A Curious Case of UX Research: Re-designing the Process of Material Selection for Interior Designers
The interior design studio where I worked until March this year is a small firm comprising 13 employees. During the design development stage of each project, 2–3 interior designers and a design assistant (me) would spend at least 15 min. to locate the right material for a client’s designs by sifting through several physical bins, buckets and shelves in the library. This exercise happened at least twice a week. Per month, this easily amounted to more than 6 hours of unskilled human labor by highly skilled professionals. Each service/task was billed by the hour, to the client. Given the time and cost-sensitive nature of each project, clearly material selection aspect of the design development process was turning out to be inordinately expensive for the client! Moreover, climbing over the ladder and crouching on the floor to shuffle around heavy bins of tiles, wood, fabrics, hardware and rug samples was physically exerting as well as time cumbersome. Such an acrobatic routine begged the question- Should it have to be so difficult to find desirable materials from the in-house library? So, I asked myself the typical HMV question:
“How might we reduce the time spent on searching for desired materials by reducing the number of steps involved in making decisions during the process of material selection?”
For the purpose of this article, let us understand an interior designer as a user of materials. Often, the designers and I were unable to find exactly the kind of fabric, tile, paint, leather and wood samples for design projects, without spending at least 15 minutes rummaging through nearly 4–5 bins full of sample swatches placed in nearly 2–3 different locations. Many times, we labored to locate the closest match, only to discover after reaching out to the vendor that the pattern is discontinued. This exercise was eating up too much of our time. Day in and day out, I thought to myself, clearly the process of material sourcing, selection and maintenance could be simplified and streamlined.
No doubt, the limited space to store all materials within the studio was a constraint. And while plans to relocate to a bigger office space have been under consideration, they are not a priority for this year. However, that was not the only problem. There was a general ambiguity about which material attributes to prioritize during selection. I’d like to refer to it as a ‘classification conflict’. Supposing a designer was looking for moss-colored, distressed leather for a custom upholstered bar stool, one designer would prefer checking for the varieties that both a retail vendor as well as direct vendor has to offer by sifting through their respective bins. Whereas, another designer would prefer sorting through color/pattern and then budget before determining the type of vendor. Some materials, such as hardware product samples, could not be classified based on vendor type as they were sourced only from direct vendors. To add to the complexity, one kind of material was stored in different locations, wherever some room could be made for more samples. Another classification criterion was use-based. If say, the leather type had previously been used for a different client’s project, its use was strongly discouraged by the principal designer for a current or an upcoming project. Thus, the existing system of material selection was not designed to address these nuances.
Most of my understanding of the problem and users’ pain points is shaped by undertaking qualitative methods. To collect data, I engaged in a service safari. The reason this worked best for me is because I was new to the firm. As I myself performed services of material identification, selection and purchase from vendors, I started noticing glaring inadequacies in the organization of objects and information. While issues pertaining to spatial architecture were tackled by interior designers, information architecture was in dire need of attention. When navigating through every step of the service I noticed frustrating interactions between co-workers about how the material library needed to be reconfigured. This allowed me to just swoop in at times and intercept a conversation or a user’s experience. Additionally, before our office shut down due to the pandemic, I was also able to interview a couple of senior designers about their experience selecting materials from the in-house library. According to Stanford’s d school, there are two main prerequisites for applying a design thinking strategy to solve a problem:
1. It needs to be a meaningful problem that is somewhere in its early stages of discovery
2. And secondly, it needs to involve a human element so that one can gather insights from the people whose problem one is trying to solve.
Following this principle, I reframed my challenge as thus: Redesign the material storage and classification system for interior designers to expedite and ease the process of material selection.
By undertaking a design thinking approach to the problem outlined, I thus charted out a course to discover, define, hypothesize and ideate a possible solution to replace a flaccid material inventory & classification system with a foolproof, robust digital infrastructure.
