Intro
Our group was tasked with designing a vending machine in a market other than the English-speaking United States. Based on our research of countries, cultures, and populations, we decided to create a meal kit vending machine for young, urban professionals in Japan. With nearly 4 million vending machines in a country the size of California, we knew it would be an acceptable means to deliver our product. Automation is popular and highly trusted in Japan's commerce ecosystem. The key was offering a product that would satisfy user needs. Through our research we found that Japan's population is mostly urban, with more than half living in 1-2 person households. High density, low crime rates, and expensive real estate make vending machines cost-effective. Because a majority of single professionals do not have time or space to make nutritious meals at home (even though many would like to), we knew that a healthy, convenient, and easy product for them to cook in their own small kitchens would be a feasible vending solution.
Process
Primary Research
To better understand our problem space, we chose to triangulate our knowledge with three research methods
1. Interviews
Our team developed an interview script to guide our semi-structured exchanges with proxies for users in Japan – individuals in the Seattle area with ties to Japan. We conducted five interviews with user proxies to collect in-depth information directly from individuals about processes and experiences, including challenges and opportunities related to food, cooking, and vending machines. We especially appreciated the ability to probe into specific issues, with a topic as complex as food. Conducting interviews also served to establish rapport with subjects, opening up opportunities to test prototypes with these subjects at a later date.
2. Survey
We used an online survey to directly gather information from a relatively larger number of potential Japanese users about attitudes, practices and challenges related to food, cooking, and vending machine use. We developed our survey after conducting initial interviews, which helped us identify key topics and terminology. We distributed our survey via social media and relevant Reddit forums.
Our survey had 10 responses from individuals with ties to Japan, which we used as a proxy for users in Japan. The survey was helpful for understanding the scope of opinions on specific features to include, and obstacles to overcome, to inform our design process. Although this number of responses reflects our limited access to enough users (or proxies) for generalizing patterns, it did provide a few key features and issues to address in designing our vending machine. Some specific findings included:
- Respondents' desire to cook at home more often
- Extensive experience and comfort with using vending machines in Japan for food and beverages
- Importance of taste, cost, health, and speed of preparation while cooking
3. Contextual Inquiries
A third research method for this project would have been contextual inquiries, but the nature of the method made it unfeasible to conduct with proxy research subjects. Nevertheless, this is an ideal method because it would have provided us with tacit knowledge that may have been overlooked in both user interviews and surveys. It ensures that we're getting the underlying motivation of the user in the moment rather than relying on a memory of a past event.
Personas
We created two personas based on our primary research, they both live in the Koenji district of Tokyo, where we decided to place our machines. This decision was based on its location as a primary commuter district with a large population of young, busy professionals. Both our users are busy working professionals, who don't have a lot of time to meal prep. Haruto's primary concern was to curb his unhealthy eating habits, so he would like to cook more balanced meals. Kana enjoys cooking and collecting the recipe cards from our meal kits. While she doesn't always purchase a meal kit, she always stops by the machine to get meal inspiration.
Secondary Research
To better understand our Japanese users we conducted secondary research, which was more immediately accessible to us. One set of findings influenced the design and placement of the vending machine itself, such as the centrality of train stations in everyday commutes, and Japan's reliance on cash and SUICA cards for payment. A second set of findings helped us think through the specifics of the meal kits themselves, such as typical Japanese diets, small household sizes, and limited kitchen space.
Japanese Culture
We examined Hofstede's cultural dimensions for Japan. The Hofstede model of national culture contains six dimensions that represent dominant preferences that distinguish one country from another. The most pertinent for us was Japan's high uncertainty avoidance score. We incorporated this into our design by giving the user as much information as possible before making a decision. Key examples include requiring payment at the end of the interaction, so that the user can be certain about their decisions and displaying actual photographs of the products.
We also analyzed Japanese design trends across posters, magazines and digital interfaces. Some key findings were:
- Most designs use multiple colors and dense, busy layouts. Animation in digital products is popular.
- Scripts do not have capital letters or italics to indicate textual emphasis, so we would need to use color and size to create hierarchies for organizing text.
- Users from such high-context cultures scan screens in a holistic, nonlinear manner so our content could be placed more "freely" on the interface.
Design
User Flow
For better user experience we chose to begin the purchase with the customer selecting products first so they can track how much they will have to pay. This removes error cases in which the user may not have enough money or has paid too much before selecting.
- Touch screen
- View products
- Add products
- Make payment
- Collect product
Prototype
The home screen tells users to touch to start when they approach the machine. The graphic is animated when the machine is not in use to attract attention. The product screen displays photographs of the real food options and shows the meal ingredients. (Keeping with the uncertainty avoidance dimension in Hofstede's Cultural Dimensions for Japanese culture.) Users simply touch buttons to add, remove, cancel, or complete the transaction. The total cost is displayed as they make their selections. (In line, again, with the uncertainty avoidance dimension). When done selecting, users choose to pay with either cash or IC card. The animated screen returns with a "Thank you" message telling them to collect their food and change, if due.
Design Mockup
Our prototype implements cute, playful, busy, happy, and colorful into the machine and packaging. Kawaii characters, which are also animated on the home screen, are found throughout. There is a positive, upbeat slogan in English and we include a recipe card as part of the product packaging that cooks can save for future use.
Outcomes
There were several issues with our research process, key being the low response rate to the online survey. Ideally, we would have been able to relaunch it to get more responses to inform our design choices. Also, we would have conducted contextual inquiries with users in Japan, in several contexts (while using a vending machine, cooking, and buying prepared food from other sources) to simultaneously observe their behavior, and probe into their thought processes. For further research, we would have liked to conduct usability testing with our prototype to understand the ease of task completion as well as the overall user flow. This would then inform further iterations of our design.
If interested in the phenomenon of Japanese Vending here is one of many articles about the topic: Architectural Digest: "Japan's Vending Machine Designs Are Like No Other Country's"
