Matthew Declercq

Determining the Client’s Expectations

We were asked by the MSU Libraries User Experience Team to research and develop various ways of improving the digital wayfinding and signage throughout the Main Library. Our research was focused on, but not limited to, the following:

• Who is using the system and why / why not?
• What are the pain points of using the system?
• Does the UI offer the information users want most, and in a clear way?
• What do target users actually want from the system?
• How does the digital system translate to the physical spaces?
• What are the main objectives and motivations for using the system?

Interviews / Surveys

We surveyed 25 random students and faculty then personally interviewed 5 of them. Our results indicated that the kiosks were exhibiting some of the problems we had initially anticipated. Many of the participants did not spend much time at the library and therefore did not have a need for any wayfinding systems. Additionally, not many people actually struggled to find resources in the library, leading to less than 25% of the participants ever having actually used a kiosk. However, despite the majority of people not having a use for a wayfinding aid, there was still a substantial population that could benefit from these kiosks. From our research it was apparent people were using the kiosks for what they are intended for, however it was also clear the user experience was in need of many improvements. The more shocking data point was that over half surveyed did not assume the displays in the library are interactive. This was something we felt the library should take action on immediately. We concluded there would potentially be a 50% increase in kiosk usage with some simple touch friendly signs added.

Card Sorting

Card sorting allowed us to get a better idea of how users would organize information without being pre-exposed to the interface. From our card sorting, we gathered that some of the information displayed on the kiosks wasn’t where the user would expect it to be. The arrangement of the cards provided us insight as to what aspects of the kiosks would benefit from having more emphasis, as well as where items should be nested.

Contextual Inquiry

What We Did // Participants were provided with a random picture of a book and its call number, or a random room name in the library. They were instructed to use one of the kiosks to assist them in finding the book or room, while explaining their thought process as they used the system. We recorded their actions, making sure not to interfere with their decision making process.

Good // In both cases our participants were able to achieve their task; however there were several moments where they were either frustrated or confused. The participant trying to locate the book was frustrated by the speed of the map animations. Perfecting the animation times was an easy fix that would dramatically increase the user experience. Additionally, by the time the participant got all the way up the stairs, he already started to forget where the map told him to go. This proved it can be a difficult task locating an item far away from the kiosk. The biggest struggle for the second participant instructed to find a specific room was scrolling through the long list of locations. Once he found the listing, he effortlessly navigated his way to the room.

Better // Both participants argued they were not a fan of the list system. They felt the ability to quickly type in a call number, or room name, would be a substantially quicker method than scrolling through lists. While the system does work, it is clear there are a few simple pain points that should be addressed. Additionally, we noticed the kiosk on the 4th floor only allowed the user to search for an item on that floor. It was not able to help locate items anywhere else in the library. Both kiosks on the 1st and 4th floor physically look the same however don’t perform the same. This could potentially cause an inconsistent user experience, causing frustration or confusion.

Heuristic Evaluation

What We Did // We took notes and critically evaluated the digital touchscreen. After doing so, following protocol, we re-evaluated everything based upon 10 UX guidelines for heuristic evaluation using our notes and heuristic skills to analyze our findings.

• Visibility of system status
• Match between system and the real world
• User control and freedom
• Consistency and standards
• Error prevention
• Recognition rather than recall
• Flexibility and efficiency of use
• Aesthetic and minimalist design
• Help users recognize, diagnose, and recover from errors
• Help and documentation

Good // Overall the screens followed MSU brand standards and consistency with the rest of the MSU Libraries media. The information was accessible and there was an “escape exit” made obvious in every corner to take users back to the homepage. The homepage showed the MSU Spartan icon, title, “welcome”, weather, date, time in header and scrolling announcement in the footer which gave off a welcoming, yet modern feel that is similar to a mobile app.

Better // There were some concerning glitches in the touchscreen interface. The call numbers A-C seemed to be the go to for when there wasn’t another option when trying to find locations or call numbers. This was confusing and would be better to not happen at all, or worst case to show an error message here. The map was useful, but would be much more useful if directions were given. The feedback screen needed to be redesigned because it directs users to the website, but the link was not actually clickable. A QR code or quick survey form would be more useful for users and the library to collect information.

