I’m a software product designer with experience in business analysis, software development, and game design. I have a masters degree in Human-Computer Interaction, and my best skills are curiosity, collaboration, and prototyping. I’m looking for an opportunity to help a purpose-driven organization make human-centered software. I’d love to hear any thoughts or advice you have for me.
"The heart of interaction design is not our interaction with technology, it's our interaction with each other through technology. It's not about us and objects. It's about us and each other." - Graham Pullin
Senior Business Analyst
- Simon Property Group, 2010-2017
Replaced an expense reporting system for 5000 employees. Improved user experience and engagement while saving money.
Launched a procurement system for 30 departments and 300 shopping malls. Improved plan compliance and reduced fraud.
Transitioned manual lease workflows to Salesforce.com. Improved speed and visibility of the company's core business process.
Consulting Business Analyst
- Fusion Alliance, 2007-2010
Developed requirements for replacing Simon Property Group's lease workflow system. Applying UX techniques increased engagement and built trust with skeptical business stakeholders.
Redesigned a clinical trials web portal, distilling actionable information from terabytes of data for 13,000 users.
Designed and built a streamlined UI to enter ECG diagnostic codes via numeric keypad, increasing throughput.
Re-engineered a clinical trials reporting system to increase throughput 200%. Improved usability, scalability, and compliance. Added transactional data processing to eliminate re-transmission of past data, improving efficiency for clients.
Expanded local kit production to global distribution. Saved $.5M annually, improved output 81%, better quality, faster delivery.
Users: Residents and visitors in the Fort Ben area
My Role: UX research, mobile product design, UI development, user testing, team leadership
For a Good Partner, Talk to Lots of Partners
Project Selection: First Semester, Weeks 3-5
Our team consisted of Evi, Monica, and myself: three HCI masters students. We reached out to 20 organizations about partnering with them for our two-semester UX design capstone project. We met with 5 organizations by phone or in person to discuss potential projects. We selected Schneider Geospatial and Arts for Lawrence because their augmented reality mobile project excited us. The timing was perfect: they weren’t planning to start development until the following year. They had the technical expertise to do the development, but were excited to draw on our design expertise to ensure an engaging experience.
Another team in our class hadn't found a project partner, so I introduced them to one of the partners we didn’t select. They completed a capstone project together. Further conversation with another potential partner led to my summer internship.
Welcome to Fort Ben
Research: First Semester, Weeks 6-11
We met to kick off the project with Schneider Geospatial, and their client for the project, Arts for Lawrence. We established a weekly meeting schedule, alternating between online Zoom meetings and in-person meetings at Schneider’s office.
Fort Benjamin Harrison was a US Army base in Lawrence, Indiana, operating from 1906 to 1995. 1700 acres of the fort became Fort Harrison State Park. The remaining acreage is residential, office, and retail spaces in the former military structures. Arts for Lawrence had received a grant to turn part of that space into a cultural campus, celebrating the fort’s history and the work of local artists.
Our mobile app project was a single line item from their grant proposal. It stated the project would include “A history-focused augmented-reality app.” We discussed ideas and goals for the app, including how it would interact with other features from the larger project.
One activity I led was a “premortem”. We pretended that we were in the future and the project had failed, and discussed what led to the imagined failure. This revealed many requirements that might have otherwise not been discovered. I followed up with the inverse, a “preparade” activity, in which we imagine the project being wildly successful. The premortem is a sobering exercise, so this brought everyone back to an optimistic mood. It also revealed future opportunities, such as spinning off the app as a product for other clients. Our partners were enthusiastic about the results of these activities.
To learn more about what people do when they come to Fort Ben, I arranged for three tours of the area from local guides.
Arts for Lawrence held a community meeting about the cultural campus grant. I led a presentation about our project. We requested attendees to fill out a survey about their relationship with Fort Ben and interest in the app. We received 31 survey responses.
Bringing History to Life
Analysis & Ideation: First Semester, Weeks 12-16
With data from our research activities, we constructed two personas. For each persona, we constructed an as-is journey map of a tour of the Fort Ben area.
