Cyrena Johnson 
Pitch Deck & Case Study; Visual Designer; Researcher
Rogue Stoppard
Animator & Video Editor; Visual Designer; Researcher
In a team, design an environmental interactive data visualization experience by allowing users to interpret data in new ways in a specific space.
This project was completed over the course of five weeks.
Modern ski resorts are still relatively low-tech, and could benefit a lot from the use of AR technology. Additionally, ski goggles are already used almost universally, and so introducing AR into those goggles would be rather unintuitive. There are also many current pain points associated with ski resorts that could be potentially solved via AR, which gave us room to explore different focuses.
When we decided upon this topic, we came in with a lot of assumptions about the current pain points of ski resorts and the user base. For example, we assumed that people at ski resorts had trouble with navigation and communication with other skiers, and that they would be interested in further gamification of skiing, by way of games and statistics. We also assumed that people who frequented ski resorts were split 50/50 between ski and snowboarders, that people often went skiing in groups, and that people wouldn’t mind renting expensive equipment like AR goggles. Some of these assumptions were later proven correct and some were proven false.
In the following research phase we addressed these assumptions head-on, in two ways. Half of our time went into figuring out who our competition was and what pain points they had decided to address, and the other half was spent surveying our audience and testing our assumptions. By combining these two research methods, we hoped to figure out how to best the competition and offer what they did not.
Smart helmet/goggles combo; AR features; Communication capabilities; Maps & Navigation; Room for improvement in visual design.
Tracking technology for friends; Social Media aspects; Check in at resorts; Leaderboards; Track your whole day and earn rewards; Resort Guides; Lacking AR.
Not quite as thoroughly built out like SNOCRU; Mostly focused on conditions; See conditions for resorts in real time; See conditions before you ski.
Current speed aspect; Navigation, finding runs, etc; Music control; Smartphone connectivity; Buddy Tracking; Lacking visual design.
In order to reach a large audience and gauge the general opinion of people who frequent ski resorts, we decided to put together a survey and distribute it as wildly as we could. The result was a survey with 13 questions, which we distributed mostly via facebook and reddit, and through classmates. We got 125 answers to our survey, and learned a lot about what our target audience was like, and which assumptions of ours were true. 
Based off of what we learned from the survey, the target audience is a skier between the age of 18 and 25, who doesn’t rent equipment, skis with a group, and is interested in games and stats. We also know that they have at least some issues with navigation on the slopes and losing their group. 
With the knowledge we got from this survey about the target audience and their desires we were able to determine what to focus the product on and start designing with specific goals in mind.
Make experience as easy and fluid as possible; Enrich/Add to experience through games, stats, racing, etc; Add sense of safety experience.
Set brand apart from others, we want to be the number 1 choice; More sense of safety provided to users; Cohesive experience throughout products, etc.
Availability to buy/rent for range of users; Appeal to large range of audiences; Attractive/intuitive product and interface.
Enhance the ski resort experience by using integrated GPS AR goggles to increase connectivity and ease of navigation for visitors on the slopes.
We came to the conclusion that the problems Mjöll should solve were navigational and communication issues on the slope, as is written in the product statement above. We brainstormed a list of relevant pain points, listed here:
1. Losing your group on the slopes
2. Not being able to locate the right lift to get on
3. Not being able to find a run or lodge
4. Not being able to call your friends without taking out your phone (or at all)
5. Not knowing that your friends are also at the resort
Our solutions to these issues were multi-faceted, to ensure that the users had multiple ways to do similar actions and could therefore behave in the most natural way for themselves. Mjoll can be controlled via eye tracking technology (look at something and blink twice), as well as voice recognition software (“Hey MJ...”). 
To solve the navigational issues, there are pop-ups that display the names of runs at the top of said runs, and pop-ups over lifts that state which runs they connect with. You can also say out loud where you want to go, and pink arrows will show up on your goggles to direct you there. For alternate routes, there will be smaller blue arrows popping up on occasion as you move. If you don’t know the area as well as you’d like, and want to see all the runs and lifts and lodges at once, you can pull up a full-size map that will hang in the air in front of you. It won’t block your vision, and you can walk up to it and away from it as you like. 
