Overview
Spin is an electric scooter company based in San Francisco. As we rapidly expanded within the US, we started to receive pushback from local government and citizens with concerns on the negative impact on mobility within cities if Spin scooters aren't managed properly. To solve this, Spin leverages Fleet Operations employees that help manage inventory, charge and repair scooters, and make sure scooters aren't negatively impacting walkways and other parts of the city. In order to scale this, we needed to design a product for FleetOps specialists to track their tasks and manage their day.
Process
Our process started with research to understand how FleetOps currently work and opportunities for product features, synthesis to make sense of our research and understand our user, ideation and design, and then delivering final product designs.
Framing the Problem
When thinking about local community concerns, our data and policy team surfaced that many of the concerns were focused on Spin's effect accessibility within cities. At the same time, we were tasked with creating a product with core business value for Spin by empowering FleetOps (gig workers) to best do their jobs. Since these are critical stakeholders to Spin, we wanted to create a better way for these stakeholders to leverage data for their jobs in an effort to better work with community partners. Our challenge was to design a new data dashboard that helped two unique stakeholders make sense of data in a similar fashion.
Research
In the first phase of research, our team took time to delve deeper into the context of Spin scooters, analyze data, and understand users’ goals and pain points. Some methods we adopted include affinity diagramming, conceptual modeling, and visualizing data with scatter plots, bar graphs, etc.
In the first phase of research, our team took time to delve deeper into the context of Spin scooters, analyze data, and understand users’ goals and pain points. Some methods we adopted include affinity diagramming, conceptual modeling, and visualizing data with scatter plots, bar graphs, etc.
Semi-structured Interviews & Contextual Inquiries
I partnered with our researcher to speak with several fleet operations workers as well as accessibility policy employees in cities such as Pittsburgh, PA (a new launch city for Spin). Through synthesizing this research, we started to understand how these stakeholders make sense of data:
Chaos everywhere—these stakeholders are often confused about what they need to do each day.
Guess and check—both stakeholders are not used to data analysis, so they often are making decisions on a whim.
Internally motivation—the work these stakeholders conduct can be grueling, but they are driven by their goals.
Stories over data—both stakeholders seemed to focus on the ‘so what?’ of data. They were interested in actionable data.
This helped us set a north star for our product—realizing that we didn’t want to provide our personas with static maps and boring graphs, but engaging and digestible information that helps them accomplish their goals. We also created user persons based on this information to help understand our users better and design with them at the forefront of our mind.
I partnered with our researcher to speak with several fleet operations workers as well as accessibility policy employees in cities such as Pittsburgh, PA (a new launch city for Spin). Through synthesizing this research, we started to understand how these stakeholders make sense of data:
Chaos everywhere—these stakeholders are often confused about what they need to do each day.
Guess and check—both stakeholders are not used to data analysis, so they often are making decisions on a whim.
Internally motivation—the work these stakeholders conduct can be grueling, but they are driven by their goals.
Stories over data—both stakeholders seemed to focus on the ‘so what?’ of data. They were interested in actionable data.
This helped us set a north star for our product—realizing that we didn’t want to provide our personas with static maps and boring graphs, but engaging and digestible information that helps them accomplish their goals. We also created user persons based on this information to help understand our users better and design with them at the forefront of our mind.
Data-driven Design
I worked with our data scientist to make use of the scooter data we had available to gain insights and conduct preliminary data visualization explorations with Tableau—giving us a sense of what type of information and insights are relevant to Spin's data structuring. We noticed the impact different graphs have on different kinds of data, especially with a dataset with rich variables like geolocations and multiple related temporal stamps.
The key takeaway was that the flow of scooters has significant spatial and temporal patterns. These could be critical insights to provide for the users through an intuitive dashboard interface, without needing to run the data through dedicated software. Projects often don't get the chance to fully leverage data science resources, so in this case it was great to get a sense of what our product could be (a temporal-based data dashboard) before diving deeper into design explorations.
I worked with our data scientist to make use of the scooter data we had available to gain insights and conduct preliminary data visualization explorations with Tableau—giving us a sense of what type of information and insights are relevant to Spin's data structuring. We noticed the impact different graphs have on different kinds of data, especially with a dataset with rich variables like geolocations and multiple related temporal stamps.
The key takeaway was that the flow of scooters has significant spatial and temporal patterns. These could be critical insights to provide for the users through an intuitive dashboard interface, without needing to run the data through dedicated software. Projects often don't get the chance to fully leverage data science resources, so in this case it was great to get a sense of what our product could be (a temporal-based data dashboard) before diving deeper into design explorations.
Synthesizing with Affinity Diagrams
After understanding the perspectives, needs, and goals of our two personas, we used an Affinity Diagram to organize our analysis of the various personas and identify insights.
After understanding the perspectives, needs, and goals of our two personas, we used an Affinity Diagram to organize our analysis of the various personas and identify insights.
We found key patterns based on each persona’s objectives, goals, and challenges that allowed us to better understand what a successful dashboard might look like for them. We then created clusters of notes based on common patterns across the personas. Through visualizing the data, we were able to understand how each persona can benefit from knowing certain data insights, and how there could be an overlap across even relatively opposite personas.
We found key patterns based on each persona’s objectives, goals, and challenges that allowed us to better understand what a successful dashboard might look like for them. We then created clusters of notes based on common patterns across the personas. Through visualizing the data, we were able to understand how each persona can benefit from knowing certain data insights, and how there could be an overlap across even relatively opposite personas.
Making Sense of Research
One of the challenges of our project was trying to build one dashboard product for two relatively distinct users (accessibility policy advocates and FleetOps gig workers), so we used value flow mapping to synthesize our learnings into a comprehensive flow and identify opportunities for a better user experience. This was also a great tool for working across Spin to gain alignment on where to focus before jumping into design explorations.
We modeled the current state, highlighting interactions between the personas and their environment along with Spin scooters and users. This process allowed us to identify the problems that our personas were likely to face in their daily lives. These ranged from scooters parked on the sidewalk, not allowing wheelchair users to safely pass, to it
being difficult for broken scooters to be located, causing more accidents.
With information gathered from the current state model, we were able to add interception points to the model where we anticipated our dashboard to facilitate the personas. Examples include providing FleetOps with a map that calculates the best routes to get to multiple scooter locations and data for accessibility advocates that allows them to know if a scooter is parked on the sidewalk near her in real time.
Design Explorations
Sketches
Once we explored the problem space and aligned our team's understanding of key persona goals and pain points, I was ready to sketch preliminary ideas for design explorations.
Once we explored the problem space and aligned our team's understanding of key persona goals and pain points, I was ready to sketch preliminary ideas for design explorations.
