.png)
With the recent rise of remote work, collaborative tools have become the predominant mode of asynchronous interaction for students and professionals.Feedback is an essential part of productive process. How it is given can affect receptiveness and the team culture & dynamics. However, feedback can be time-consuming to write, sensitive, and easy-misinterpreted.This is because tone of a written feedback can be easily misinterpreted as feedback can sometimes be sensitive.
Emotions is what exactly the core issue with feedback that make it so time consuming and inefficient. Due to the sensitive nature of feedback, the reviewer often has to take a long time to craft a well written and thoughtful feedback response. Additionally, well written response can take the reader a long time to read and extract important information.
Therefore, our solution focused on removing the emotional aspect of feedback. We try to apply our ideas to redesign Figma comment feature.
Each member of our group came up several ideas to achieve objectives. After discussing, we distilled the main ideas we wanted for our system.
We created 3 sketches based on initial ideas and analyzed their corresponded competitors.
To reiterate, there are two main workflows of our system: text and voice input. We focused on refining the process so that the user can easily give feedback and the recipient can easily interpret the feedback.
Inner rationale of Summary Card
After honing in our ideas, we moved on to developing our prototype. We decided to apply our idea, SmartFeedback, to redesign the Figma comment feature. SmartFeedback enables people to leave feedback more intuitively, and read feedback more efficiently.
.jpg)
Conclusion
In conclusion-Feedback is an important part of the productive process. Yet writing it can be time-consuming or sensitive. We sought to address these issues by designing a feedback system in Figma which makes feedback more precise, actionable, and detached from personal intent. We leveraged AI and NLP to make feedback easier to author and interpret. Moreover, we allow greater flexibility by giving them both text and voice input options. All in all, we hope our system is a good stepping stone to streamlining collaborative iteration.
Future step
Looking ahead, some future improvements we were pondering, especially in light of how error prone NLP can be is error identification and recovery. We'd also like to examine how well our system generalizes beyond Figma.Other steps our team can take to refine our product is by conducting user research and testing to explore what type of feedback users find most helpful, depending on the use case. For our existing product, we'd like to test whether users find it easy to use, helpful, and whether it makes feedback authoring more efficient, understandable, and less prone to misinterpretation.