|Duration||Mar 2019, 5-day design challenge|
|Key Words||Conversational Interfaces
As a person who hasn't decided to settle down in one place, Charlie always signs yearly leasing contracts for apartments. He just moved into Rosemont, a community in the coveted Atlanta suburb. Apart from the great scene with a reasonable price, this community attracts him because the management team serves people super well. In the past, when he encountered amenity issues, he has experienced various kinds of “toing and froing” situations that exhausted both the leasing office and himself.
Now, he doesn't need to suffer from the inefficiencies any more! Rosemont management team has adopted a new product ———— LifeMate for every residents in the community.
On the day Charlie moves in, the leasing agent sent a LifeMate invitation link to his registered email. The link redirects him to joining the Rosemont community on Lifemate, and he only needs to confirm the email address and apartment number.
With the chatbot, he can report issues and request maintenance without any effort. The AR mode is especially helpful when his parents come to visit him from China.
Talk to Google assistant to report the issues. He doesn't even need to open the app! Also, it's multi-lingual now!
With LifeMate, Charlie can track the progress of each request and take actions correspondingly.
The process of maintenance requesting vary significantly from community to community. I started to narrow down the scope by analyzing keywords in the prompt and listing its variations. Taking reachability into consideration, I decided to focus mainly on young leasees in leasing communities. Based on my previous knowledge and some assumptions, their characteristics are:
To quickly understand users' thoughts & deeds on existing solutions, I talked to 6 leasees with drafted interview questions. I tried to balance the age, gender, nationality and experience in living in leased apartments. Also, I approached 2 leasing agents for stakeholder interviews. I coded the interview notes and synthesized findings via affinity mapping.
Understanding how people get amenities repaired right now gave me a better idea on their painpoints.
To my surprise, the biggest difficulty users encounter is not how they suffer from inefficient feedback (Well, I made this assumption at the beginning), but how they give up in the first place due to the inconvenience to initiate a request.
Something you may need to know before I move on to explainig my design process:
From previous research activities, I noticed how "lazy" people could be and how they regard "direct conversations" as a way to quickly get attention.
With such questions in mind, I started brainstorming solutions. A smart digital agent became my core design direction. My first version of design was an AR-based chatbot that could help users to identify the issues. For users who're not on spot and cannot use the AR function, I proposed to use voice assistant like Google assistant to initiate the conversation. Then I returned to the users I interviewd beforehand for feedback
Good news! You only have to answer at most 4 questions (sometimes 3) in a few words, or maybe don't have to think about a single word but just "tap tap tap"!
After refining the user flow of the system, more importantly, designing the conversational flows, I then created a relatively comprehensive wireframe for user testing. I intentionally chose two different types of issue as examples in the prototype, so users can understand the major functionalities easily during the test.
How can users get access to the reporting system conveniently? Inspired by project management service like Asana and Slack, LifeMate is a platform for any leasing real-estate corporate to create their "online communities". Residents will get an email invitation to onboard the system. No registration work is needed because your profile has already been established by the management team!
How to avoid redundant manual work when reporting problems?Users can simply talk to the agent bot to report issues. If they don't know how to address the problem in detail properly, the system will provide answers!
Thanks to the development of machine learning, AR & language analysis in voice interactions, I was able to utilize these trending technology and designed the conversational interfaces for apartment management in the future.
Although in my project scope, reporting public issues is not a necessary and frequent practice. I still consider the edge cases:
As the flow B-5 suggests, the system will automatically detect if the reported case is a public issue, and will ask if the resident would make this incident visible to neighbors or the whole community. By doing this, all community members involved in the issue will receive a notification and follow up with the maintenance progress.
I wasn't able to conduct a thorough usability testing due to the time constraints, especially for the conversational interface. However, I did asked users to quickly test on my final design. The overall feedback was quite positive.
There were two small iterations I did after conducting the evaluation session:
I still remember when I approached the first user, he didn't think there was any big deal in issue reporting situations. "Why do you choose to solve such a boring problem? I don't think there's any better way than filling out an online form or directly making phone calls." He asked. To be honest, I felt discouraged at the moment, doubting myself if I could actually make something creative and useful in the end. But I didn't give up. After I showed him my final crafts, he was convinced that a product like LifeMate would definitely make his life easier.
For designers, there's never a "boring" issue. It's important to dive deeply into the context and dig out insights. I really enjoyed doing this design challenge because:
Of course, due to the time and resource constraints, my final deliverables are still far away from "final" if it is a real product. Future steps: