ENTERPRISE SEARCH PLATFORM

Product Design, UX Design, UI Design, Visual Design,
Accessibility, User Testing, User Surveys, AI,
Cross-Function Collaboration


College Board

Designing a complex student segmentation and data purchasing system that enables institutions to identify, refine, and act on prospective student audiences.

Student Search allows colleges to define and purchase student datasets based on academic, demographic, and behavioral attributes. These datasets power recruitment — emails, mailers, and outreach students receive.

This is not a single feature — it is an ecosystem.

Student Search spans:

  • search creation

  • filtering logic

  • saved searches & orders

  • subscription constraints

  • data uploads & segmentation

  • account + criteria management

Each part contributes to a single outcome: defining the right audience to purchase. I worked across the system to improve clarity, feedback, and decision-making in a highly complex filtering environment.

Role: Senior Product Designer, end-to-end ownership.

Partnered with: Engineering, QA, Content, Accessibility

Led: UX strategy, workflow redesign, specs

Scope: End-to-end product design, UX strategy, workflow restructuring, specs, prototyping, research synthesis

Timeline: 1 Year

Tools: Figma, Zeplin, High + Low Fidelity Prototyping

The Problem

The existing experience had grown into a powerful but fragmented system over time.

While feature-rich, it created friction across nearly every stage of the workflow.

Key issues

1. Lack of clarity in filtering logic
Users could select filters, but struggled to understand:

  • how filters combined

  • what impact they had on results

  • whether their audience was accurate

2. No real-time feedback

  • Audience size was unclear until late in the process

  • Users couldn’t confidently iterate

  • Filtering felt like trial-and-error

3. Fragmented workflows
Search creation, saved searches, orders, and uploads existed as:

  • separate tools

  • inconsistent interfaces

  • disconnected mental models

4. High cognitive load

  • Dense tables

  • Nested dropdowns

  • Overwhelming inputs

Users were forced to translate system logic instead of focusing on strategy

Legacy State

The legacy system reflected years of incremental feature additions:

  • inconsistent UI patterns

  • minimal hierarchy

  • poor visibility into system state

Users frequently lacked confidence in: “Did I actually build the right audience?”

Transform a complex, opaque system into one that:

provides continuous feedback

  • supports confident decision-making

  • scales across expanding datasets and features

  • connects the full workflow from search → purchase

Goal

Making a Complex Data System Legible

Student Search is an enterprise SaaS platform used by enrollment teams to identify, segment, and order student audiences from a large and highly detailed dataset.

Because the system supported complex business rules, dozens of filters, subscription-based access, and multiple connected workflows, simply removing functionality was not an option.

Instead of reducing the system’s complexity, I focused on making that complexity understandable, visible, and actionable.

My design approach centered on three principles:

  • Expose system behavior

  • Structure information clearly

  • Reinforce cause-and-effect relationships

Approach

Turning Filtering Into a Data Visualization

A primary challenge was helping users understand how each filter affected the size and composition of their audience.

Previously, users could apply multiple criteria without clearly seeing the impact of their decisions. They were often operating blindly and had limited confidence in whether their search would produce a useful audience.

I introduced a persistent Students Included visualization that updated as users adjusted their criteria.

As filters were applied:

  • Audience size updated immediately

  • Increases and decreases became visible

  • Users could compare the impact of different criteria

  • Potentially restrictive combinations could be identified earlier

  • Users could refine their strategy without leaving the workflow

This transformed the result count from a passive number into an active decision-making tool.

Rather than simply showing how many students matched, the visualization created a continuous feedback loop between user input and system output.

Result

Filtering became a feedback-driven decision system in which users could test, evaluate, and refine their audience with greater confidence.

Structuring Complex Filters & Data insights

Student Search gives enrollment teams access to a rich dataset containing thousands of student records and dozens of segmentation criteria. The challenge wasn't reducing complexity—it was helping users confidently navigate and understand it.

To support this, I designed both the filtering experience and the accompanying data visualizations to work together as a unified decision-support system.

