How Docker navigates product analytics for an open source community
“Raise your hand if you use Docker,” asked Marshall Reaves Louis, Mixpanel Machine Learning Engineer and notable host of Mixpanel Office Hours. A room full of developers raised their hands.
For those who don’t know (or don’t code), Docker is the open source platform for developers to build, ship and run distributed applications.
And millions of developers use Docker across the globe. So we were excited to have Ben Bonnefoy (a.k.a. “French Ben”) join us for our February Office Hours so he could share Docker’s philosophy behind product analytics, and how Mixpanel helps them build better products.
(If you weren’t there, you didn’t miss the boat entirely. We’re hosting it again in an exclusive virtual event on February 28th – you can join here.)
Ben, a software engineer at Docker, shared how the world’s leading software containerization platform navigates product analytics and communicates the benefits of tracking to users and team members.
But there’s a Catch-22. In an open source community, sometimes the idea of tracking user behavior can come with its sensitivities, both within the company and for its users. In having a single source of truth, Docker better understands its platform’s usage and how to make it even better for its thriving community. As the saying goes, a rising tide lifts all boats.
Here’s a glimpse into how Docker uses Mixpanel to understand its users. Once product analytics was instrumented into several of Docker’s products, like Docker for Mac and Docker for Windows, Ben and his team had to decide which metrics where the most important ones to measure, and why they were so crucial to the development of their products:
During the Q&A, one audience member asked a question that everyone was thinking: In an open source community, does tracking irritate your users?
Not necessarily, said Ben. According to the software engineer, honesty is the best policy. Clearly communicate to your users on how your behavior analytics will better the product and help a company understand its usage.
And especially for open source solutions, providing users with the ability to “opt-out” allows them to choose whether or not they want to contribute their engagement metrics to the betterment of the platform.
But you might be surprised to learn that contributions to the community come from the most unlikeliest places. Ben shared how in Beijing, China – even with its notorious firewall – users (through a VPN, of course) will still opt-in to sending their metrics to the containerization company.
According to Ben, product metrics give you visibility into how well your product is doing. It’s also a source for invaluable insight when uncovering new ways to iterate and optimize products. Plus, data is the necessary component to making important decisions with your team.