We knew that our augmented reality feature would be to align historical photos in the app with the structures as they stand today. This was inspired by a photo in the Schneider Geospatial lobby. It shows a black-and-white photo of graduating chaplains, held up from the position in which the photo had been taken. We were all excited about this feature, but it would prove more problematic than we had anticipated.
Our team brainstormed other features. We sketched storyboards from the ones that seemed most interesting.
We discussed these storyboards with our partners, and settled on a feature to provide fictional characters as virtual guides. They would lead the users on walking tours, while describing their historical experiences at Fort Ben. This concluded our first semester on the project.
Experiments in Augmented Reality
Prototyping: Second Semester, Weeks 2-11
I led a task analysis exercise to identify the steps required to complete two features:
Take a day-in-the-life tour with a virtual guide
Overlay images from a historical timeline into an augmented reality display
From those steps we remotely collaborated on a low-fidelity prototype in Adobe XD, and demonstrated it to our partners. The resulting discussion revealed some concerns about the augmented reality feature:
Mobile device positioning sensors aren’t precise enough for the app to recognize when the historical image was aligned with the camera image.
As users attempted to align the images, they might intrude on private property, block traffic on sidewalks, or even wander into the street.
We provided a slider to control the opacity of the overlaid historical photo, but we had concerns that users wouldn't understand opacity.
About this time, Monica experienced serious health problems which temporarily removed her from the project. And Evi and I disagreed upon the technology to use for our next prototype.
Evi wanted to simulate the photo alignment feature using Torch AR, a prototyping tool.
I had already experimented with AR using Unity, and thought I could quickly code a working prototype that would be more enlightening than a simulation.
We agreed to each create our own prototype for one week. At our next partner meeting, we demonstrated both prototypes. Both of them worked, but Evi did not want to use Unity. She volunteered to develop our high-fidelity prototype if she could use the prototyping tool of her choice. She later abandoned Torch AR in favor of ProtoPie, and ultimately created our testing prototype with Sketch and Principle.
Testing in the Cold
User Testing: Second Semester, Weeks 9-14
While Evi developed the high-fidelity prototype, I wrote our user test protocols and recruited participants. The test consisted of the following parts:
Meet at the Fort Ben community campus. Go through a brief orientation and pre-scenario interview. Begin audio recording.
Day-in-the-Life scenario: Follow overhead map directions to a historical location. Experience the location according to a virtual guide’s description. Participant was to think aloud.
Historical Timeline scenario: Return to the starting location to look at information about the Theater at the Fort. Align the Theater’s current appearance with a historical photo. Participant was to think aloud.
End of session questionnaire, followed by a brief post-scenario interview.
We conducted a pilot test with our instructor. That November was already quite cold and windy, so we halved the walking distance to the first test location. I also arranged for permission to begin and end all tests in the lobby of the Theater at the Fort, to protect our participants from the weather. Two participants canceled their test sessions when they came down sick. One test session was postponed due to an early blizzard.
Ultimately, the participants for our six user test sessions were as follows.
Session
Persona
Age*
Gender
AR Experience
Notes
S1
Resident
Older
Male
None
S2
Resident
Younger
Male
Astronomy apps
S3
Resident
Older
Male
Worked on entertainment apps.
S4
Visitor
Younger
Female
Pokemon Go
Performed offsite at the IUPUI campus.
S5
Visitor
Older
Male and Female
Pokemon Go
Married couple performing the test together.
S6
Visitor
Younger
Female
Pokemon Go
*Compared to a reference age of 50.
We wanted testing insight from at least one of our classmates. But they were mostly foreign students without transportation to the test location. So we conducted test session 4 on our school campus, while the participant pretended she was walking at Fort Ben. The conditions were not ideal, but she provided valuable insight and data.
I facilitated each test session. Evi provided her mobile phone as the testing device. I compiled the quantitative data (time on task, numeric questionnaire responses) for our test report. We compiled notes from our interview responses, observations, and recordings. Then our team met online using Miro to group the notes by related themes in an affinity diagram.
