1 product manager, 5 designer
UX/UI Design
User Research
November 2022 - January 2023
After PM initiated the idea of redesign of the remote office platform experience, we decide to focus on the redesign of Zoom. We aim to improve the meeting user experience of Zoom in emojis, interaction, whiteboard functions, etc., to make the meeting interaction more immersive. This redesigned zoom encourages communication and collaboration anytime to help users reduce the loneliness caused by remote work. Also, it helps to create a more efficient work environment and makes collaboration more convenient.
During my internship, we shipped four rounds of the iteration and successfully drove up the users' satisfaction with the Zoom platform experience through several rounds of user testing. Throughout the design process, I designed and participated in multiple user research sessions, helped narrow down the feature prioritization and impacts, made several versions of the design, mainly focusing on emoji iterations within its visual and functional aspects, and held regular meetings with PM.
Since 2019 the pandemic has started worldwide, and remote work has been a trend. Although it benefits people in some aspects, such as saving commute time, it harms the relationship between workers and company culture—apps quickly respond to this trend to meet the demands of remote workstyle on the market. Our challenge is finding out the users' pain points that Zoom hasn't solved and entered the highly competitive on-demand remote workstyle worldwide. Zoom built up on its core function- online meeting. Centred on this function, we made some improvements to better meet users' demands. With new functions and improvements, we hope to strengthen user relationships and create a healthy community culture.
We found that 95% of workers feel bored at remote work, 77% of workers would prefer an easy and fun way to communicate during work, and 68% of workers think engagement at work is important to them, 59% of them needs proper tools to increase the work productivity and willingness to share valuable knowledge with co-workers.
Remote interviews were carried out with twenty interviewees between the 14th and 18th of November 2022. Interviews were conducted with Zoom frequent users. The key findings from the interviews are:
.png)
.png)
.png)
Based on the research findings, we created the two personas and a Journey Map to visualize what we learned from the interviews:
We proposed a Zoom redesign and add three new functions focusing on enhancing interactive communication and increasing meeting productivity.
We categorized ideas together along impacts and effort lines to help in making decisions and balance/evaluate suggested actions before committing to them. Based on our researches on workers’ satisfaction on remote work experience, We decide to focus on:
How might we encourage workers to share knowledge and tips?
How might we make it interesting for Gen Z workers for online employee training and virtual team meetings?
How might we create a easy and fun way to share knowledge with the co-workers?
How might we create a friendly competition environment for workers when they do online meetings and presentations?
How might we…………
.png)
We concluded that several parts of the Zoom platform were unsatisfactory to the users. To continue, we reframed the problems that we need to address with our design solutions.
.png)
Using emojis to response more accurate and quickly.
Through researches, we understand that users not always use the emojis through their official meaning,thus, instead of using the official descriptions, we decide to characterize the meaning from the context where emojis are used.
Introduce to LINE model
LINE provides two settings, which preserve first-order and second-order proximity respectively. The first-order proximity between two nodes is the local pairwise proximity, which is directly measured by the weight of the edge between the two nodes. In the co-occurrence graph, the first-order proximity measures how likely the two tokens co-occur. The second-order proximity between two nodes, however, is the similarity between their neighborhood network structures. That is, how likely the two tokens occur in similar contexts. In this work, we trained two LINE models that preserve the first and second order proximity, and obtain two sets of embeddings. We will refer to the two models as LINE-1st and LINE-2nd respectively.
Geometrical visualization
The LINE embedding enable us to find the tokens that are semantically similar to another token, as the semantic similarity can be directly computed as the euclidean distance in the embedding space. Tokens geometrically close to each other in the embedding space are also semantically similar. Therefore, the semantics of a token can be represented by its nearest neighbors in the embedding space.
Using emojis to response more accurate and quickly.
Using emojis to response more accurate and quickly
Customization is a way to express personal identity, and create users shared identity. It provide means to say “this is us” beyond the content of conversations, which can strongly contribute to a sense of belonging in online spaces.
In the context of personal communication, customization can help create “mini communication places”within an app reflecting the intimacy of a relationship in the interface itself:
“I like changing the default emoji [in Facebook Messenger], it’s like a sign of closeness and intimacy. Like, here are my actual friends, people I talk to on a regular basis. It creates a differentiation from the default blue, the default [Facebook] 'like' thing”
.png)
.png)
.png)
.png)
.png)
1) The design of option 2 is more helpful to find drawing line icon
2) The users prefer pen feature instead of drawing line feature
3) 3/4 of the users can’t find laser icon and 1/2 users seldom use it.
.png)
.png)
.png)
Provide screenshot function and enable easy saving and sharing.
.png)
Next Steps
1. Find more users to test our products; if anyone drops off in the middle or cannot find some functions, we will interview them to understand their needs and resons.
2. Work with PM to improve the wiki function.
Takeways
1. Be humble! I learned quite a lot from PM and other designers. Since I am mainly responsible for the Emojis part, the initial interface is not as good as the final version. I have iterated many times with PM. During the weekly team discussion, other designers also gave me valuable ideas.
2. I am good at researching and finding related resources; my academic years of studies contribute to this part which always allows me to have more comprehensive perspectives and seek insights behind phenomena.
