Risk Insights Dashboard
Crafted + Samba Safety
Overview
Let's provide actionable insights from complex data!
Problem
Goals
Samba Safety is an industry leader in driver risk management with their B2B solution helping companies manage and assess the risk of their driving fleet through the monitoring of driving records and algorithmic insights.
How can Samba Safety surface insights on professional driver records that can be used to reduce risk and improve road safety?
-
Simplify a complex process involving multiple products and disorganized record keeping for users to gain insights on driver data.
-
Use Samba's unique Driver Score to provide a quick view for customers take action to manage risk and reward safe drivers.
-
Introduce the Balanced Team approach to Samba Safety and establish a more refined design process.
Challenges & Constraints
Learnings & My Experience
-
Designing around unknowns in the data that was the foundation of the product
-
Ambitious sales goals and deadlines set prior to project kickoff
-
Limited access to users for research and testing
-
Technical constraints on data visualization that required the Balanced Team to creatively problem solve and iterate
-
Ramping up on a complex domain very quickly to understand the user and providing value to the client from kickoff
-
How to adapt a design sprint to be effective with a team juggling other projects
-
Bringing Sales and other teams into the process to build trust and learn together
My Role
Team Members
Stakeholders & Departments
Lead Designer
-
Conducting user research & testing
-
Creating design assets
-
Implementing UCD strategy & practices for Balanced Team
-
Collaborating with cross-functional teams on launch strategy
-
Directing & mentoring junior designers
Crafted Balanced Team:
-
Product Manager
-
Designer
-
Engineers
-
Paired with the client Balanced Team (~10 total)
-
CEO
-
CTO
-
EVP of Product
-
Director of Product
-
Director of Sales
-
Marketing team
-
Data team
How Did We Do It?
1
Research
2
Ideate
3
Prototype & Test
4
Iterate
5
Repeat
1
Research
During our research process, we employed various methods to ensure a comprehensive understanding of the industry and effectively meet the client's goals. First we engaged with stakeholders to establish alignment on objectives and processes, which is especially important as consultants entering a new industry and forming a partnership with the client. Next we conducted interviews with subject matter experts (SMEs) which helped our team ramp up very quickly to a complex industry and long-standing product line. The SMEs also gave us crucial information to prepare for user interviews by formulating relevant questions and familiarizing ourselves with the appropriate terminology.
For the user interviews, we brought in the Account Management and Sales teams to gain access to existing customers as well as a few prospective ones. We also used Pendo as an in-app tool to recruit users for interviews and future product testing. Implementing this research approach presents its own challenges, as it requires establishing trust in organizations unaccustomed to UCD practices. We successfully demonstrated the value of Discovery sessions by linking our findings and data to product and design decisions.
Interview Topic Map
Market Mapping
Important Takeaways from Samba Safety Customers:
-
Users rely on experts like Samba Safety to be the industry leaders on risk insights and best practices meaning education is an important aspect of new products, policy ideas or data
-
Users need to see data in simple ways that make the necessary action steps clear, numbers without context doesn't help them do their job
-
Users swivel-chair to a lot of software programs so a one-stop-shop would make their jobs easier and less info falls through the cracks
-
Users' main goal is to impact driver behavior therefore decreasing risk for their companies and the public
-
Users want to retain their drivers so they want to reward good behavior as much as correct bad behavior
-
Companies require reports on the driver pool and users need to know industry benchmarks to show success
-
The data has to be trustworthy and too much unorganized data causes more pain for them
-
Insightful data can help them create better company driver policies and track outcomes
The development of personas played a crucial role in our project. Considering the wide range of opinions about the users, we focused on validating proto-personas and identifying the most critical goals and pain points we aimed to address. These personas served as a guiding beacon for our team throughout the entire project
Original proto-personas to be validated
Research takeaways for validated persona
2
Ideate
During the ideation phase, we employed key steps to foster creativity and collaboration. We organized a design studio where the team generated a wide variety of product ideas. Workshopping the user's Jobs to be Done and completing a story mapping exercise helped us understand user needs and journeys. Sketching and engaging with engineers allowed for rapid exploration and iteration. We also evaluated our product risks and assumptions and devised experiments and other ways to mitigate and validate these through the next few phases of the process. These steps led to innovative ideas, a user-centric approach, and early incorporation of technical considerations into design and product decisions.
