OVERVIEW
How might we empower young DIYers to carry out home repair and improvement projects with confidence?
My Contribution
Concept development, UI/UX design, Motion design
Timeline
Aug ‘23 - Oct ’23 | 6 Weeks
Tools
RESEARCH
Interviews and contextual enquiry for gathering insights first hand
We visited our local Home Depot and interview the employees as well as customers. We asked the customers about their in-store and online shopping experience, struggles with regard to home improvement projects and aspirations. We asked the employees what do customers most frequently need help and their ideas about how the store could be improved. Apart from direct questioning we also made observations of their behavior to capture what might be missed in conversations.
Trouble diagnosing
If faced with a home repair or maintenance challenge, users struggle in diagnosing the problem and as a result do not know how to solve it.
INSIGHT 1
Uninformed about right products and quantities
For any project, most users don’t have the expertise to know the right tools and materials they need for the task as well the right quantity needed for their specific needs.
INSIGHT 2
Overwhelmed at the store
Our target users reported often feeling overwhelmed at the store, resulting in always needing to ask for help.
INSIGHT 3
PROTOTYPING
Storyboarding as a form of brainstorming and building empathy
All the team members created storyboards for potential solutions. Everyone presented their ideas and told their stories which led to a productive brainstorming and ideation session.
Reimagining the complete experience by designing user journeys
All the team members created storyboards for potential solutions. Everyone presented their ideas and told their stories which led to a productive brainstorming and ideation session.
Our solution?
Codi, voice enabled smart virtual assistant that helps users by diagnosing, purchasing, navigating and providing step by step instructions for completing a home improvement project from start to end.
Why VUI/CUI?
Our research showed that users often needed assistance while they were in middle of a home improvement projects when their hands were occupied. In addition, our target user group is very accustomed to interacting with assistants like Siri or Alexa, ensuring interactions with a new CUI/VUI wouldn’t pose a learning curve.
Meet Codi
AI Image recognition for diagnosis
To diagnose a problem, Codi simply asks the user to take a picture of the affected area. Then using AI powered smart image recognition, Codi precisely detects the problem.
Personalized and context aware
After diagnosing Codi suggests multiple ways to solve the problems based on factors like the user’s skill level, budget, time required and other contextual cues.
Adjustments based on user inputs
Codi recommends the best option for the user but also makes necessary adjustments to them as requested.
Location activated dynamic store-mode
When Codi detects user is near any Home Depot location, store-mode is automatically suggested which unlocks a host of ‘in-store features.’
In store navigation and assistance
When inside the store, Codi’s store mode helps user navigate through the gigantic shelves and get exactly what they are looking for.
Human in-loop when necessary
Codi recognizes that certain tasks are best done by humans! The right store assistant is requested for the user based on specific needs.
3D scanning for task assessment
Codi utilizes 3D AR scanner to recognize the target area for tasks like painting to suggest precise products, quantities and time required for the project.
In store navigation and assistance
Codi gives careful instructions for each step of the process using HD's existing resource library and provides foresight into what to expect.
Dynamic conversations
Users can ask questions to Codi regarding the task in progress to which Codi's smart AI feature can provide answers.
Reflection
VUIs can be incredibly helpful at guiding users through tasks that require hands on work, like home repair and improvement projects. Home Depot already owns a large repository of assets like DIY tutorials and instruction manuals. These can be made more accessible and contextual using smart virtual assistant. Lastly, a smart eco system of customer facing as well as worker facing virtual assistant could increase efficiency and satisfaction for users.