Existing Information Architecture
At the moment, an old school method of sorting through physical objects in the library is complemented by a digital database of products and services. Albeit convoluted, there was a system for organizing information. One of the popular platforms used by many interior designers is the Studio Webware (SWW), or as it is now referred to as Studio Designer 2.0. The version that we use in our studio is SWW 1.0. While the digital database includes visual and textual information about the items purchased for a client as well as those purchased for the styling library, there is little and incomplete information on which materials are currently held in the material sample library. Further, there is no way to track whether a product sample was actually used in a design before or had it only been considered for potential use (in which case, it is still usable for future projects). Therefore, the only way to cull fresh samples for developing a design is by physically rummaging through bins and pulling samples from the materials library. Over time, due to lack of storage space many materials had receded to the back of the shelf. Some had been stuffed awkwardly in gaps. Lighter or smaller yet heavier materials would sometimes fall to the ground and need to be shoved back into an overflowing bin or shelf space. And only a few fabrics, tiles and metal finishes were classified in distinct bins based on vendors. For visual clarity I have illustrated the existing process of material selection via a user flow diagram below. As an example, I have chosen ‘fabrics’ as the material type to be retrieved from the library.
I realized that a unified system of archiving all the materials digitally would help bridge the information gap and resolve the classification conflict. In the existing system, occasionally, some material samples were found under the option, ‘Z Samples’. This folder could be accessed from the ‘Items’ page in SWW 1.0, after selecting the client’s name and room type from the drop-down bar. In some instances, the material description included links to ‘Order’ form and ‘Invoice,’ indicating that the materials were first ordered by the designers on behalf of the client and then charged to the client’s account after their approval. In this manner, the existing system involved an elaborate set of steps — first physically shuffling through sample swatches in the library — and then checking inconclusive digital records suggesting prior use of a certain sample. The reason this system was proving to be so problematic and time cumbersome is that:
a) The samples that showed up under ‘Z samples’ were only few of the samples that had over time, either been considered or used in a specific client project,
b) The SWW digital database did not contain any record of the total sample collection held in our design studio’s library;
c) In the absence of any location codes, there was no way to link images/information of samples in SWW with the actual physical location of the sample in the library.
d) Lastly, the information regarding the price/unit of any material sample was mutable depending on the vagaries of the market, collection updates and price regulation by a certain vendor.
The existing system of material archiving had no scope to keep up with changes to pricing, style, or pattern in the collection. The provision for doing so however does exist in Studio Webware. It was only a matter of taking notice of certain relevant features and investing time to adapt them to internal design needs by training employees to use a single database most effectively. In the existing system, data was harbored in multiple sites- the physical library, Studio Webware, the Excel spreadsheets documenting standard operating procedures, the Google Drive and desktop computers.
How then could such a multivariate system of material classification be integrated? To streamline each set of operations, the most essential series of sub-tasks included-
· weeding out all discontinued patterns from the sample library
· determining location codes for every row and column of each shelf containing sample bins
· and finally, compiling a directory of all categories of the library with Studio Webware, such that it may be linked with ongoing client projects, past projects as well as vendors.
A Potential Solution
I began by drafting a revised user flow for an improvised system of material selection. Visually illustrating a revised flow of tasks enabled me to pin down the start and endpoints of the entire process while listing out the necessary steps involved in making decisions at specific intervals. The goal ideally is to maximize the breadth of the search for new materials with minimum effort, in a minimum duration of time.
Simultaneously, I embarked on the task of flushing out discontinued patterns by material type while retaining only those that are still in continuous production. In order to reorganize the complete material inventory, one would have to sort through each bin, bucket and box of samples of wood, rugs, fabrics, leather, tiles, paint and metal finishes, shades, counters top stones, hardware, trims, etc. It was no doubt, a labor-intensive and time cumbersome project. Vendor names are almost always listed on the back of samples. So, I started scouring through websites of vendors to verify whether they still carried the styles and patterns which we held in our library. At times, a few vendors would oblige by replacing old samples when coming over for presenting their latest collections. Occasionally, I would stumble across samples that were being considered by designers for ongoing projects. In such cases, I would immediately note down their price per unit, pattern name/number, the client name and date of update, so that in future these details could be promptly inputted in the online database on SWW. Since I joined the design team in January, I was exposed to fabrics, leathers, tiles, wood and rugs more than any other material. Hence, to begin with, I decided to focus on segregating fabrics, leather and tile samples by listing them out on a commonly shared document in our company Drive folder. Sometime earlier during ideation, I had met with our project manager who advised me to improve the specificity of location labeling and taxonomic classification to serve the immediate needs of the designers.
