As a trained chemist, Dan DeAlmeida had spent his entire career in and around laboratories before agreeing to join LabVoice as director of product management. Although he had worked in all sorts of labs, and had a good idea of how complex and dynamic of environments labs can be, he wanted to spend a lot of time at customer sites.
Dan would often set up as many as four or five regular customer meetings per week to learn about the customer flows, to gain a clearer understanding of the day to day experience, but also to test voice workflows.
Since LabVoice is creating a voice assistant for the modern laboratory, their product either succeeds or fails based on how well it performs in labs.
“I’ve got 12 inches of bench space and I need to mix my reagents where do I put my computer? Or I have to hold on to a rat while I’m taking some measurement, how do I write down my values? It’s that visual learning perspective that really helps open people’s eyes and say, oh I get it now…”
Dan has been using his time on-site working with prospects to help define and iterate voice dialogues; what the system needs to say, and what the users typically say in response.
How Dan Tests How Well Dialogues Match Customer Flows
To learn, Dan creates happy path dialogues for various lab scenarios. He then hands them off to scientists, who then act them out a bit like theater.
Although dialogues are usually 60% okay the first time around, this process helps them identify where users get stuck and there’s a need for exception paths, provide feedback on the flows and the language, but also helps scientists build familiarity with the technology.
Dan says: “To them, it’s just another software technology, but it’s really not. It changes the way they can interact in the lab and the value they can get from it.”
Through the process of providing feedback on the dialogues, scientists start to understand the general concepts, and the type of value voice can bring to their work.
These on-site collaborations with basic dialogue prototypes help Dan and his team significantly improve the product, iterate, and improve their product-market fit.
You can learn more about how Dan and his team are learning from customer flows below:
This post in an excerpt from Solving Product. If you enjoyed the content, you'll love the new book. You can download the first 3 chapters here →.