Sojourn Fare

PROJECT INFO

A 3-week UX design sprint with Daniel Nullmeyer and Josh Chiang for a food tech startup in Chicago. Tools included Sketch, Illustrator, and InVision.

OVERVIEW

Sojourn Fare is a nascent food tech company based in Chicago, building “hardware and software that automates and assists the process of cultivating mushrooms.” Why mushrooms? Paul Stamets answers that question pretty convincingly in this TED talk. The company is currently prototyping an onsite growing unit that produces a continuous supply of mushrooms week after week, and the model has been successful in a large corporate cafeteria in Chicago. They are now in the process of optimizing the system and refining the business strategy to expand into new markets. Sojourn Fare asked our team to develop an internal tool that helps drive insights and optimize the growing process. At a minimum they wanted a mobile platform for monitoring and controlling the environmental conditions in each grow unit.

Our proposal for this tool, Spore, not only allows the Sojourn Fare team to manage the environmental conditions of multiple grow units in different locations, it provides features and functions to track yields and optimize mushroom harvests. With these added features, the design strategically positions Sojourn Fare to tackle a completely open market: medium-scale grow units for gourmet mushrooms in grocery stores and large kitchens. We collaborated with our client to clarify the business strategy and design a tool that is highly tailored, yet flexible enough to allow for adaptation and evolution.

RESEARCH

Building a product for an internal team required a strategic approach to research, with a particular emphasis on understanding the market for gourmet mushrooms and the inner workings of Sojourn Fare. The following objectives guided our research: Understand the domain and potential market competitors. What makes this platform unique? Learn about the internal team. How do they work, and what type of system do they need?

We evaluated a number of existing platforms specifically designed for growing mushrooms and, more broadly, platforms for growing other types of plants in controlled environments. Freight Farms, for instance, is a company that is exploring aeroponics and hydroponics in repurposed shipping containers. In trying to identify qualities that would make our platform unique, we uncovered a significant gap in the market, and a promising direction for Sojourn Fare: medium-scale units in grocery stores, where consumers can get direct access to fresh gourmet mushrooms.

The market for consumer-facing products is almost saturated. A number of companies, such as NIWA, that currently offer home-grown marijuana kits are in the process of adapting their systems for mushrooms.

To some degree, our client was already aware of this market opportunity, and its potential impact on the Sojourn Fare business model. However, in order to develop an application that could support this “Whole Foods model”, we needed to learn more about the internal team and build consensus around key operational processes and procedures. To do this, we scrapped our formal client presentation in favor of a hands-on workshop, where we jointly created an ideal user journey map.

TOP: Using workshops and design exercises in place of traditional presentations allowed our team to rapidly align our thinking and move our design forward in a compressed timeframe. Roman and Valerie from Sojourn Fare are making comments on our journey map. BOTTOM: It was particularly helpful to get comments from the development team, who provided insights about what needed to happen on the back-end for each step in the user journey.

We derived two main insights from our research. First, Sojourn Fare needed a way to keep track of system maintenance and cleaning records to comply with USDA and Health Department regulations. If they are operating out of a grocery store, regulatory compliance will be a huge part of their business and something for us to streamline in our mobile application. Second, analytics related to harvest optimization are better suited for the back-end of the system, where there is more room for experimentation with data. The internal team will use the mobile application to input data, but mobile isn’t the right platform for manipulating and analyzing large datasets.

PROJECT SCOPE

From our research and through interactive workshops with the client, we established a clear direction for the project and built consensus around the company’s immediate needs. We learned that we weren’t just designing an environmental control system; we were designing a platform that enables Sojourn Fare to capitalize on a specific market opportunity. The challenge was how to design the platform to help the internal team control, monitor, and optimize each unit’s environmental conditions and growing processes as the company prepares to bring their product to market. With help from our client, we developed three guiding principles for this challenge:

1. Let me see it grow. This isn’t a home automation system or even a farm management tool. We’re growing mushrooms in a box, and that’s awesome. We need to reflect the beauty of that process in the user experience.

2. Keep it ship-shape. We need to show how the system is running, and offer the functions and features necessary to optimize operations. Specifically, we need to be methodical in our approach to handling information related to regulatory requirements.

3. Don’t make me panic. When we present system status notifications or error alerts, offer actionable solutions instead of just flashing panic lights. We need to prioritize errors that are serious over routine maintenance reminders.

The problem statement and three design principles helped establish much-needed constraints for our team and set the tone for concept development. At this point, we had a clear direction and a framework for evaluating our design concepts.

EARLY CONCEPTS

To achieve a minimum viable product, we needed three main components in our application: a way to control and monitor environmental conditions in each mushroom growing unit; a way to manage harvests and record yields; and a way to track maintenance activities. Even within this fairly well-defined scope for the project, we developed three approaches to our design:

Concept 1: A responsive web app allowing users to manage multiple units, and emphasizing harvest management over environmental controls. We decided that this prototype would allow for maximum flexibility and give users the ability to experiment with the mushroom growth process. This was the most inward-facing of our concepts, and was geared toward helping the company fine-tune its operations in advance of developing a more consumer facing product.

Concept 2: A native app featuring a bottom tab navigation system and a store for purchasing new growing substrates. We included the store in this concept to test whether the internal team needed a way to order new substrates from the mobile platform, or whether that aspect of the business could be handled by another system. Including this features sparked a number of great conversations about how our platform fit into the larger ecosystem of products Sojourn Fare uses to manage its operations.

