Prefab home customization
Scaling sales by helping home buyers explore products independently
Project overview
Summary
Blu Homes, a high-end prefab home builder, introduced a novel 3D home customization tool that allowed customers to explore product lines, configuration options, and pricing, while also visualizing both the interior and exterior spaces.
Problem and context
Blu Homes manufactures homes in a factory and delivers them to customer sites, offering customizable designs similar to configuring a car. Instead of choosing between two- or four-door models, customers select options like a three- or four-bedroom layout, or whether to upgrade the kitchen. When I joined the company, sales relied on traditional PDFs to present around 50 categories of customization options—such as cabinet layouts and countertops—but these lacked real-time pricing, spatial or visual feedback. This created a significant burden on sales, finance, and design teams, who had to manually explain options, create estimates, and produce personalized renderings. To address this, we developed a suite of 3D tools that allowed customers to visualize their homes, explore configuration options, and receive real-time pricing. At the time, nothing like this existed in the industry.
Here are some photos of completed Blu Homes:
Promo video explaining 3D home customizer value proposition
My role
As the UX lead for Blu Homes' digital tools, I played a key role in the company’s rapid growth from 5 to 300 employees. Over five years, I transitioned from UX Designer to UX Manager to Director of UX. I led a team of eight internal employees, including UX designers, architectural designers, 3D artists, game developers, and software engineers, and managed several external cross-functional teams. My responsibilities ranged from designing initial UI mockups and storyboards to developing and optimizing 3D virtual models, as well as building a custom content management system from the ground up. As a product evangelist, I conducted customer training sessions to ensure users could maximize the value of our tools. I also spearheaded the patent process as the first inventor, working closely with our legal team to write a patent related to the 3D home customizer’s user interface.
Impact
Post-launch feedback was overwhelmingly positive, particularly from our sales team, who appreciated the ability to transfer the specification selection process to customers. Home buyers also enjoyed the flexibility to customize their homes at their own pace, free from the pressure of rushed decisions. The launch of these tools garnered significant media attention, with coverage from prestigious outlets including the New York Times, Wired, Bloomberg, Venture Beat, Gizmodo, PC World, Builder, Entrepreneur, Jetson Green, Grist and Thrillist. My contributions to the patent application led to a successful patent publication, and this, along with the product launches, significantly enhanced our intellectual property portfolio and was instrumental in helping the startup secure $60 million in funding. By the time I left Blu Homes, three years after the tool’s launch, customers had saved over 30,000 home configurations.
— KEEP SCROLLING IF YOU WANT TO GET INTO THE WEEDS —
Process and approach
User testing 3D navigation with prospective customer at a conference
1. Discovery and research
Stakeholder Alignment: I collaborated closely with the CEO, VP of Technology, VP of Marketing, and our sales leads to align on project goals, gather insights and feedback, define product requirements, prioritize features, and ensure cohesive messaging across all departments.
User research and competitive analysis: To gain insight into the challenges of the current system, I observed salespeople interacting with customers during lengthy meetings in our office conference room. These sessions involved examining floor plans and images of homes while laying out tile, countertop, and flooring samples across the conference table, revealing significant scalability issues. I conducted a competitive analysis of online design tools used by other home builders, which were primarily 2D but offered some configuration functionalities we could leverage. Additionally, I spent considerable time reviewing car configuration tools that featured much more sophisticated functionality, for example swapping of 3D components and texture mappings. I also tested early versions of the user interface with salespeople to evaluate various 3D navigation methods, as well as gathering product feedback from prospective customers at trade shows.
The inside of our home building factory in Vallejo, CA. The 3D home customization tool had to account for real-world challenges, such as fluctuating prices and product options we offered going out of stock.
2. Strategy and planning
I hired a team of internal and external specialists, created a timeline of detailed milestones and wrote detailed requirement documents to ensure the following needs were met.
Usability: A lot of the homebuyers were older and often not tech-savvy so it needed to be exceptionally intuitive and easy to use. The tool also had to be lightweight enough to run smoothly on regular consumer desktops, especially since home customization could be a time consuming process. We needed functionality for users to save their configurations and share them with a salesperson for further discussion. Additionally, in-app guidance was essential to help users navigate the tool and interact with the 3D environment. Lastly, we wanted it to be simple to select groups of materials (e.g. flooring, countertop, cabinet color, backsplash tile) that look good together and don’t clash aesthetically.
Detailed 3D models: It had to accurately reflect our product offerings, configuration options, and real-time pricing. At the time, we only had 2D floorplans, so we had to bring in 3D artists to create detailed 3D models, embedding all possible configurations—such as different bedroom layouts, kitchen setups, and fixture choices—within a nested tree structure. Each option had to be positioned correctly in the model, with interchangeable textures, like flooring or tile, seamlessly applied to 3D objects.
Flexible and granular control for content management: To handle the complexity of options and pricing, we needed a flexible content management system that allowed us to quickly update materials and pricing based on availability and market changes. We also had to implement complex conditional logic (AND, OR, NOT operators) to manage dependencies between options. For example, a wine fridge was only available with the highest-tier kitchen package, and countertop costs fluctuated based on several interrelated choices.
Delivery speed and budget: Given the company's tight financial runway, speed and financial efficiently were critical. We had to deliver a solution on a shoestring budget in less than 1 year that would quickly boost sales and demonstrate to investors our ability to launch effective digital tools.