I had provided voice recordings for the virtual guide, and we included audio effects of marching soldiers. Our test participants liked the immersive experience from these audio features.
The AR photo alignment feature intrigued our participants. Even though it had some issues, they really wanted it to work.
A Pleasing Finale
Delivery: Second Semester, Weeks 12-14
By now Monica had recovered enough to resume work on the project. She and Evi prepared our class presentation and project poster while I prepared the final report. Evi updated the prototype based on our test findings.
Our project partners told us they were very satisfied with our work:
I was fortunate to be able to work with Carl and his talented team. He is creative, engaging, and well organized. He has a unique ability to empathize with users, and turn this understanding into intentional experiences.
-Daniel Mallinak, Director of Software Development, Schneider Geospatial
Carl was thorough, creative, and professional. His team tested and delivered a prototype that I am confident will entertain and enlighten our visitors.
-Judy Byron, Executive Director, Arts for Lawrence
I connected Schneider Geospatial with a programming instructor who helped them start development on the app as a project for his students. And our presentation won a People’s Choice award at our school capstone event.
Telepathic
Game Design
Goal: Create an elegant cooperative tabletop game
When: Published in October 2019
Context: Independent Study in Serious Games, IUPUI
Users: People who want to play with a partner instead of against them
My Role: Product design, local and remote user testing, marketing, digital game development
I wanted to design a cooperative game to play with my spouse. I started with colored markers on index cards and a simple set of rules:
Two players must arrange tiles to match both their win conditions.
Neither player knows the other’s win condition.
The players cannot speak.
Initial playtests were encouraging. A friend volunteered the artwork. I designed the components and box in Adobe Illustrator. After playtests indicated the design was solid, I wrote the rulebook, laid it out in InDesign, and conducted blind playtests to validate it. I set up a shop page on The Game Crafter. Friends helped me create a promo video. By product launch I had tested the prototype with over 80 players, from family and friends to expert game designers and publishers.
For a Serious Games class in Fall 2019, I conducted tests comparing the experience of playing Telepathic face-to-face versus playing with a remote partner over the Internet. What I learned helped me to improve the manual and component design, and to start creating a digital prototype in Unity.
My spouse doesn’t care for the game, but my crowd sale sold over 100 copies.
My Role: UX research, web and mobile product design, local and remote user testing
Graspable Math provides an online tactile interface for solving algebra equations (see above). Their product had good traffic, but poor user retention. As a UX intern they asked me to improve the onboarding experience.
I documented cognitive walkthroughs of competing products. I calculated user churn from Google analytics data. I attended training sessions, surveyed target users, and interviewed expert users. I documented critical usability issues, such as:
The product was designed for large screens, despite the reliance upon tablets and mobile devices in classrooms.
Personal classroom materials could not be organized.
Public classroom materials were undiscoverable.
To discuss how to address the issues, I created a concept map and a story map. I created a wireframe prototype in Adobe XD which overhauled the main interactive canvas (see below, with menu open) and provided materials organization. I conducted four online usability tests to validate the prototype.
My reports offered Graspable Math many suggestions for improving their onboarding experience. They hired a UX designer as my internship ended, to carry on the work that I started.
Context: Interaction Design Practice class, final project, IUPUI
Users: Healthcare patients and providers
My Role: UX research, mobile product design
Healthcare portals connect patients and providers via the Internet. They provide an alternative to time-constrained and time-consuming communication by phone. To research them, we gathered information from the following sources:
Two healthcare portals
Five providers (a physician assistant, two dentists, a doctor, and a pharmacist)
One patient who manages 20 prescriptions for herself and her spouse
We discovered that portals are not realizing their potential:
Portal vendors focus on providing features instead of usability.
Portals often use medical jargon instead of information patients can understand.
Liability and regulations constrain portal use.
Portals lack the personal connection that patients feel with voice communication.
Based on our research we designed a prescription management subsystem for a healthcare portal. We selected this area because it is a common and important interaction involving multiple parties (patient, caregiver, physician, and pharmacist) in a workflow. We created storyboards, whiteboarded screens, and created a prototype with Adobe XD. Four HCI experts and a physician assistant evaluated the prototype, and provided much helpful input.