Concept 3: A hybrid mobile app with similar interaction patterns to popular platforms like Nest. This prototype also allowed users to control multiple growing units. We intended this to be the simplest of the three concepts, but also the most closed. We didn’t include any features that would allow for error checking or diagnostics. As a result, this was the most consumer-facing of the three concepts.

During concept testing, we learned that having the ability to manage large numbers of units (20-30) is essential to making the Whole Foods model work. In addition, the entire Sojourn Fare team liked having direct and immediate access to all of the app’s features through the bottom tab navigation system.

Borrowing an interaction pattern from Slack, we included a side drawer that allows users to switch between units in our converged prototype. After showing the prototype to Sojourn Fare, we discussed the site map and allowed the company to comment on each screen.

REFINEMENT

One of the benefits of working in close collaboration with an internal team is the ability to test concepts quickly and achieve rapid alignment. After two rounds of concept testing we had a solid direction for our prototype and began the process of refining interaction styles and cleaning up usability issues, developing a clean set of mid-fidelity wireframes. From our conversations with the Sojourn Fare team we learned that they needed a prototype to control conditions and track harvests at the unit level, but didn’t need any extra granularity in the data for experimentation. At one point, we discussed having the ability to record yields for individual substrates (blocks) in each unit, as a way to set up experiments and see if small variations in nutrient composition affect mushroom growth. We included this feature in concepts one and three, and Sojourn Fare initially expressed interest in the idea. However, once we started talking about the logistics of managing multiple yields across multiple units, we determined this feature introduced unnecessary complexity to the platform. Having longer conversations during concept testing allowed us to thoroughly vet each design decision and evaluate the importance of each feature.

We found that testing two people (Romand and Valerie from Sojourn Fare seen here) at the same time, rather than testing individuals, led to interesting discussions and deeper insights about how the platform supports Sojourn Fare’s day-to-day operations.

Given the complexity of the app, and the number of features that it needed to accommodate the Whole Foods model, our biggest challenge was reducing the number of screens–the number of distinct steps involved in each process. We found that it was useful to revisit our earlier journey maps to determine which steps and which process could be combined. Harvesting the mushrooms and cleaning the unit, for example, always happen consecutively. Therefore we combined these processes in a single flow and reduced the number of distinct screens in the app.

We also built out a feature that allows users to define environmental setpoint schedules for each unit. In our original concepts, we experimented with graphical inputs (e.g. Nest), text inputs (e.g. Wink), and different ways to display deviations from predefined setpoints. We settled on a system of “rule cards” with each card representing a specific setpoint change during a 24-hour schedule.

Continuing our practice of using workshops to discuss each prototype in depth, we worked with Sojourn Fare to address usability issues on each screen. Ultimately, having a visual record of these conversations was invaluable for the UI team.

FINAL DESIGN

We used a bottom tab navigation system to organize the application into four distinct sections: monitor, control, crops, and maintenance. The monitor section allows the user to view histories for temperature, relative humidity, carbon dioxide, and light levels. The control section in turn allows users to adjust those same parameters and override default setpoints and schedules. The crops and maintenance sections respectively track harvests and system maintenance issues. The prototype also features a side drawer that allows users to browse through and select other growing units.

Having four relatively self-contained sections of the app establishes a simple framework to address questions about development and product evolution: What are the most essential features for a minimum viable product? What degree of customization is possible for different members of the Sojourn Fare team. How would this platform change as a consumer facing app? As part of our final presentation, we laid out a roadmap for releasing the product–called Spore–based on this framework:

Spore 1.0: The MVP includes the monitor and control sections, which allow the team to diagnose system malfunctions and control environmental variables.

Spore 2.0: This is our proposed design, including all four primary sections of the app: monitor, control, crops, and maintenance.

Spore 3.0: To adapt this product for the consumer market, we will remove the maintenance section and streamline features, such as scheduling, that are designed for more advanced users.

Overall Sojourn Fare expressed enthusiasm for the design, and appreciation for how the design process clarified the company’s business goals. We note only delivered a comprehensive prototype for managing the grow units and harvests, we created a flexible platform that could easily adapt to Sojourn Fare’s evolving business model.

For ease of use, the most frequently used features are always within one or two clicks from anywhere in the app. We pushed more advanced, and less frequently used, features onto secondary pages.

UI HANDOFF

Our deliverables to the UI development team included all of our research, sketches, clickable prototypes, and annotated wireframes. The UI handoff also presented a great opportunity to get objective feedback on our design and explore potential refinements to the user experience. Most of our design decisions persisted through to development. Further user testing validated the information architecture, navigation strategies, and core feature set. However there was one substantial change that emerged as a result of further user testing: combining the monitor and control tabs. This change helped clarify the relationship between existing conditions and setpoints, and didn’t introduce more clicks to typical user flows.

As a result of further usability testing, the UI team consolidated the monitor and control pages, with a system of cards to view each parameter. UI screens developed by Kit French.

My biggest takeaway from working on this project was the value of using workshops rather than presentations to actively engage clients and build rapid consensus around design decisions. This format doesn’t work for every client, but when used in the right situation it greatly accelerates the design process. The workshop artifacts were also a great record of our discussions, and a huge asset for the UI and development teams during the final handoff. In addition, this project was a great opportunity for me to see the connections between design and business - specifically how companies can employ a user-centered design process to drive business insights and guide product development. Good design should be proactive.