August 2025 - December 2025

Verizon

Worked as a UX Designer for the Digital Unity team, worked on developing direction to shape future app experiences

At-a-glance

In an 8-week discovery sprint, I explored how Verizon's core consumer apps could evolve into a single, unified experience enhanced by AI personalization. Partnering with product, engineering, and research teams, I drove early discovery work spanning literature reviews, information architecture exploration, and adaptive concept design to define a structural model that could scale across diverse user behaviors.

Goals

Consolidate four separate consumer apps into one cohesive, navigable experience
Define an adaptive architecture that serves both power users and infrequent ones
Identify where AI personalization adds clarity without creating dependency or overwhelm

INDUSTRY

IA & AI Personalization

ROLE

Research | Conceptualization | Design

TIMELINE

6 week sprint

TEAM

Jeffrey Powers

Steven

Bazarian

Lillian

Lin

The Problem

Verizon's consumer experience is fragmented across four separate apps, making it hard for users to discover features and maintain a consistent mental model. The challenge was architectural: could one unified structure adapt to vastly different user behaviors without overwhelming either group?

Starting with Behavior, Not Structure

We began with a “maximalist” persona, a highly engaged customer interacting across multiple services. This stress-tested the upper bounds of navigation depth and structural complexity.

Supporting artifacts included journey mapping and intent analysis to identify task frequency, friction points, and discoverability challenges.

Diane, Maximal Customer

AGE

Middle age

FAMILY

2 Children

STATUS

Partnered

BIO

Highly engaged, tech-forward, uses many tech products and features to manage life and save time.

OCCUPATION

Tech VP

LOCATION

California

TECH LITERACY

High

Informing Adaptivity Through Human–AI Patterns

To evaluate how adaptivity should function, I studied emerging human–AI interaction models across industries.

Rather than designing AI features in isolation, I focused on where intelligence builds trust, balances user control, and clarifies rather than overwhelms.

These insights informed how adaptivity should integrate structurally into the app.

Consolidating Through Hierarchy

With behavioral framing in place, I mapped detailed information architecture models to explore consolidation within a unified framework.

This pass improved logical grouping and reduced fragmentation. However, deeper hierarchies began to bury infrequent tasks. The system felt cleaner but increasingly rigid and biased toward highly engaged users.

Hierarchy alone was not enough.

Reframing Adaptivity as Structural

Rather than refining hierarchy further, I expanded the lens to include minimalist users, those who engage infrequently and rely on a small set of core features.

The challenge shifted from grouping features to defining where adaptivity should live within the system.

Within a two-week sprint, I generated rapid conceptual wireframes to test alternate architectural models.

Centralized vs Distributed Adaptivity

CONCEPT A - HOME-CENTERED ADAPTIVITY

Intelligence is centralized. The home surface predicts frequent tasks and reduces reliance on navigation.

CONCEPT B - ADAPTIVE NAVIGATION HUBS

Navigation reflects user-enabled feature hubs. Core areas remain persistent while hubs adapt based on relevance and preference. Intelligence is distributed, preserving user agency.

Prioritizing Scalability and User Agency

Concept A centralized prediction.
Concept B distributed adaptivity.

Concept B provided a more scalable foundation. It supported both minimalist and maximalist behaviors, reduced reliance on system guesswork, and created a clearer mental model as features expanded.

Comparisons & Trade-offs

Criteria

Concept A

Concept B

Where adaptivity lives

Home feed

Navigation + Home

Navigation model

Stable, fixed tabs

Dynamic, hub-based

User control

Implicit (behavior-driven)

Explicit (enable / disable hubs)

Cognitive load

Managed through content surfacing

Managed through feature removal

Scalability

Moderate

Higher

Designing Modular Navigation for Scalability

Concept B introduced adaptive navigation hubs, modular groupings of related features that appear based on relevance and user preference.

This allowed complexity to expand without overwhelming infrequent users while maintaining structural consistency across behaviors.

Returning to Structure with a Clear Model

With a conceptual framework established, I revisited the information architecture.

Features were re-evaluated based on purpose, flow, and relationship to each adaptive hub. Structure was now guided by the adaptive model rather than isolated grouping exercises.

This phase focused on structural validation rather than polished UI execution.