Design Studio
We wanted to ensure that the teams who would be using, marketing and selling the product felt that their opinions were heard and were confident in the work we were doing. From the initial project kickoff, we made it a priority to include key stakeholders and product advocates in collaboration sessions.
Consolidating ideas from a Design Studio can be one of the hardest parts of the process.
Jobs to Be Done & Story Mapping
These exercises are helpful to do in conjunction with a design studio with a core group of vested stakeholders and SMEs because it is such specific and tangible idea generation. It also gives a framework to communicate the necessary balance of user needs, business goals and technical feasibility to the range of roles in the room, many of whom have been dreaming very big about their product for years.
Story Mapping as a Balanced Team ensures the business, user and technology needs are all represented.
Sketching & Collaborating with Engineers
This is where the rubber meets the road (no pun intended, I say this on every project). Something I learned with this client is how to manage the challenge of consolidating the huge variety of ideas generated in a design studio or other ideation session into a workable initial prototype. Crafting and facilitating workshops effectively so the output is useable and all the participants feel heard is a skill I've honed over the course of my entire career.
Sketching is the fastest way to evaluate ideas with the Balanced Team.
Feature prioritization 2x2 done with engineers ranking Level of Effort vs Value
3
Prototype & Testing
We started our usability testing of sketches internally with SMEs to verify the accuracy of data and industry concepts before proceeding to customer testing. The existing Samba Safety product already housed a lot of driver data that customers needed. Samba created a new Driver Safety Score that compiled customer data from Incident Claims with the standard MVR Driver Score to create a more robust and informative at-a-glance Driver Safety Score of each driver. To learn more about that project as background to this one please read my Driver Safety Monitoring Case Study.
Insights Dashboard
One of the main challenges with the existing product workflow was users could view plenty of data but it was disconnected which made it difficult for them to determine what they should do about the data. They couldn't quickly tell the cause of low MVR scores so they have to chase more information in order to decide what to do next. They also couldn't easily track actions taken or improvements for the driver fleet overall or specific drivers to report to their management.
Existing driver data is one-dimensional and not clearly actionable.
The main goal of the Insight Dashboard is to surface the most important data in a way that users can distill into top level action steps and understand the state of the driver pool on any given day. Our testing was primarily focused on understanding what was most actionable data for different user personas and how best to visualize that data for quick understanding.
Round 1 Testing Priorities
-
Started w a pie chart for driver scores that certain stakeholders strongly advocated for which the team believed wasn't the best way to explain the data to users but we wanted to test it to verify our assumptions
-
Give high level metrics on the most important driver stats for a user to take action on each day
-
Kept the existing convention of listing flagged drivers so we could hear users' feedback about that feature
-
Showed driver trends over time and allowed users to filter dates and types of data
-
Allow users to sort by driver groups that they custom configure in the platform (i.e. by location or by sub-brand, etc)
-
Surface at-a-glance violation data
-
Gave industry benchmarks to compare against and use for reporting
Round 2 Updates & Testing Priorities
-
The pie chart was confirmed by testers to not be the correct data visualization so we moved to a bar chart for the driver scores with a filter to give more nuanced data
-
We learned what data is necessary on home dashboard vs what is clutter that can still be easily found by drilling down into a driver's profile
-
Collapsing the flagged drivers list into high level actionable metrics on the dashboard was a controversial change but repeatedly validated by users (this is why we test before building!)
