Lead UX Designer & Researcher
I’m the one who asks, “Wait, why do we do it that way?” Then I get curious, ask a few more questions, and find out what could work better.
First, I get close
to the work.
A confusing screen might actually be a broken handoff. A slow process might be asking someone to do work the technology could handle. And a seemingly simple fix might create a whole new problem somewhere else.
I love figuring out those connections—and helping teams decide what to do about them.
Start here
Different challenges, different people, and a lot of asking questions before deciding what to change.
Observed task time · same task segment
About 25% lower observed task time.
Start here · 1 of 3
For a Customer Care agent, a reply starts with figuring out what happened. I led research and designed CareConnect so agents could check the AI’s analysis, edit the draft, and decide what to send.
Verified outcome25% less time preparing a reply · 13:29 → 10:05
Role: Lead UX Designer & Researcher
Service journey · 5 agent interviews
Start here · 2 of 3
A receipt could be on file and still show up as ‘no results.’ I used agent interviews to find where the search broke down and help shape what travelers and agents could do next.
Verified outcome~50% fewer support calls · ~15% more self-service · 0 major outages reported after launch — program results reported by PM Josephine Wilson
Role: UX designer and researcher
Handoff sequence · 5 agents, 3 customers observed
Start here · 3 of 3
One screen said complete. The customer’s iPad kept loading. After observing payments at Seattle Cargo Station, I designed clearer signing, shared status and recovery across desktop, iPad and card reader.
Research scope5 agents + 3 customers · desktop, iPad, and reader
Role: Lead UX designer and researcher for the portal’s agent desktop and customer iPad
I like getting close to the actual work—watching what happens when someone is busy, when something goes wrong, or when the process doesn’t quite make sense. Sometimes the answer is a better interface. Sometimes it’s helping a team realize they’ve been solving the wrong problem.
More about meEspecially the ones that begin with “Wait, why?” I want to understand the problem underneath the request before jumping into solutions.
A conversation at a busy counter can reveal things a requirements document never will. I listen, observe, and follow the work beyond the screen.
The confusing error. The missing next step. The person using an older device. Small details can have an outsized effect on someone’s day.
Research matters most when it leads somewhere. I turn what we learn into clearer choices, practical priorities, and designs people can use.
Different problems call for different kinds of research. Here’s what I studied, measured, and brought back to teams.
Measured result · CareConnect
~25%
less time preparing a reply
That’s 3:24 less in the observed post-rollout comparison. Agents checked and edited the draft before returning the reviewed reply to their case.
Read the CareConnect case studyWho and what I studied—sample sizes, not performance results.
5 + 3
Agents and customers observed across desktop, iPad, and reader.
Cargo Payment Portal5 · 16
Agent interviews and documented UAT scenarios.
Receipt & Inquiry6:54
Mean failure-to-success time across 6 controlled-study participants.
QR Pay-by-Link7
Stakeholder interviews synthesized into a shared research baseline.
Atlas 2.0Let’s talk
I love a complicated problem, a thoughtful team, and a good conversation. If you’ve got something that could use a fresh perspective, I’d love to hear about it.
elise.ivory14@gmail.comFor agencies: I’m happy to share detailed work samples and process documentation privately under NDA. Just reach out.