3. Collaboration and application design
Collecting feedback on 3D home customizer prototype from company leadership, salespeople and designers
Collaboration: I spearheaded the early stages of the 3D customization tool by designing initial UI mocks, which I used to solicit bids from software development firms. I later managed an external UX design firm to improve the 2D user interface, and partnered with a 3D game developer and a web designer in France to build the application. Internally, I led UX designers in developing graphical tutorials for the UI and later transitioned the application development in-house, where I managed a software engineer and a team of 3D artists and architectural designers, who were responsible for generating the 3D content. Additionally, I collaborated closely with the estimating team to ensure accurate pricing for the many configuration permutations, and worked with an external PR firm to craft the tool’s public launch strategy. Regular feedback sessions with salespeople and customers were also essential for refining the tool’s usability and overall effectiveness.
Learning what didn’t work: The 3D Home Customizer went through a two major iterations built on different tech stacks before we landed on the one we launched with. The first was a 2D only tool and the second was a 3D tool built with a game engine called Virtools. Here are some UI samples:
Prototype #1: 2D Customizer
The earliest version of the Configurator was in 2D. To make changes to the exterior of the home, the user selected elevation line drawings and could click on a mirror toggle that flipped the home. We quickly realized that it would be necessary to use 3D if we wanted customers to understand the space.
Prototype #2: Virtools 3D Customizer (option categories)
The 2nd version of the tool required customers to install a browser extension that was buggy, cumbersome and not cross-platform compatible (it didn’t work well on MacOS), which was a showstopper. Even though we didn’t end up moving forward with this prototype, we still learned a lot from it. For example we discovered that having customers select finishes one at a time led to poor aesthetic outcomes, as choices for flooring, countertops, cabinets, and backsplash tiles often clashed. This insight inspired us to introduce a new feature in the next iteration, allowing customers to choose coordinated material packages that ensured a cohesive look.
Prototype #2: Virtools 3D Customizer (navigation)
In the earliest navigation system, users could move around a blue circle on the floor and click it to walk to that location. Reorienting the camera was accomplished by clicking on the edges of the 3D window. Although this system is sometimes used in PC games, user testing taught us that it was not suitable for our customer base, which was unfamiliar with this UI. People did not know the purpose of the blue circle, and even when they did, they had a difficult time positioning it accurately. The map in the upper left showed one’s position from bird’s-eye view within the house but was eventually removed because it cluttered the UI and was not especially helpful.
Figuring out what worked well: Eventually we rebuilt the 3D home customizer in the Unity game engine, which was more stable and cross-platform compatible. We also improved the 3D navigation, render quality, 3D mesh size and introduced material palettes.
3D navigation tutorial
The navigation system we ended up using had three modes: exterior orbital, interior walk-through and 3D birds-eye pan/zoom. We conducted user testing with customers, internal employees, friends and family to make the system intuitive. Several rotation and walking speeds were evaluated so customers could move around efficiently without feeling nauseous. Users can explore the homes by clicking and dragging on the environment, tapping on screen buttons or pressing keyboard arrows. We also enabled USB Xbox gamepads. Customers could even change their viewing height to personalize the home exploration experience.
Making 3D models more lightweight
Because the customizer is often used by customers with slower computers, we had to keep mesh and texture resolution low to minimize the amount draw calls and time spent downloading assets. The left images show a chair before I reduced the polygon count, the right images show after. Note there is not a noticeable impact to visual quality.
Simulating kitchen palette selection
A useful feature the Blu Homes Customizer gave users is the ability to select and customize palettes. In an earlier build, the Configurator had over 50 categories in which to choose specifications. This gave customers a high level of control but it was time consuming, designs often came out unattractive and it often overwhelmed users. Palettes solved this by grouping together several categories of specifications in a given room, such as cabinets, flooring, countertop, backsplash and wall paint color. Blu's product designers make sure materials in the different palette categories look cohesive together. Customers could make a single selection instead of five. If they still wanted to tweak an individual category however, the configurator still allowed for it.
4. Launch
The Blu Homes Customizer launched to much fanfare with this press release. Here is a demo of the UI we went public with:
This video shows the UI that we launched with in action. The user selects a home model, customizes it and saves their selections to be accessed later. The 3D visualization and pricing update throughout the process.
After the launch I ran several online and in-person workshops to teach promote the tool and teach customers how to use it.
Prospective home buyers walking around the exterior of a virtual Blu Home in VR
Learnings and subsequent developments
Leadership growth: I learned from this project how to effectively grow and manage diverse teams. It taught me the key qualities to look for when hiring in the design field, such as creativity, adaptability, technical proficiency, and a strong collaborative mindset—essential traits for UX practitioners—and how to evaluate them. I also learned best practices for working with remote teams, including establishing clear communication channels, setting measurable goals, and fostering regular touchpoints to maintain alignment and engagement across different time zones.
Complementary tools: After the successful launch of the 3D home customizer, we expanded the toolset with several complementary products. These included a customizer admin console, VR home tours, an AR site visualization mobile app, and a construction progress tracker, each designed to further enhance the home buying experience and streamline project management.
Let’s connect
Want to discuss customization tools or how this project relates to yours? Reach out!