Building an AR room-scanning app
Client
CITY Furniture
Role
Digital Product Designer
Year
2024
Summary
You're looking for a new piece of furniture after purchasing a new home, or you're simply looking to replace an old sofa. You make your purchase online, receive it, but it just doesn't fit the way you thought it would.
This was one of the many issues our customers faced when ordering from our website. This case study follows the journey of how the design team tackled this issue and designed a product that was developed but, ultimately never made it to market. A multi-year, cross-functional effort that involved collaboration with stakeholders, engineers, and our customers.
01 Understanding the problem
Background
City Furniture, a leading Florida-based retailer, has seen a rise in digital sales after the onset of the pandemic shutdowns. Accompanied by increased customer complaints about fit, sizing, and styling in online returns.
A thorough and deep understanding of why returns were increasing was needed.
Historical return rate
≈ 6%
2020-2022 Return rate
≈ 14%
Initial steps
We started off consulting with the UX research team on customer grievances that were documented as part of the bi-weekly user tests that were ran to gauge consumer sentiment post-purchase.
Time and time again customers stated that sizing and material choice were a real miss having shopped exclusively online and not in our brick and mortar stores. Our sizing descriptions and swatch pictures were not enough to predict how the piece of furniture would fit in their homes.
Ultimately, the UX team recommended a solution in the form of helping customers better visualize products in their homes.
This ultimately resulted in the development of two digital products. We will be focusing on the mobile app as part of this case study.
02 Research
We kicked things off by diving into the e-commerce furniture world, checking out what competitors were doing and spotting trends to better understand what customers were looking for. Turns out, a bunch of companies—both in and outside our niche—were already tackling similar challenges. Some had cool 3D room planners, while others offered simple tools to customize room layouts.
It was really inspiring to see how these solutions weren’t just functional but also visually stunning and easy to use. But it also hit us that the bar was set pretty high. We knew our customers wanted to see how our furniture would look in their spaces, and we needed to make that happen fast.
Mapping solutions
So, we set our sights on creating a scanning tool and a room-building product that would let users generate 3D versions of our furniture into their own scanned rooms.
The scanning tool was to accurately map out how context in which our customers could see furniture. We needed to have a very detailed and exact representation of their spaces. The challenge? Making it not just technically solid but also enjoyable and intuitive to use. This digital product is what we will be focusing on as part of this case study.
Here are a couple of examples that inspired us, especially their guided instructions to help users avoid mistakes.
03 Design Process
Ideation and Brainstorming
After exploring solutions implemented by other companies we knew we had to address any technical concerns in order to create a solution that our dev team could deliver on.
In talks with our engineers and product team we hit our first 2 roadblocks:
Roadblock 1
The scanning process required multiple scans to create reliable mockups, making it more tedious for users than we expected.
Roadblock 2
We relied on Apple’s LIDAR for room scanning, limiting the app to iOS users and excluding Android entirely.
We started by brainstorming how to make the scanning process simple and intuitive. While we initially aimed for a single-scan approach, the limitations expressed by our engineers led us to split it into two scans: one for the room layout and another for finer details like wall texture, color, and objects.
As for limiting our exposure to Apple users, we decided to focus our efforts on iOS for the first release and address the Android issue in future releases.
Wireframes
Using very rough wireframes, we mapped out an ideal state solution for the entire customer flow having kept a two scan approach.
Knowing how technical this process could be for some customers we added instructional steps before each scan to avoid any mistakes or hiccups our users might encounter. The last thing we wanted was to have a user repeat any of the scanning tasks.
We also included saving checkpoints after every scan where our users could review and check that everything looked as it should.
User Testing and Feedback
Testing revealed that while the two-scan process made sense technically, it felt overwhelming for users. Our focus here going forward was to really make this process as easy as possible. Users loved the idea of instructional videos but wanted clearer prompts during the scans themselves more than what we had in the mockup. Based on this feedback we noted to add a step by step how-to during the entire scan process to ease any user uncertainty and possibly implement a progress bar for timing expectations.
Refinement and Hi-Fi Mockups
After collecting all our feedback we met with our product team to cross check the project requirements without adding any scope creep. We were all on board with the design direction and started producing her fidelity mockups.
This is where we started addressing the granular aspects of the mobile app. UI and visual language was heavily borrowed from our main CITY Furniture brand but adjusted for iOS.
At this point we were on track to deliver a great experience.
Scanning screens
As the design team was concluding hi-fi mockups the engineering team notified us that some of the custom UI we created for the scanning screens wouldn't be possible to implement without considerable dev lifts. The most that could be done was adding instructional popups through the process but that was it. Any other UI would have to be set aside and we would assume the default scanning programs UI.
Although the flow was unaffected the visual consistency of the app might seem off during the flow since then native UI was not 1:1 with our brand. We agreed to have this addressed post launch.
04 Final Designs and Handoff
Collaborating w/ Creative
With the bulk of the design work completed at this point it was time to start adding in all the assets needed for the release of the app.
We worked hand in hand with the creative team (photography, video, communication design) in order to accurately show users exactly what was needed for them to use the app without any hiccups. This included some environmental shots for the introduction page, a series of instructional videos for the scanning process, and a series of still images of how to hold the phone.
This is really when everything started coming together.
Dev Hand-off
Throughout the process we really pushed for an agile approach so by this time in the process the development team was 100% aware and building the framework for the work that had been completed. In addition to this we had a few rounds of meetings at the tail end of our design process to communicate in granular detail how some of the interactions and flows were supposed to function.
05 Project Retirement
As much as we loved our app, the tech world moves fast—and competitors moved even faster. While we were busy perfecting our two-scan process and making the experience as smooth as possible, other companies rolled out features like real-time 3D rendering and AI-powered room customization. Suddenly, our app felt a bit behind the times.
After some tough but thoughtful discussions, the business decided to retire the project and focus resources on other exciting opportunities. It wasn’t an easy call, but in the end it was the right one.
06 Reflections
While it’s always hard to say goodbye to a project you’ve poured your heart into, this experience taught the team so much. It reminded us how important it is to stay nimble, and always be ready to push innovation to next horizon. Even though the app never made it to the app store, the lessons we learned—about design, collaboration, technology, and adapting to change—are still shaping how we tackle new challenges today.