Dear Danielle,
Our teams are constantly being asked to do more. Do more, and faster. Do more, and better. Do more, with less.
This is especially true now in the1 age of AI. The expectation for knowledge workers is that with the help of AI, we can and should do more, and do it faster, better, and with fewer resources.
And so today, I want to share a success story: the research team at 1Password. Over the past year, we’ve taken full advantage of the AI tools at our disposal, embraced injecting AI into our workflows, and have since sped up our end-to-end research workflow by as much as 50%.
Yes, you read that right. 50%.
In concrete terms, this means we’re helping speed up the overall PDLC. We’re shortening the upfront discovery windows, getting in front of customers sooner, and bringing their feedback back faster, which gives our product, design and engineering partners more opportunities to iterate on our features.
So how did we do it? Not with one magic tool (unless one considers AI to be magic), but with an iterative system, built one step at a time.
Here’s exactly how we approached it, in five steps: audit, assess, build, measure, and re-focus. I’ll walk you through the four questions we asked before touching a single AI tool (surprise, the first one isn’t actually about AI at all!), the custom agent that took the “where’s that study?” pings off our researchers’ plates, the exec dashboard that asks “Are we building the right things?”, our before-and-after workflow maps, and why we’ve refused to speed up certain part of our process.