The User Interface Design
During the on-boarding process, users have the option to take a virtual tour of the application to get familiar with all of its features. The AI-powered chat function is there to receive commands and answer questions.
Under the "Devices" tab, users can add devices and control them individually or as groups.
Automations are suggested to the users based on their observed and recorded habits. Automations can be made active or inactive.
Users can select and edit automations from the suggested list, or create new ones themselves. The suggested automations will have pre-configured settings that the user can edit. It will also contain an explanation of why the automation is suggested, with supporting data.
Users will be able to view their habits and data patterns in the "Insights" tab.
The Problem
Existing smart home management applications and devices provide an experience that does not feel personalized or tailored to the user. Assuming that AI technology can learn from human habits and track routines, can smart home living be fully automated and rendered bespoke?
Role
User Research, Competitor Analysis, Design, Prototyping & Usability Evaluation.
Team leader: Liza Karimova
Supporting members: Gaelle Noel, Saurabh Khatri, Harshal Nanavare
The Solution
Home Pal is an AI-powered mobile application that learns from habit patterns and automates smart home control, uniquely tailoring it to the users preferences. It offers suggestions backed by digestible data that can help track and modify habits.
Constraints
12 week timeline
IOS platform
Mobile-first design
User Research
-
A survey was created using Qualtrics
-
29 participants that have experience with smart home devices were recruited
-
Participants were asked questions about what features they would like to see added to existing smart home mobile applications
-
7 participants were interviewed to gain further insights on what they would like to see improved in the way they manage their smart home devices
Interview Analysis - Affinity Diagram & Quotes
“I would like a section in the app for suggestions based on my usage.”
“One thing I do enjoy is seeing my own habits in a data quantified manner, like with screen time.”
"Ideally, everything should be automated. If you can read my mind, I don't have to do anything."
“I want to know how I can reduce my energy use while still having the same experience”
"Any relationship has to be developed. Like, even if it was a human Butler, right. I wouldn't immediately trust this person to do anything for me."
"You can have my data, as long as my life is easier"
User Research Take-aways
79%
Participants want cost savings suggestions & sustainable living
69%
Participants want personalized suggestions & privacy not a concern
Pain Points:
-
Voice control not working well
-
Automations complicated to set up
-
Data statistics not presented in a digestible or exciting way
62%
Participants want a dashboard with statistics of their usage
<50%
Participants want a chatbot & calendar integration
Features to focus on:
-
Suggestions & Recommendations
-
Dashboard with statistics
-
Automations
Low Fidelity Prototype
sketched on paper & tested in the Marvel App.
High Fidelity Prototype
used for Usability Testing (Version 1)
Usability Evaluation
-
6 participants performed 3 tasks
-
Particpants were asked to think out loud while performing the tasks
-
The task success, time, errors, efficiency & satisfaction were measured
-
Participants then completed the SUS questionnaire.
Interview
Pain Points
& Themes
A second version of the high fidelity prototype was developed to respond to the findings from the usability evaluation. The re-design focused on making the navigation and automation set up more intuitive and providing explanations and tooltips.
Design Changes
Liza Karimova. 2023.