Usability Test

What We Did // We created a list of tasks for participants to complete on the touchscreen based on thoughtfully created questions:

• How easy are the tasks to complete?
• What tasks were most difficult to complete?
• What frustrated the user?
• How long does it take the participant to complete the task?
• How does the user interact with the touchscreen?
• If prompted to repeat something, does the user remember?
• How accurate was the user in completing the tasks?
• What worked well on the touchscreen?
• What parts of the touchscreen need improvement?
• How helpful is the interface to the participant?
• Once the person understands how it works, do they seem to be more efficient and comfortable?

We then created a task list for the participant to do:

• Find an area of the library based on call number.
• Find an area of the library based on room name.
• Find a specific area on campus.
• Find information on a specific staff member.
• Reserve a room.

Good // Some notable things that went really well in this test was the process of reserving a room. The user felt it was easy to do and finished quickly. Secondly, the task of finding a room by location went well for our participant. She noted that the process to find a room by name went smoothly as well. The touchscreen was helpful in this case.

Better // We believed that the interfaces could greatly improve through some changes in the navigation and location of items. It is important that the user knows where to go on the screen in order to complete their task. We removed some of the less useful tabs and made sure the correct content was in the places where the user expected it to be. Our user got disheartened quickly when they could not find the thing they were looking for.

After providing the client with a comprehensive UX analysis of the current systems in place, we opted to create a redesigned user interface that reflected what we learned from our research. Our redesigned screens were imported into Adobe XD where we built a functioning prototype for the client.

Check out the high resolution redesigned screen samples below and watch this video to learn more.

Using the Flora GPS to Build a Consumer Wearable

The Adafruit Flora Wearable GPS is arduino based kit that consists of a microprocessor, GPS receiver, 8 neopixel LEDs, a battery pack and conductive thread. For this project our team of four worked together to prototype a consumer wearable product. It was crucial for us to utilize our combined skills and knowledge in different areas to accomplish the build.

The Jacket

Before sewing everything down to the jacket, we first assembled all of circuitry using alligator clips to ensure the code we developed was going to be functional. After some quick successful testing, the sewing began. We were thrilled to find out everyone on our four person team had sewing experience! This made what could have been a daunting process go very smoothly. Our jacket was designed to guide the user using two methods: the color of the LEDs, and the rate in which they blinked. If the user was moving towards the phone, the lights flashed green. If they were moving directly away from the phone, the lights would flash red. Any angle in between 0° and 180° would proportionally mix the green and red together. Additionally, the closer the user got to the emergency phone, the faster the lights would blink.

To Summarize

While building our consumer wearable device was a tremendous learning experience, our user testing proved true many of the issues we expected, given the limited amount of sensor input. Our biggest limitation was when our users were standing still. With only using a GPS sensor as input, it was impossible to calculate course over ground (COG), which our algorithm required to determine the color of the lights on the jacket. With the addition of a magnetic compass module, we would likely be much more successful in guiding our users to the emergency phones. We also determined while the jacket concept worked well given the equipment we had, our idea could possibly be better executed on a smartwatch platform. Identify User & Problem - Create User Flow - Make Prototype - Test. Do it all again.

Identifying the Necessary Improvements

After meeting with the client, we started by documenting the intended user flow. Then we moved on to creating an affinity diagram to help us organize our thoughts and intentions. We created a sticky note for each element of the site, then organized them how we felt would best enable the intended user flow. This is a one of the most cost effective ways to get a rapid prototyping project in motion.

Low-Fidelity Mockups

Once we were happy with the organization of our affinity diagram, we moved on to creating a few low-fidelity mockups, working with fellow team members for critique and approval. From there, we went on to creating our high-fidelity designs to present to the client.

MSU Libraries Digital Repository is a massive database that only continues to grow. Our biggest concerns for our client was how they handle their Information Architecture. We offered several different suggestions, but the most important being the need for them to develop a top level navigation categories, breadcrumbing and checkbox filters for search. As this was only a week long rapid prototyping project, it was important to focus on improving the elements that would make the largest positive impact on the site. That focus was user interface heavy, and developing a new information architecture hierarchy is something we believe the client would be better suited performing themselves. They know their content better than anyone.