Lead UX Designer, UX Researcher

ROLE

Aug - Dec 2024

TIMELINE

Contextual Inquiry, Heat Maps, Eye Tracking

METHODS

User Interviews, Guerrilla Usability Testing

OVERVIEW

Transforming Cox Enterprises’ TechHelp Express Kiosk 🖥️

We partnered with Cox Enterprises to help transform their tech-support kiosk (TechHelp Express) over the course of the Fall 2024 semester. We aimed to increase adoption rates of the existing kiosk by expanding its functionality to include other features other than just tech support. We researched about the company’s culture and connected with a variety of employees to learn about their daily work life to find areas that could be improved by our kiosk.

PROBLEM STATEMENT

How might we design the help desk remote kiosks to be functional, intuitive, and inviting?

RESEARCH METHODS

Competitive + Comparative Analysis

Goals

  • Gain a deeper understanding the strengths and weaknesses of different tech support modes at Cox, as well as different kiosks at other companies

  • Discover how these strengths and weaknesses can differ for the users vs. the IT team

Details

  • Our Cox shepherds gave us a detailed explanation of available tech support methods on the Cox campus; we included common and important measures of success, such as efficiency, accessibility, etc. in our study

  • Weakness: Only comparing a select few things, could possibly omit other features/examples, could be affected by creator-bias

Findings

  • Kiosk had the highest overall score in the comparative analysis, and the Tech Bar placed close second

  • Kiosk scored the highest in areas such as amount of service features, efficiency based on timing, and effectivity

  • Many kiosks are used for check-out retail and restaurant purposes, usually with specialized capabilities

RESEARCH METHODS

Landscape Analysis

Goals

  • Gain a better understanding of floor layouts in the main Cox Enterprises tower

  • Understand employee traffic-flow on campus, locating the most walked paths to discover optimal placement of future TechExpress kiosks

Findings

  • We learned that the current kiosk on the 9th floor is in a decent location, but not the most optimal

  • Near elevators and outside conference rooms are the most optimal kiosk locations as they tend to have the highest population density

RESEARCH METHODS

Survey

Respondent Demographics

  • Respondents: 18 Cox Enterprises employees

  • Age Ranges: 55% are 35-54

  • Departments: IT, Communications, Marketing  

Findings

  • Most Cox employees are satisfied with company tech support options

  • The majority of respondents prefer live messaging or in-person support

RESEARCH METHODS

Semi-Structured Interviews

Details

  • Who: 8 Cox Enterprises employees who have recently contacted tech support

  • My Role: I helped write the interview questions, moderated around half of our interviews, and wrote notes for each (180+) onto our Google Sheets to then be used for affinity mapping 

Findings

  • The vast majority of interviewees are satisfied by current forms of tech support at Cox

  • Some employees think the tech support process could be streamlined, specifically ticketing

  • Most technical issues occur in the office, such as connectivity issues with office equipment

ANALYSIS

Interview Affinity Mapping

Out of 8 interviews and 179 notes, we found various patterns across our data: 

1. I don’t like going out of my way to receive tech support

Analysis:

  • Employees often will deal with minor issues themselves, even if they can’t resolve the issue (in-office monitor, keyboard, mouse connection)

  • Employees are hesitant of tech-support: some are unsure if there are better options, some feel like getting support takes too long

2. I like getting my problems fixed in the best way I know how 

Analysis:

  • Employees think some aspects of tech support could be streamlined

  • Most employees enjoy the live-chat and call options to receive quick, personalized support

  • Those who prefer to submit tickets do so for documentation and team purposes

3. I hope the kiosk can solve my various problems on campus

Analysis:

  • Employees face the most technical issues while working on campus

  • Employees use apps to order food, reserve rooms and spaces on campus

  • A few employees like exploring and walking around rather than staying at their desks

  • Many employees dislike having to walk to the Tech Bar for support when they could call stationary from their desk

DESIGN

Wireframing

Based on all of our conducted research, we came up with a table detailing our top findings -> user needs -> design implications -> value case. We then began to start sketching and wire framing our ideas.

Wireframing

Lo-Fi Feedback Sessions

To gain initial feedback on our lo-fi concepts, we held 4 expert interview sessions where we asked members of our Cox team a series of questions to gauge overall opinions and usability.

Med-Fi Feedback Sessions

After we modified our wireframes to implement some of the feedback suggestions, we again held interviews with 5 expert users on our Cox team to receive more feedback before we transitioned into high-fidelity prototyping.

EVALUATION

Usability Testing @ Cox Enterprises

After our wire framing feedback sessions, we moved to high-fidelity prototyping. To get one last final round of feedback, my team and I visited the Cox campus and held guerrilla user testing sessions. We set up our prototype on a vertical, touchscreen monitor in the Tech Bar, a place where employees can get in-person tech support help. We had an iPhone set up to record users with permission, as well as eye-tracking glasses to discover hot spots of our prototype.

Testing Participants

32 employees at Cox Enterprises:

  • 18 male, 14 female

  • 12 user-based testing

  • 32 eye tracking and SUS forms

Testing Materials

We asked participants to complete a list of 7 tasks. Afterwards, we asked them to fill out a basic SUS form.

ANALYSIS

Eye Tracking

Cognitive Load

Outliers

We used a box-plot for the cognitive load indicator to find outliers in SUB01, SUB02, SUB25, plus SUB11, SUB30 who encountered tech problems during recording. The valid number of participants is 27.

Design Requirements

  • DR1. Lower the learning curve of interactive map for users

  • DR2. Simplify the streamline of Live Chat function

Task Duration

  • Task 5 takes the longest average time, at 22,493 ms, followed by Task 2 with 19,350 ms

  • Task 6 and Task 7 take the least time, at 8,749 ms and 6,091 ms respectively

ANALYSIS

Overall Testing Findings

Due to time constraints, we only had time to test our final prototype and make a list of needed changes, not actually implement them. Therefore, our final prototype exists as a v1 before changes from usability testing were made.

REFLECTION

Thank you, Cox team! 🎉

I thoroughly enjoyed partnering with Cox Enterprises and working with a team of fellow students in the MS-HCI program at Georgia Tech. Working directly with stakeholders led to so much growth and achievements I’m proud of!

Learnings

01: Keep a level-head under time constraints. We had so many due dates, so it was crucial for us to use our time wisely.

02: Adapt to your stakeholders’ voices and schedules. Listen to your stakeholders; be polite, yet also challenging.

03: Learn your team’s strengths. As my team had a range of backgrounds, it was important to cover each others’ weaknesses.

04. Feedback, feedback, feedback! Constant iteration added with constant feedback produces the best, most usable designs.