For this project, our team were asked to design a mobile service that makes use of emerging technologies for Construction Junction (CJ), a local non-profit that promotes conservation through the reuse of building materials. We designed Trove, a service that incorporated augmented reality and a conversational user interface, to help CJ shoppers find unique items.
Lu Yang, Tiffany Jiang, Nishchala Singhal
Mobile Design; Emerging Technologies; E-commerce
I conducted brainstorming session and field research with the team, and was responsible for designing the user flow, high-fidelity UI Screens, prototyping and creating micro-interactions within the app.
Homeowners travel far away to visit CJ’s warehouse hoping to purchase unique items from a trove of donated building materials. However, they are not guaranteed to find items shown on CJ’s website as the information is not updated in real time. Even if they do find the item they want, they are not sure it will be a perfect match for their homes.
INTRODUCING: CJ TROVE
Chatbot Personal Assistance
Trove’s chatbot provides a quick and easy way to get personalized item suggestion.
Purchase from any location
Customers can save their favorite items to their shopping list and shop from anywhere.
No Guessing Required
AR Home Preview makes it easy to find a perfect in-home match.
Through online research of the Construction Junction site, we identified key stakeholders, which include CJ employees, donors, customers and partners. We created a stakeholder map, which accounts for all the individuals and business associated with CJ. The map identifies opportunities in exploring CJ’s relationship to its customers as we assume it drives the most revenue.
Using reverse assumptions as a method to brainstorm ideas, we listed out assumptions on the customer’s pain points and the opposing statement, and then imagined how a new mobile service could address these pain points. This helped us identify a problem we would like to focus on:
How might we help CJ customers navigate the warehouse to find items they need?
We framed 20 questions to generate as many ideas as we could.
For example, we asked:
1. Can CJ obtain customers’ shopping lists and optimize the route to pick up items?
2. Can CJ provide an inventory expert to guide in-store shopping experience?
3. Can CJ match their inventory items to pictures customers bring in?
4. Can CJ suggest items to customers as they walk through the store?
Framing 20 questions
Writing down reverse assumptions
Our ideas included:
1. Computer vision would allow customers to upload a picture and find similar items in the warehouse.
2. AR technology could help customers navigate to the items in the warehouse.
3. Predictive analytics could help suggest items to customers as they peruse the warehouse.
4. A chatbot would help customers find their unique item.
EXPLORING THE PROBLEM SPACE
From the above exercise, we arrived at this assumption: CJ customers have trouble finding items that suit their needs because both the inventory and warehouse are gigantic.
Our proposed service would provide them with a chatbot that use computer vision to closely match in-store items to the inspiration images they provide. The customers could then choose the item they like and use AR optimized route that appears through their phone to locate the item.
We presented wireframes in class and received the following feedback:
Some customers may actually like perusing the aisles in the warehouse and may not want an optimized navigational route.
Tracking where all the items are in the store for the navigation feature seems infeasible. Would we need to place GPS tags on each item?
It could get tiring for customers to navigate through the whole store while holding their phones up to use the AR feature.
CJ wants to see an increase in sales. Would this app accomplish that?
REFRAMING THE PROBLEM FROM FIELD RESEARCH
From the class critique, we quickly realized additional research would help us better understand how we could improve the shopping experience. We took a trip to CJ and conducted guerilla interviews with four customers (homeowners) and three employees. The site visit helps us pinpoint specific problems in our next iteration.
1. Staff were very helpful in finding a specific item.
Acting as customers, we walked around the warehouse to see if we could find the crates that matched an inspirational image we found online. Unable to find it by ourselves, we sought for help from a front desk employee. Within a minute, he came back from the warehouse inventory with the item we wanted. We were surprised that he could match the item so well and so quickly!
An inspirational image
A crate the staff found for us
2. Customers tend to shop with specific requirements in mind
People cared to find items that matched the dimensions they needed. They also prioritized the look and uniqueness, but they often forgot to bring photos of their home for reference, so there was no guarantee that what they buy would actually match their home.
3. The online inventory is not comprehensive and slow to update.
We learned that each item over $5 has a barcode and unique SKU number from CJ’s dock coordinator. The barcode and SKU number are for internal management. Staff members have to update the online inventory manually.
4. Customers travel far to CJ for the unique items and the great deals.
A couple travelled 1.5 hours to get here and look for tiles that complement the peachy tiles in her shower. Cost was not a huge issue because prices are already significantly cheaper than what is offered by competitors.
DESIGNING A PERSONALIZED SHOPPING EXPERIENCE
We learned from the site visit that home owners are CJ’s primary customers, and they travel from far away to look for unique items for home renovations. They tend to come here with specific requirements in mind and in-store navigation is not a big issue for them. The customers are unsure if they could find the item they need in store, and they worry if the item will actually match well with other furnitures in their homes. Based on the insights, we wondered:
How might we help CJ customers find items they need with ease and purchase them without wondering how well the items will look like in their homes?
We designed Trove, a mobile personalized shopping experience that enables CJ customers to purchase unique items that meet their requirements with the option to visualize how the item would look in their homes.
A CUSTOMER’S JOURNEY WITH TROVE
ITERATIONS ON LISTS
We considered how customers would view their shopping lists. After a customer clicks the “Save to List” button, we surface the lists they have made in the past, and they can add the item to one of their curated lists. The initial design directs the customer to another page in order to save items in the list. We decided to make the list a pop-up window to minimize distraction and help customers focus on their key task.
ITERATIONS ON AR PREVIEW
The original AR feature put the frame on the door. Customers drag and resize the object to fit the frame. After looking at conventions set by apps such as IKEA Place and Houzz, we chose to position the frame on the door. We made it seem like the customer is placing the door rather than fitting the door to the frame. Originally, customers didn’t have the option to put items they like directly into their shopping cart with AR preview. We added the cart function on the assumption that it will increase conversion rate.