Verizon
Enhancing the Bring Your Own Device Experience
Platform
Web
Role
Experience Designer
Duration
May 2026
Iteration #1: Quick Assist and vzGPT
While exploring solutions, one of our developers introduced an AI-powered concept called Quick Assist: contextual prompts that could answer common questions and explain what customers were doing and why.
The concept was powered by vzGPT, Verizon’s integrated AI assistant.
I took the initial concept into Figma and refined it into a branded, scalable experience that could work beyond BYOD.
Rather than treating Quick Assist as a one-off solution, I designed the interaction pattern so it could be reused at other high-friction points throughout the buying experience.
That work eventually expanded into experiences including cart and Bring a Number.
Selecting "Find my IMEI", opening the drawer with more more detailed information.
Making IMEI entry easier
The next question was more specific:
Could vzGPT actually help customers retrieve their IMEI?
Instead of asking customers to navigate through their device settings and manually transcribe a 15-digit number, we introduced an Upload IMEI concept.
Customers could take a screenshot of their device information, upload it to vzGPT, and use the assistant to identify the IMEI.
They could then copy the result directly into the Verizon IMEI field.
This moved us from explaining how to find the IMEI to helping customers find it.
But there was still a problem.
On desktop, the approach was promising, but on mobile, the experience still required customers to leave the primary BYOD flow and interact with a modal.
We had reduced the effort of finding the IMEI, but we were still asking customers to leave the context of the purchase.
With launch only one week away, we shipped the initial experience and tabled additional improvements for post-launch exploration.
The results
Quick Assist produced a measurable improvement in conversion.
+6%
New-customer conversion
+26%
Existing-customer conversion
The results validated the broader direction: giving customers contextual assistance at moments of friction could meaningfully improve their ability to complete the purchase.
Iteration #2: Bringing vzGPT into the flow
After launch, I explored a more integrated version of Upload IMEI.
Instead of sending customers into a separate modal, I designed an IMEI drawer directly within the entry page.
The customer could upload their screenshot without leaving the flow, and vzGPT could identify the IMEI2 and populate the field for them.
The interaction became:
Upload → Identify → Populate → Continue
This removed an unnecessary context switch and brought assistance directly to the moment when the customer needed it.
The concept is currently pending development.
Next steps: 1-Page BYOD
With the Quick Assist and vzGPT work underway, my product partner and I stepped back and asked a bigger question:
What if BYOD didn't need to be a multi-page journey at all?
Our next goal was to condense the experience into a single page to make device management more accessible.
In the existing experience, customers had to rely on the browser's back navigation to revisit previous steps. That interaction could feel unreliable and made editing an earlier decision unnecessarily difficult.
A single-page experience could keep the customer's progress and their decisions in one place.
Iteration 1 — One-page BYOD
I began by bringing the existing IMEI entry experiences into a single page using expandable sections.
The first explorations used accordions underneath the Add Lines section, with different approaches to confirmation and progression.
The goal was straightforward:
Let customers see where they are, complete a step, and move forward without losing access to what they've already entered.
Version 1.1 — "Confirm" button at the bottom of the device entry drawer.
Version 1.2 — "Confirm" button in the corner to reduce confusion between drawer-level confirmation and page-level confirmation.
Learning from feedback
Design reviews quickly surfaced two issues.
First, the Confirm button pattern felt unnatural. Second, the overall experience still felt too heavy—the customer had to scroll through too much content to understand what they needed to do next.
The feedback shifted the direction of the design:
Don't make the customer navigate the interface. Let the interface guide the customer.
Final direction — Progressive drawers
For the final iteration, I designed a more focused, progressive experience, where each step became a compact drawer.
Customers would:
Complete a step → Confirm their decision → Move to the next step
The next drawer would automatically open, creating a clear sense of progression without requiring the customer to search for the next action.
At the same time, completed steps remained accessible, allowing customers to revisit and edit previous decisions without relying on browser navigation.
The result was a BYOD experience that felt less like moving through a sequence of pages and more like completing one cohesive task.
I also partnered with our content strategist to refine the language throughout the final experience, making each step clearer and more actionable.
What I learned
Reducing friction isn't always about removing steps. Instead, it's about making the necessary steps easier to understand and complete.
Across multiple iterations, the work evolved from contextual help, to AI-assisted IMEI retrieval, to a fundamentally different approach to the BYOD journey.
The strongest solutions came from repeatedly asking:
Where is the customer working harder than they should have to, and how can the product take on more of that work?
When my time with Verizon ended, the final designs were ready for continued discussion with my product partner around business requirements and technical feasibility before moving into development.