Sway is a video communication tool to replace meetings with more in-person like communication for remote teams. And by today Sway is now open source, available for all developers on Github.
Watch this video to see the Sway in action:
What's Sway?
The idea behind Sway is basically to reduce coordination cost; which slows down remote teams by average 20%. I saw how coordination delays build up and slow down work in the teams I lead since Covid; coordination is delayed by default in remote work and every delay has a cost. A quick catch up about a PR review gets delayed to tomorrow, or a quick coordination between 3 people in the office turns into a meeting with 10 people that needs to be scheduled.
To address this pain-point, Sway places an icon in OSX' menubar. You can see your team room and quickly jump in and out of conversations fluidly;
In the same menubar popup, users can set their availability status and choose their "flow". Notice how Sway always highlights where the focus is in the UI. Its UI state kept track of focus as well;
Besides of the menubar shortcut, Sway offers a main window where you could see your entire workspace, chat with others via video or text;
And in the main window, users can enjoy a very polished command palette which allows users to switch devices just using keyboard;
Sway's command palette is not just list of commands. You can list / test / switch video / devices devices, or even choose which desktop to share right in the command palette, using only keyboard;
Reflections
Looking back, here's the learnings I gathered from building Sway:
1. Market exhaustion
Users validate the problems and like the solutions, but they do not want to add another communication product to their setup. They're already exhausted about too many tools.
2. Privacy worries
Users are worried that they may accidentally be on camera or microphone in an undesired situation. And a new product -no matter how well it's built- can't guarantee that it will never happen.
3. Think marketing first, build later
We often hear advise like: solve a problem you have and care about. I truly cared about this problem and obviously that was not enough; there should be a market with potential customers who's willing to pay for the product. Founders should focus on marketing first, building later.
What's next?
I'm building Mitte, an AI image generation & editing platform for creatives, in addition to providing other companies AI research & development through my company Ray Labs.