Much work would remain to complete the prescription subsystem. We learned some things about the needs of elderly users and the requirements of healthcare systems. We also gained practice with Adobe XD and creating validation scenarios.
Context: Prototyping for Interactive Systems class, final project, IUPUI
Users: AR game players
My Role: Mobile product design
Our final project team wanted to make a digital game that encouraged players to do the following:
Be physically active in the real world
Work together with their friends
We decided to design a cooperative AR game for smartphones.
Nexus is like a Ghostbusters tower-defense game you play in a park. Your team runs around to gather resources and build towers. The towers zap evil spirits that travel along ley lines toward your base. If too many spirits reach the base, you lose.
After some initial sketching and whiteboarding, I researched some other AR games and wrote a game design document. For our first prototype we drew some storyboards and selected some basic tasks:
Forming a team
Communicating with teammates
Gathering resources
Building defenses
We tested the prototype in Balsamiq with three participants. From those results we made a higher fidelity Sketch prototype and an experience video. Our class enjoyed our presentation, and several people encouraged us to keep developing the game.
120Water helps water utilities find, track, and remove lead service lines that deliver water to people's homes. The onboarding process starts with importing the client's service location data into our system. For a large client, this data could reflect 500,000 locations. This process was taking our client services team, on average, two months. We wanted to reduce the process by 90%, to about 4 or 5 days.
Our project team consisted of the VP of Engineering, three engineers, a product manager at about 30% utilization, and myself as product designer.
Why is this taking so long?
Root Cause Analysis
The import employed a user-hostile data template with no documentation, finicky column name matching, and unhelpful error messages. It didn't even match the export format for the same data.
The import process was slow and fragile. The process provided no notification on success or failure. Users had to break large files into 5000-record files to reduce the chance of failure, but each file had to be submitted one at a time. And if any file processed more than once, the engineering team had to find and delete duplicated records.
The root cause was code debt, caused by startup engineers solving existential crises. The import process needed to be redesigned and rebuilt from scratch, and it couldn't be done piecemeal. We estimated that our MVP was a three-month effort for one-third of the company's engineering staff.
XXX clever title here
Research and Ideation
I used multiple sources of information to inform our redesign:
Existing Jira tickets on the topic
Interviews with our client services team
Analysis of raw client data files
Import processes in other tools (HubSpot, Dromo)
Best UX practices for long & interrupted workflows
Based on what we learned, we decided upon a two-stage import process:
The user formats their input data according to the template and submits it. The system matches the inputs to known records by comparing data IDs, or comparing normalized addresses via the Lob API if it finds no matching ID. When complete, it notifies the user which inputs will create new records, which inputs will update existing records, and which inputs are not valid.
The user reviews the stage 1 outputs and approves or rejects the import. If they reject it, they can use the feedback from the system to correct the input data and try again.
XXX clever title here
User Interface Design
I started the UI redesign by wireframing the existing process in Balsamiq. I updated those wireframes with our improvements:
Added a batch ID to aid troubleshooting.
Added a progress indicator to Status column, and link to details.
I designed a new batch details page to show any warnings, and describe what would change if the file was committed. It allows download of the imported file, a file of all valid rows, and a file of all invalid rows.
I redesigned the import template to provide user documentation for each column. I documented rules for robust column matching. I wrote helpful error messages that include accepted data formats and values. I designed test cases for handling expected data conditions.
XXX clever title here
Product Outcomes
What I'd do now
Learnings
I'd do some things differently if I had a do-over for this project:
I'd validate the UI designs by asking our internal users to walk through some test cases. I was scrambling so fast to keep the engineers busy I forgot to even ask if I could do this.
I'd design just for an MVP instead of detailing so many data edge cases. That could have freed up time for UX testing.
I'd define more objective metrics to verify the project's success, and monitor them after deployment. That would have proved the project's ROI, and guided opportunities for future improvement.