-
Alerting users to new information is a common challenge for dashboard design so we probed this issue with users to understand what info to communicate, when and how
-
Added section to highlight changes in trends over time which users had a very hard time creating with limited reporting in the existing platform
-
Cleaned up UI elements like section headers and filters using design system
Round 3 Updates & Testing Priorities
-
Each round of testing we learned more about how to educate users on the meaning of new data incorporating the Driver Risk Score
-
Added more layers to the data in the bar chart for Driver Risk Score
-
Adjusted the colors for the two different scores so they didn't appear to align with each other which caused confusion in the first rounds of testing
-
Sorted the driver errors to be more clear on the type of error
-
Changed the industry benchmark categories based on feedback of what metrics are most important to track
4
Iterate with Balanced Team
There were many driver metrics to track and various ways to visualize that data. Intense debates ensued about the best way to present the data sometimes with an emphasis from stakeholders on "innovative" designs, but usually trying to reinvent the wheel is a fool's errand. This is common in growth-stage companies that want to set themselves apart in the market and have ambitious sales goals that can push people to create designs in a vacuum that look great but aren't technically feasible or won't actually perform well with users. This is why we advocate so strongly for a Balanced Team philosophy and a methodical problem solving process to save everyone time and money creating untested and unvalidated products that flop. We had to cut through a lot of assumptions and risks throughout this project to land on a solution we were confident in.
Data Visualization
One of the misses on this product from an accessibility perspective is the heavy reliance on the use of color to convey meaning in the data. If you aren't color blind then you have no problem using the colors on the dashboards to understand the information at a quick glance. However, that is not the case for everyone and I wish we'd been able to include better accommodations in our designs. This is a perfect example of how oftentimes accessibility is an afterthought or a "nice-to-have" in a product when there are many competing priorities and demands on the engineers' time.
Final MLP Updates:
-
After further discussion between the Balanced team and the data analysts we realized that the Driver Risk Score wasn't quite solidified enough to conceptualize at a high level like we were testing so we took that chart out for the MLP (Minimum Lovable Product)
-
We knew from several rounds of testing and similar data concepts (i.e. Credit Scores) that Samba needed to develop and include educational components so users knew exactly what the data means and how to act on it
-
Added a layer to the violations chart to show break down of specific violations that occurred which helps the user take action (i.e. training, communication, rules for discipline, etc)
-
Added driver group comparisons which we heard from users could help them understand variations in performance across groups and determine possible actions to improve specific groups
Creating a dashboard with multiple levels gives the user all the data they need without overwhelming them.
Design System
As we increased the fidelity of the designs we leveraged a design system, which provided a structured framework for design and engineering, resulting in increased efficiencies and consistency across the project. By utilizing a design system, we streamlined the development process and maintained a cohesive visual identity.
Utilizing a design system streamlines the process and saves everyone on the Balanced Team a lot of time.
5
Repeat Process
As the designs take shape and move towards production readiness we work as a Balanced Team to do rigorous prioritization of features and functionalities based on their level of effort and impact to the business and users. This prioritization allowed us to allocate resources effectively and focus on delivering the most valuable aspects of the designs first and then determining how to phase in more value over time.
Post-MLP Releases
It is crucial to design an MLP with the future in mind while considering what is on the roadmap and how the design can adapt. Anticipating future features and functionalities ensures that the MLP's design is scalable and can accommodate future enhancements without requiring significant rework or disruptions to the user experience. This forward-thinking approach allows for flexibility and scalability, enabling the product to evolve alongside user needs and market demands. By aligning the design with the product roadmap, the team can create an MLP that not only meets immediate user requirements but also sets a strong foundation for future iterations.
.jpg)
Users had many ideas and requests for tracking their action steps within the platform.
We found creative ways to incorporate calm and useful alerts to the constantly changing status of drivers.
What Does Samba Safety Have to Say?
Product Manager, Samba Safety
“I had not worked with a UX Designer prior to the Crafted engagement. The designer brought a much-needed perspective to the team. She was always focused on the user, workflows, and validating our assumptions.”
EVP of Product, Samba Safety
"We added a new project on top of existing roadmap commitments, Crafted helped us quickly ramp up a Balanced Team to define the problem, frame the solution and execute on the concept in weeks instead of months."
CTO, Samba Safety
"Working with the team at Crafted gave us a leg up through the definition and execution of some of our new and innovative products."
Deliverables
-
Competitive Analysis
-
User Personas
-
Generative Research Reports
-
User Journey
-
Design System Specs
-
Interactive Prototypes
-
Usability Test Reports
-
High Fidelity Mockups
Tools & Technology
-
Figma
-
Jira
-
Survey Monkey
-
Pendo
-
Miro
-
Zoom