Product Design Lead, Project Manager
Headings, paragraphs, blockquotes, figures, images, and figure captions can all be styled after a class is added to the rich text element using the "When inside of" nested selector system.
Translate Thicket's workshop process into a digital data collection interface with live insights and analytics
Develop a survey process that helps network members pursue the best solution for their community as a whole
A streamlined data collection process that could be performed and analyzed live
Thicket Labs's primary service was data collection workshops. In these workshops, we had been collecting information using post-it notes, stickers, and paper surveys. To analyze the data, it would all have to be recorded via video and photo as well as physically collected afterwards. Then, we’d transcribe those records into our APIs, which was a very time intensive and manual process as well. We needed a way to apply the workshop at scale and provide meaningful insights during the workshop itself. We believed this would allow for the greatest impact and feedback in the moment.
Although post-its and stickers are standards for most brainstorming activities, it has its flaws. This process makes it difficult to have all voices heard, it's time-consuming to collect the data, and work is often incomplete or incorrect. There's no way to check this at scale with one to two facilitators and 50 participants. This is what inspired us to pursue a digital collection process for workshop settings that could develop into a SaaS product of its own.
The core issues needed to be addresses were speed, limitations on insights, and scalability. Therefore, our main measurements of success were faster group collection, more inclusive perspective, and scalable for large groups.
The workshop process is directly pulled from a research methodology called Fuzzy Cognitive Mapping. The data collection output from this approach allows for predictive simulations, meaning that it can show how a system is affected in various scenarios. For example, if we increase exercise (or at a larger scale, provide fitness centers) in the system below, then the risk of heart disease will decrease. This helps individuals and communities test for the most effective and efficient solutions to pursue.
I designed The Possibility Engine to provide an interface for our APIs. This interface was for internal use only and was developed as a proof-of-concept prototype. There were a lot of bugs and hiccups due to constrained development resources and no external testing. We had drastically limited the possibilities (pun intended).
When I prototyped the Pocket Strategist (seen above) I shifted our focus to a more accessible, consumer-facing, and mobile iteration. I conceived a single user survey process to create a personal network map. This was a great way of gathering data directly from individuals, but it lacked the ability to input collective data to show immediate insights.
Moving forward, I took the best from both products – the robust collective insights of The Possibility Engine and the consumer-friendliness and personalization of the Pocket Strategist.
As opposed to the messy seven step process, I narrowed it down to five steps that could be performed non-linearly:
Ideate - Gather solutions from members to address network goals
Prioritize - Hone in on the right solutions based on community needs
Rate - Gather perspective on how impact and feasible solutions are
Relate - Discover how solutions may impact each other
Plan - Find the most effective and efficient solution through testing simulations
These five steps can be performed linearly measuring a single idea the whole way through the process or non-linearly measuring multiple ideas that will help contribute to the whole. This flexibility offered users the option to brainstorm their own ideas and weigh in on others at their convenience, allowing for quick interactions and returns to the platform instead of a full lengthy process each time as was the case before.
When planning out each step of the user flow, especially a non-linear one, I wanted to make sure the user would not get lost in the process. Two specific tweaks I made were to direct the users from sign up to creating their first goal and to add more instructional text and pop ups to help lead the user through the process.
The way this research methodology works is that the more feedback each user provides, the more powerful and accurate the predictive simulations and analysis can be. This meant that was extremely important to capture as much information as possible as accurately as possible.
These adjustments gave gentle direction, freedom to explore and helped build trust in the product – versus the previous prototypes with a forced linear path and no explanation as to why the information was important or necessary.
We wanted to find the least overwhelming, most flexible, and simplest way for users to view and respond to the data collected. We turned to massive visual data sets like Pinterest and Tumblr for inspiration. You can also see this applied to text based content in products like Google Keep. This interface allows for skimming and quick high level reactions as well as diving deep into the details if desired. We kept the masonry layout as a consistent element throughout the collection process and then diverged once a user entered the simulations and other data visualizations.