Organizing Complex Filters

The platform included filters across multiple dimensions, including:

  • Academics

  • Geography

  • Testing

  • Demographics

  • Student interests

  • Behavioral data

  • Environmental and neighborhood insights

Rather than exposing every option at once, filters were grouped into logical, collapsible sections using progressive disclosure. This allowed users to focus on one category at a time while maintaining awareness of the broader filtering system.

Design decisions

  • Organized filters into meaningful categories

  • Reduced visual noise through accordion sections

  • Maintained consistency across filter interactions

  • Surfaced advanced options only when needed

Impact

  • Reduced cognitive overload

  • Improved scanability

  • Increased task completion efficiency

  • Supported large-scale enterprise workflows

Beyond Search: Turning Student Data Into Actionable Insights

Creating a student audience was only one part of the recruitment workflow. Enrollment teams also needed to understand the composition and potential value of their selected audience before moving forward.

I helped design the Segment Analysis Service (SAS), a connected reporting experience that transformed large student datasets into accessible visual insights.

The challenge

Enrollment teams previously relied on dense tables and exported spreadsheets to analyze student populations. This made it difficult to quickly identify patterns, compare audience segments or determine whether a selected audience supported their recruitment goals.

Users needed to answer questions such as:

  • How is this audience distributed geographically?

  • Which student segments are most represented?

  • How do different cohorts compare?

  • Are there gaps or unexpected patterns in the data?

  • Does this audience align with our recruitment strategy?

Designing the dashboard

I designed a reporting experience that connected interactive filtering, data visualization and detailed student data within one interface.

The dashboard allowed users to:

  • Compare student cohorts and clusters

  • Explore geographic and demographic distributions

  • Review historical audience performance

  • Apply filters without leaving the report

  • Move between summary visualizations and detailed tables

  • Export reports for additional analysis

Rather than presenting every control at once, I organized filters into progressive sections. This helped users focus on the information most relevant to their analysis while preserving access to advanced criteria.

Connecting search and analysis

Student Search and SAS served different parts of the same decision-making process:

Define an audience → Evaluate its composition → Compare segments → Refine the strategy → Place an order

By connecting the filtering and reporting experiences, the platform supported more than audience creation. It helped enrollment teams understand the students represented by their selections and make more informed recruitment decisions.

Outcome

The redesigned experience shifted Student Search from a static filtering tool into a more transparent decision-support system.

User impact

  • Gave users immediate visibility into how criteria affected audience size

  • Made complex filters easier to scan and understand

  • Helped users identify overly restrictive combinations earlier

  • Connected audience creation with deeper segment analysis

  • Reduced the need to move between disconnected tools and spreadsheets

Product impact

  • Established scalable filtering patterns across search and reporting

  • Unified previously fragmented search, analysis and ordering workflows

  • Created reusable interaction patterns for future datasets and features

  • Improved visibility into system behavior throughout the user journey

Business impact

  • Supported more precise audience targeting

  • Reduced the risk of purchasing an incorrectly configured dataset

  • Helped institutions make more informed recruitment decisions

  • Strengthened Student Search as a strategic recruitment platform rather than a transactional purchasing tool

Reflection

The central challenge was not eliminating complexity. Student Search needed to support detailed data, business rules, subscription constraints and advanced institutional workflows.

My role was to make that complexity visible and understandable.

By pairing structured filtering with continuous feedback and interactive reporting, we transformed the experience from a collection of inputs into a connected decision-making system.

User Flow

Selecting filters

Outcome / Impact

Reflection

Visual Dashboard

Beyond Search: From Finding Students to Understanding Them

Student Search wasn't limited to finding the right audience. It also included tools that helped enrollment teams analyze, compare, and validate their search results before making strategic decisions.

As part of the broader Search ecosystem, I designed experiences for the Segment Analysis Service (SAS), where users could transform large student datasets into interactive reports and visualizations.

Previous
Previous

SAT Suite Mobile Score Reports

Next
Next

Scaling Ordering Experience