It's been about a year since I started my first full-time role out of college, working as a data scientist at Squarespace. Recently, I decided to change things up and join a fast-growing startup called Hugo to work on all things Growth.
This detour away from further specialization in data science surprised a lot of people. It still surprises me a little bit. I learned a ton as a first-time data scientist from both a technical and a functional perspective.
This post is my attempt at preserving a few of those ideas, while hopefully providing you with some valuable takeaways from my first role as a data scientist.
Invest in Your Team
When I played baseball growing up, I was always focused on doing what I could to help us win, rather than putting up numbers for myself. Sometimes, this was even to my detriment.
I never really carried that thinking over to my career until this past year. At internships and early on at Squarespace, I was laser-focused on proving myself and establishing credibility. It wasn't until I had settled into my role and felt vetted as a data scientist that I was able to shift my thinking to where I wanted the team to go and how I could help us get there.
Working at a startup, I've noticed this mindset carries over nicely. At the end of the day, organizations aren't that different from sports teams. You can still focus on personal improvement while working to bring the team closer to some goal. Often, they turn out to be one and the same.
Scoping out Projects Is Hard
Surprisingly, planning out projects that would drive impact was the most difficult thing about being a data scientist. I started out taking a top-down approach. I would pay attention to the questions that stakeholders were asking, and then try to come up with ideas for potential answers. This was somewhat effective, but I often ended up getting stuck trying to untangle the messiness of a vague question without immediate actionability.
As I progressed, my process became more and more bottoms-up. I would look at what I could do today that would drive smaller but more relevant impact. I still think that I naturally lean towards top-down thinking, but I now see the benefits of approaching problems bottoms-up in organizations. This might differ depending your particular situation. Try both approaches and see what moves the needle.
When in Doubt, Prototype It
There's a classic line, "don't ask for permission, ask for forgiveness." This is a bit extreme in most cases, but there is some truth here. If you have an idea that you think is really important, it's not always best for you to take the time to try and get buy-in from stakeholders. Just show them.
Find the least amount of viable work that you could do to prove the project out and start there. This could be a proof of concept or something as simple as a 1-pager project brief. Just the act of getting your thoughts laid out clearly in a short document goes a long way.
There's also an added bonus here that once you do a little work and the project has some promise, others will be much less inclined to push back on next steps. I've found that this is the quickest way to drive impact on larger teams.
Having Opinions Is Good
It's difficult to avoid group-think while making decisions in organizations. Things get complex awful quickly, especially when you take into account variables like company politics and history. The reality is that you can't control everything, but you can be helpful by doing one simple thing: Have an opinion.
This doesn't mean to be the loudest person in the room or give unsolicited "hot takes" even when you don't know what is going on. Instead, this means that when you're informed about something, have a strong point of view and share it. Most discussions end up with everyone sitting somewhere in the soft "maybe" range. Anecdotes get thrown out and information is shared, but the team doesn't get any closer to making a decision.
Even if others don't agree with your perspective, you're still helping move the conversation closer to a decision. It may be a little uncomfortable, but others will appreciate the contribution. Faster, more comprehensive decisions are good for everyone.
Make Sure Somebody Owns It
Lastly, this one isn't rocket science, but turns out to be pretty important. Assign an owner to literally everything. Everyone has a million balls in the air. Everyone is understaffed. Everyone has different incentives and projects that matter to them. If you make the assumption that someone else will pick a project up, you're making a mistake.
The key is to be explicit. The workplace gods punish the vague wish and reward the specific ask. If you want someone to do something, ask them. If you want to automate that ask, assign an owner. Keep these two things in mind and your life will be much easier.
I have no complaints about working as a data scientist. In fact, I loved it. Data science is a rapidly growing field with high-leverage that has the added benefit of being really fucking cool. It was incredibly fun to go into work everyday where I could perform in-depth analysis, build cool things, and inform product decisions.
With this being said, many of the counterpoints about working in data are true. When working as an analytics data scientist for a larger team, you are rarely the primary decision maker and things are often out of your hands. You will be in the room for lots of decisions, but in the best case, you serve the role as the right-hand man or woman.
Ultimately, I decided to go to the other side of the spectrum by joining a startup where I couldn’t avoid being a primary decision maker if I tried. As someone who likes to move fast, is a self-proclaimed generalist, and wants to be a founder someday, this feels right.
I'm still a data scientist at heart, but I have a lot of other interests that I want to explore as well. Besides, I'll always be a data scientist at heart. I won't leave you quite yet, Jupyter Notebooks.
Thanks for reading! If you enjoyed this post and you’re feeling generous, perhaps follow me on Twitter. You can also subscribe in the form below to get future posts like this one straight to your inbox. 🔥