At Slack, I led the design of a new audio/video collaboration feature called Slack Calls. This was my main project for two years, which I guided from its inception to a general availability release.
I joined Slack in December 2014 as a part of the Screenhero acquisition. Screenhero was a collaborative screen-sharing tool for remote engineering teams. I was a founding designer at Screenhero, and we brought our experience and tech into Slack.
As a lead designer on Slack Calls, my role involved everything from product strategy and product design, product writing, all the way to sound design.
We designed and built iteratively, prototyping the path forward, starting with audio call foundation, moving to video, and lastly, collaborative features such as screen-sharing. I worked closely with engineers, validating our assumptions at each step through real-world usage.
Even though Slack’s core value is in text communication, talking is sometimes faster than typing. Therefore Slack’s customers needed a way to have reliable face-to-face conversations from their immediate context and optionally share their screen – collaborate. Impromptu conversations needed to happen instantly and be shareable with other people.
In anticipation of a long development process, constantly altered with new discoveries and observations and changing technical limitations, we created a set of high-level design principles, guiding our decisions in the future.
Through shadowing a wide variety of meetings, I identified many types of calls based on the context in which they are happening – 1:1s, daily standups, large video conferences, small group calls, and scheduled meetings. While they share common needs, there are also tangible differences.
Based on the research findings and competitive analysis, we focused on small teams without enterprise-grade conference rooms that can benefit the most from having a simple, readily available default experience directly in Slack.
Calls can launch from any conversation (1:1, group, or channel) anywhere in Slack in a single click, taking advantage of the tight integration.
All calls are represented as living objects in Slack’s conversations so they can be quickly joined, shared, and persist in the message history. This was also a foundation for later features like transcription, attachments, recording features, and keeping a call record with all its related information.
Slack Calls had to work both in a browser and as part of a standalone app. However, we could enhance the standalone app experience with custom incoming call notifications (on the system level) and a mini call window, always present when the main call window isn’t in focus.
Reactions during a call started as a playful and novel idea from Andy Locascio – an engineer on the team. After a quick prototype, we quickly saw positive engagement and decided to push it to a general release. In the years following, reactions (emojis) became ubiquitous in video conferencing software.
We focused on making the mobile experience as close to a familiar native calling functionality as possible, taking advantage of actionable notifications, ringing, and the incoming call screen.
In the first four days of releasing voice calls, ~102.5K calls were made. The number of daily active users (DAU) of Slack at the time was ~750K. After releasing video calls, the usage went up by 56%.
Video calls were the #1 customer feature request, so releasing the feature significantly enhanced Slack’s selling proposition to potential customers and increased perceived value among existing teams.
Although Slack Calls provided the basics of face-to-face conversation, it was just a start. Meetings of any form (virtual or in person) require significant investment in time and attention. In the years following my departure, Slack introduced features such as Huddles (simplified impromptu audio chats directly in a channel) and Clips (quick recordings) tailored for the emerging hybrid or remote future of work.