Aug 6, 2015 by Eric Muntz

What We Learned on Our Engineering Retreat


Last month, we took 7 members of our engineering team to an off-the-beaten-path, sustainable community south of Atlanta called Serenbe. We called the week an Engineering Research Retreat, and its purpose was to completely disconnect from the day-to-day responsibilities of building and maintaining MailChimp’s products to spend some time researching.

A few of us who had previously attempted using vacations or conferences to do this kind of deep research failed miserably. Meanwhile, we were feeling pretty bummed by our lack of focused research time in 2013. So last year, Chad and I came up with the idea of forcing the issue by going somewhere, bringing along some engineers, and locking ourselves away to research. He emailed our CEO Ben to see if we could get budget approval, and got this response: “You had me at ‘Hey Ben.’ Budget approved.”


Scaling down and setting expectations

We tried to avoid making rules this year. During our 2014 retreat, we said, “Don’t work on MailChimp problems.” But that was too restricting for a) some folks who are really into MailChimp problems or b) so new to development that that’s kind of all they know. So this year, I changed it to: “Try not to work on MailChimp problems.”

We also encouraged rather epic projects last time, telling attendees that if they finished their work during the retreat week, it wasn’t big enough. We totally scrapped that approach this year, suggesting instead that the work they do should be meaty enough to engage them for a week, but not necessarily something that’s going to change the world.

Last year, our entire 20-person engineering team went on the trip. That's a lot of people! We had a great time, but everyone grouped up in their usual cliques and didn’t socialize or explore topics with other people as much. This year’s plan was to have 3 retreats with about 10 people each, in order to keep the trips a little more manageable.

Our July group ended up with only 7 attendees, which initially felt too small. But as it turns out, 7 is the perfect number! We stuck together as a team the entire week—no cliques, no separate groups, just 7 united engineers working together to solve problems and have fun.


Who went and what we researched

Our group was comprised of 3 women and 4 men, ages 24-40. We had an email developer, an infrastructure engineer, a junior engineer, our engineering services manager, our lead API developer, our staff engineer, and our VP of product.

We researched all sorts of stuff, including learning Photoshop, hacking on hardware, learning Go, Python, and Elixir, and building a ray tracer in C. Staying true to nerd form, we sat in a room together and communicated through a chatroom about our GitHub commits. We presented together to start and end the trip, and it was so much fun seeing what everyone accomplished and how they felt about it.

Listening hard and changing fast

We stayed at Serenbe for 5 days. While it was amazing, I think we’d shorten it to 3 or 4 next time. That’s a long time to be away from family, and maybe too long to research a greenfield project. By day 4, I was actually longing for some production bugs to crack, or maybe even a few meetings to attend. (Oh, hello, I’m the VP in the group.)

One other fun thing we did was present awards to each other at the end. Our email developer used puns and inside jokes we’d made throughout the week to determine the awards—and then we gave her an award for coming up with the best award names. It was a fun way to wrap up the week. I can’t wait to see what the next group comes up with. Oh, and if you'd like to be a part of that group, know that we're hiring.