Day 1

Understanding the users

"It's hard to stick to a budget."

" There's a item looks good online, but I'm not sure if it will look good in my living room."

" I hope there's an interior designer knows my taste and just tell me what to buy."

" I know the look I want, but I always got overwhelmed and end up not buying anything."

" I want to stick to small items."

" I don't want to spend on only one piece"

Highlights from user interviews

After reviewing the existing user interviews and persona, I began to ask a few "how might we" questions to enhance users' shopping experience.

HMW help people to stick to a budget while they still get the look they like?
HMW help people to know the items will look well together in their own apartment?
HMW help people to get the right items to achieve the look they like?
HMW help people to get the biggest impact within their budget?

Initial end-to-end user experience

Day 2

Competitive analysis

Lighting demo

I conducted competitive research focused on how other products help users visualize decorative items in their own space. I found many interesting insights on how these products use cutting-edge technologies like AR to enable users to customize their choices.During my research, I discovered that sticking to a budget is a key concern for users. Therefore, I also investigated ways to help users keep track of their spending.Based on users' needs, I compared two established competitors' shopping apps and Pinterest. Although Pinterest is not a direct competitor, it was frequently mentioned during user interviews and its features are a great reference.

Ideation

The crazy 8s sketches & solution sketch

After synthesizing all of my research, it is now time to pick up the pencil and start the fun part! I used the solo "crazy 8" method, where I sketched eight distinct ideas in eight minutes. After completing the "crazy 8", I reviewed my ideas and chose one solution to be my final idea.

" There's a item looks good online, but I'm not sure if it will look good in my living room."

The crazy 8

" I always look through Pinterest for inspirations, but I still don't know what items I should buy to pull it off."

Day 3

Storyboarding the solution

Before I sketch out the storyboard, here are a few considerations that led me to choose my solution:
‍
・It helps users visualize the items in the room.
・It helps users control their budget.
・It requires the least amount of effort from users.
・It utilizes the most advanced technology available.

Then I sketched out around 11 frames to visualize the possible steps that user might take.

Final solution storyboard

Research on AI technology

I have decided to incorporate AI into my solution. While it is a new technology that is still in development, there are already a few AI products that I can start with. I have focused on identifying what users can do with AI and which features would be useful in my solution. I am also prioritizing visual AI tools like MidJourney, as I am trying to solve a visual problem for our users, rather than chat-based tools like ChatGPT.

After the research on AI technology, I found a couple features that are already in use that can be implemented in my solution:
・Text prompt to images
・Users can upload their own images
・Users can customize the results

Day 4

Iterations

How to use AI to help users to find the right item combination is the key for the solution. AI can help users put their ideas in their head into a practical visual solution.

First iteration

To get started, users can upload a photo of their room and describe the look they want in their own words. The AI will then generate a couple of looks for them using the existing items on the website.

"They look good online, but will they look good in my living room?"

Second iteration

Instead of requiring text input, I have replaced it with image input. Based on user interviews, I have noticed that users often use Pinterest as their inspiration resource and have a hard time describing the accurate style they are looking for. Therefore, image input is more intuitive and requires less work from the users.

" I want my apartment feel beachy, but I don't know which items to buy, something wooden maybe."

Final iteration

Persona is the guideline that I always refer back. After looking at the frustrations of the user again, I combine the first two iterations, users can upload their own room as the canvas, then use other inspirations as the paint to create their dream room.

Final design

This is part of the 5-day design sprint so I only focus on the flow of the buying process.

Upload inspirations and user's own room

Customize the room

Check out

Day 5

Validate the design

On the last day of the design sprint, I did a usability test with five users using a moderated remote user testing method due to time limitations. I sent out the Figma prototype to the users and did zoom interviews with them. I was able to get the screen recording and keep everything single move on the screen in documents. I also asked the users to think out loud and say whatever crossed their minds during the testing. This allowed me to understand what users think and their frustrations during the process.

Next step

After the testing, there are a couple areas I can improve the design:

Label individual item's price for users to compare.
The customization process can be improved. For example, users can move items around as they wish.

Reflection

I am working with a new technology in this project, so if possible, working with a machine learning engineer will give me more insights and more accurate feedback about the technology.

Another area I will do differently is usability testing. The users from the testing are not the ones from the user interviews, since all the user research was provided for me. If I can recruit users from the original interviews, I can get more accurate feedback.

House2Home E-Commerce Website

Day 1

Day 2

Day 3

Day 4

Day 5

See other case studies