognitive load.
Our Goal
‍
Reduce friction in discovering, interpreting, and responding to student learning trends through smart, embedded insights.
🗄 My Process
As this work is still in development, I can't disclose the design details
‍
Contact me if you want to learn more

Synthesized input from multiple sources

Reviewed years of client feedback, past research, and product requests to identify persistent problems around data access and use.

Facilitated team ideation workshops

Ran a 2-day workshop to brainstorm potential instructor-facing insights and mapped out what data would be required to support them.

Define insight logic and data model

Collaborated cross-functionally to document the data required for each insight and define how it should be calculated. This step required deep alignment, as multiple interpretations of the same data often emerged.

Define intervention suggestion

Created a tiered model of instructor interventions—from simple nudges to more sophisticated automation—so that insights could scale and evolve over time while delivering value early.

UX Research + Design Iteration

Collaborated with a ux researcher to test our proposed approach with 8 instructors. Research finding informed design direction and release plan.
🎯 The Result

A sneak peek into the solution

‼️ Highlight insights that need attention
Monitor the platform and alert the instructor about student performance and engagement with course activities (quiz, assignment, content, etc.)
📈 Progressive disclosure of more information
Instructors can click into the insight to gain more information (e.g., what data are associated) and inform action accordingly
✨ AI supported intervention
Based on the data, AI will provide relevant suggestions about content and learner intervention

e.g., AI helps instructor to rewrite content or contact struggling learner
✅ Outcome of the work

Positive feedback from
UX Research

I think this is very exciting and our faculty would be very excited.
Our faculty are very more so in the post-COVID era and much more data-minded then they have been in the past...
and they would be excited to have the data at their finger tip then having to dig for it. ‍
Participant 2
UX research
If AI is going to do that for me, I find that usefulI.  I could do all the analysis, AI could do it a lot faster than I could, possible that AI is determining which questions go with which activities...
Participant 3
UX research
This is a time saver . To me this is a win-win and it would hopefully increase the way some student would perform.
Participant 7
UX research