There are also several solutions for the pain point of connecting with others. The primary one is the ability to create and manage groups on the Mjoll phone app, which connects with the headset. On the app you can make and join groups from your contacts, or join open groups with strangers. Each group stays as one of your ‘active’ groups for a day or until you leave it. On each recurring ski trip you can rejoin old groups or create or join new ones. When you’re wearing your goggles you can see the locations of your group mates who are nearby you, as little location markers show up over their heads. You can also search for group mates who are farther away and get directions to them with pink arrows. Additionally, there is a call function in the headset itself, so you can call and talk to your groups while you’re skiing or when you’re looking for them. You can also chose to race one of your contacts or a stranger in racing mode, and check your stats to see your daily leaderboard rank.​​​​​​​
Our overall goal with the visual design was to make something modern, minimalistic, sleek, and unobtrusive. As both of us weren’t very practiced at designing for AR, it took a while for us to achieve that goal. With each iteration of the design, the look got more simplified and futuristic. Another aspect of the design that we experimented with a lot was the organization of the buttons on the goggle screen, and the flow of information. With this as well, we ended up with a more minimal design than we had in the initial concepts.
1. Testers were consistently mentioning that the pop-ups on the screen, specifically big ones like the main map, needed to be made transparent or smaller to avoid covering the vision of the wearer.
2. Some things, like information on how the eye tracking selections worked, needed to be explicitly explained to the user. 
3. The organizational structure of our buttons on the goggle screen was overall too complicated, and it was suggested that we simplify the number of levels to click through.
There was also fairly consistent confusion about the ability to build groups out of older groups. This issue, however, could have been because of the low-fi nature of the paper prototypes, so we decided not to scrap it quite yet. (However, it did get scrapped after the second round of testing.)
The biggest pieces of feedback we received in the second round of user testing were:
1. The design was overall still too clunky, dated, and large. The arrows directing the user, the nav, the icons, the type face, and the pop-ups all needed to be changed.
2. Users would want more than one way to make selections using the AR goggles, and that we should consider using something more besides just clicking via blinking. Specifically, voice recognition was suggested.
In the early stages of our design, we considered several alternate routes that eventually weren’t implemented. The first big change we made was in deciding not to making the Mjoll goggles rental-based. The initial idea was that, because the goggles were likely to be rather expensive, users would need to rent them once they had arrived at the resort. This would also mean that each pair of goggles would only need to contain the map information for one specific resort, rather than all of them. However, when we got our survey results back and realized how few people rented equipment, this idea was discarded in favor of purchased goggles.
The second big change was in adding the connected phone app. This was added because we realized early on that it would be an issue for people to need to plug in complicated information like who to add to a group, or even their own login info, on AR goggles without a keyboard. Adding the phone app also had the added benefit of allowing the users to communicate with their groups when they weren’t wearing their goggles, like they would be in the lodge or cars.
The third big change we made early on was to move away from having recommended routes down the runs themselves. The initial concept for that was that the AR goggles would route skiers and snowboarders around each other to avoid collisions. However, in doing preliminary sketches, we realized this would be far to much information to have on the screen at once, and would likely be more annoying than useful.
There were less big changes made in the later stages of the design, but the changes made were still significant to how the system worked as a whole. The biggest such change was the previously mentioned inclusion of voice control in the AR goggles. This change allowed us to simplify the design a bit more, because we realized that confused users could always use the voice option in a pinch. Even having the voice option meant that far less people would rely on the nav bar to control their goggles, which made the nav bar overall less important.
The second big change later on was the switch from a static map projected on your goggles, to one that appeared to be hovering in the air near you. This decision allowed us to keep the map at all, when previously we were debating deleting it entirely to avoid obstructing the user’s vision.
Before making the video, we completed an additional round of research to gather design inspiration, and re-hauled the visual design of the app and AR display to match the feedback from our second round of user testing. The final iteration of our design was created almost entirely in Adobe After 
Effects, and so the best way to view it is by clicking the link to the promotional video below. Because of the nature of our product, the user interactions are highly animation based and static frames don’t fully convey how it appears in use. However, it is also depicted in screen caps below to highlight specific features.
In this project, we had to overcome many visual and organizational challenges. For example, there was a learning curve for both of us in figuring out how to make a sleek modern design for an AR device when neither of us had much experience using AR. Additionally, it was a challenge figuring out how to build out a large list of features on AR goggles and keep them well organized enough that the navigation wasn’t frustratingly slow for the user. We also had to find the balance between having intuitive, useful way-finding and an overcrowded visual design. Because Mjöll has many features, it was also a challenge figuring out how to depict a complete version of how our system worked without being able to fully build them all out.
This project helped us learn more about designing for AR and data visualization, and how to do complicated 3D effects in After Effects. It also helped us learn how to make decisions on a very open-ended project with many viable paths to chose from. As it was a long-term group project, it helped us learn how to better divide and conquer tasks, and figure out when we could work remotely versus when it was better to make decisions in person. It was especially interesting learning how to work around technical limitations, both in terms of building out our promotional video and in designing a product that existed, for the most part, without a keyboard and had to integrate into a variable outdoor location.
Back to Top