Materials being considered for use in each client project are stored in bins, with each client bin having a dedicated shelf space. One day, a senior designer asked me to unload the contents of an old client project bin by putting away all unused samples in their respective locations in the library. The standard procedure entailed clicking photographs of each sample and uploading them to distinct client folders in the company’s Drive. In fulfilling this task, I realized that meta tags of ‘used’ vs. ‘unused’ could thus be created to differentiate materials. Those that had passed the client’s approval and had actually been used in the final interior designs should be separated from those that were considered as options but not used in the design. It made me think of what could possibly be the best way to have information on all these materials listed in the database such that there is access to all the details about each sample including client project, room type, price/unit, color, composition, pattern name and availability. I was determined to give it a go. So, I opened Studio Webware 1.0 with the objective of creating a prototype digital catalog of samples from the library. My hope was to be able to present it as a mock demo to the entire team after its completion. I noticed that the ‘Products’ section had not been used and was still an unexplored space. I began toggling around. Then, I started building a comprehensive list of samples as illustrated below. In the sidebar on the left-hand side, you will notice that there is a provision to filter products by material types such as leather, fabric, furniture, floor covering, etc.
I realized that another way to link information on the ‘Products’ page is via a ‘sales code’ option, which is visible upon clicking on each product. Having a sales code is useful to categorize items based on type such as furnishing, finishes, hardware, etc. which serves as a useful indicator while listing the item for resale to the client. This feature’s dropdown menu was already pre-populated as it was being used to index purchased goods on the ‘Items’ page. However, what I was curious to know is whether there was an interconnecting pathway between the ‘Products’ page and the ‘Items’ page. Therefore, I added another option of ‘sample’ to the drop-down list of sales codes. Then, I made sure that the vendor and client were listed correctly. After that, I closed the ‘Products’ page and opened the ‘Items’ page which is where our regular business was conducted. Upon selecting the client, ‘Wong’, the vendor, ‘Room & Board’ and the room type as ‘Z Samples’, the sample image and information that I had recorded under the ‘Products’ page instantly showed up, with the sales code- ‘sample’. This way a link could be established between the Items page and the ‘Products’ page. Now all those photos of product samples from finished client projects that were stored in Google Drive, isolated from all the other relevant information on the ‘Items’ page on SWW, could be integrated into a single SWW database under the ‘Products’ page with a link to ‘Items’. This would then make it so much easier to distinguish the used samples from unused ones!
This infrastructure comprising a coordinated labyrinth of data used for storing visual and textual information pertaining to products and people involved in the functioning of a business is nothing but a design system. Regardless of how friendly and intuitive the interface of such infrastructures seems, I’ve come to observe that their adoption, adaptation and complete integration with an interior/architecture design business takes time and sustained upgrades. Clearly, the design system at this interior design studio that I worked with, was not being optimized. Data was being stored in 4 different places. There was a lot of scope to expand the digital database while substantially minimizing physical and multi-sited storage of materials as well as information. The in-house library could be made spacious enough simply through a dedicated and collective effort to digitize all information in one place. Granted that the core function of this business is designing interiors, but as the business matures, popularity rises and client projects begin to increase every year, it’s probably a good idea to prioritize a robust well-integrated digital infrastructure to keep things from falling apart. Furthermore, a foolproof and robust digital infrastructure inspires confidence in clients.
Of course, this case study is in no way conclusive. The last (and incomplete) leg of this journey would be to get feedback on the improvised user flow from my fellow employees and build a prototype digital inventory. Unfortunately, this part of the process was interrupted by pandemic closures and lay-offs. Nevertheless, the purpose of this essay has been to broach a conversation on a cross-disciplinary application of user research in a field less researched. Given our move towards digitization post-COVID-19 in every industry, this kind of user research may be adapted to any workplace to solve problems regarding storage/cataloging or distribution of physical products that may be indispensable to the core business. Alongside design development, database creation and management may be carved out as a role in itself. The way I see it, transporting a chunk of this material selection process on-line and maintaining the digital inventory in itself presents UX Design and UX Research opportunities in interior design firms. One fully functional, an all-inclusive digital database can store information regarding all samples, whereas only the materials with relatively newer product variations/patterns may be stored in the physical library. I strongly believe that a unified design system that is uniquely tailored to individual design firms can increase productivity, streamline operations, inspire client confidence and expedite collaborative decision making.