Revamping Ford’s analytics dashboard to enhance the vehicle quality.
As the sole product designer, I redesigned Ford's dashboard to solve key usability issues such as complex drill down and cluttered interface, boosting user satisfaction from 36% to 84% within 8 months of launch.
Timeline
August 2020 - June 2021
Client
Ford Motor Company
Role
Product Designer
Team
1 x Product Manager
2 x Data Scientists
4 x Software Engineers
Problem
Ford’s critical analytics tool, held back by outdated UX.
The dashboard had become an important decision-making tool for 5000+ engineers at Ford to improve vehicle quality, increase customer satisfaction, and stay competitive in the automotive industry.
Despite its importance, the dashboard was plagued with user complaints that it was too complicated to use. Users found it difficult to navigate, filter, and drill down to find the information needed — not meeting the core objectives of a dashboard.
Design Goal
How might we simplify the complex dashboard to help Ford engineers easily understand, analyze, and present key insights?
Solution - Before & After
No overview page
The old dashboard did not have a consolidated overview page so users had to take extra efforts to monitor and access important data. Users have to constantly shift between pages to search among the mixed data.
View essential data at a glance
The new dashboard layout is designed to present the data in a quick, easy-to-scan format with the most relevant information understandable at a glance. It displays elements (chart, table, last visited filters) in a minimized view with the ability to customize and drill down for more details.
No user onboarding
Many users found the original dashboard hard to learn. Even the power users often felt lost and did not have an easy way to get help. It also had a lot of features that users were not aware of due to a lack of user education.
Before
The new design has a friendly onboarding that helps users get started with the dashboard and reduces the cost of learning.
Clear onboarding & guidance
After
Too cluttered interface
The dashboard had too much information on a single screen which made the users difficult to focus on the essential information and miss the key insights.
Before
Prioritize in a visual hierarchy
After
I organized the elements and subsets logically to minimize users’ cognitive load and time spent searching.
Before
Hard to find specific data
The initial dashboard had lots of unused space, while customer data appeared densely packed. Data points were not clearly separated, causing labels and tags to merge visually. As a result, users experienced an increased cognitive load to locate and comprehend the necessary information.
Reduced complexity and user effort
After
I discovered that users want to know exactly where they can find important information and need an easy way to search for it. To address this, I employed a table layout, which optimizes space usage, minimizing the need for scrolling. This arrangement presents information in a manner that allows users to effortlessly read and optimize the time spent searching.
Primary User Groups
Identified primary users using daily usage rates.
It was difficult to scope down the primary user because there were 5000+ users and 30+ user groups. I used the usage rates as the metric to determine the power users:
Quality Analyst, Vehicle Engineer, Quality Engineer, Quality Investigator, Design & Release Engineer
User Research
Conducted user interviews to uncover actionable insights.
Most common pain points
No overview page
Users could not see key data at a glance and had to constantly shift between pages to search among the mixed data.
I had a good starting point as there were users using the existing dashboard. Through conducting contextual interviews with 20 users, I was able to learn about their goals, mental models, needs, and pain points.
Hard to drill down to find specific data
“I want to deep dive, filter by any specific topic, model, feature, anyway it can be possible to filter. It is all about trying to get to the specifics.“
High learning curve and no user education
The dashboard had lots of features and capabilities but did not have any user onboarding or help page.
Visual clutter
Visual clutter, especially on the header, caused cognitive loads to understand and use the dashboard.
Personas
Translated the pain points into the form of questions to capture the user’s relevant characteristics.
Solution Generation
Triangulated research insights to generate effective design requirements.
As a team, we brainstormed design requirements to address each pain point. Then we did affinity mapping to discover patterns and themes in people’s thinking and see which ideas were most common.
MVP Prioritization
Prioritized design decisions for MVP.
We generated lots of ideas and had to scope down for the MVP. I facilitated a scoping workshop with teams associated with quality analytics and conducted a 100 Dollar Test to gain insights from all stakeholders' perspectives. With a risk-value matrix, we prioritized design decisions by considering a balance of user needs, business goals, technical constraints, impact and feasibility, and user feedback.
User Flow
Re-designed the user flow to enhance the ease of movement and create an intuitive interface.
Visual Benchmarking
Dark-themed UI help reduce eye strain and support visual clarity within the interface.
I explored various styles for the dashboard interface to determine the overall look and feel. Because the dashboard is often used for several hours straight, I considered using dark-themed UI.
I also considered a single-page style that displays all the relevant data on a single page and allows users to click on specific sections to view more details or drill down.
Wireframing
Explored the most intuitive ways to represent data based on the context.
I tried using a gauge chart instead of a pie chart for the Net Promoter Score (NPS) to easily understand what makes up the score (promoters, passives, and detractors) using the sequence of colors. I went on and conducted user interviews to validate and make informed design decisions.
Hi-Fi Prototype
Refined and evaluated through prototyping.
Based on the findings from the wireframes, I created a high-fidelity clickable prototype to conduct a detailed task analysis to examine the success rate and satisfaction rate of using the new dashboard. I conducted 20+ usability tests and iterated 5 rounds to finalize the MVP design before handing it over to the development team.
User satisfaction increased by 139% with the overall task success rate of 94%.
Validation
The user satisfaction rate rose significantly from 1.8/5 to 4.3/5 with the new design - a 139% improvement. The overall task success rate was 94%, averaging over 5 areas (onboarding, navigation, finding data (drill down, filtering), retrieving previous search, modifying overview page).
Additionally, the heuristic evaluation score improved by 480% - from a severe 24/40 to a much lower 5/40, signifying fewer usability issues. These scores reflect a marked enhancement in user satisfaction and overall experience.
Easy to drill down
"Love having this level of data drill down, laid out in a user-friendly way."
- Vehicle Engineer
Clear, intuitive, and customizable
"I am super happy with how easily customizable it is. The key info can be prioritized and are effectively presented in a limited space."
- Quality Analyst
The dashboard was launched globally, achieving +525% user growth in 8 months.
The Outcome
The revamped dashboard has been successfully launched worldwide, catering to over 5,000 Ford users across the globe.
This represents a substantial growth of 525% in just 8 months, up from the previous 800 users. During this same period, the usage rate has skyrocketed by 1,500%, with daily report generation rising from 500 to 8,000.
The improved user experience of the dashboard enabled Ford to make informed, data-driven decisions that fueled customer satisfaction and fostered innovation in the automotive industry.
I was recognized for this achievement by Ford’s leadership.
Project Wrap-Up
Reflection
Serving multiple personas
One of the biggest challenges was serving multiple personas because this dashboard was used by a multitude of roles (30+ user groups). I identified the primary user groups based on daily usage rates and conducted interviews to understand where their needs overlap.
Communicating data effectively
I learned to convey a clear story to users by leveraging suggestive visuals, labeling, progressive disclosure techniques, and animation. Consistency in the use of colors, fonts, and formatting was also key to enhancing the visual appeal and readability of the dashboard.
Minimizing user efforts
By understanding users’ goals, I was able to provide the most meaningful information at the forefront for each user with an opportunity to drill down into more details. Through rounds of user feedback and design critique, I was able to iterate on the design to be more simple and powerful.