How to work smarter (not harder) in data

In the third episode of The View on Data, hosts Faith McKenna, Erica (Ric) Louie, and Grace Goheen dive into a familiar challenge for anyone working in data: trying to do too much. From over-engineering solutions to chasing edge cases that may never happen, they explore what it actually looks like to “work smarter” in a fast-paced, high-stakes field.

They also reflect on burnout, perfectionism, and the practical habits that have helped them create better balance, both in and out of work.

This episode is for anyone who's ever tried to brute force their way through a project, fix every edge case, or write 100+ lines of regex to solve a problem that could’ve been handled with copy/paste.

Please reach out at podcast@dbtlabs.com for questions, comments, and guest suggestions.

🎧 Listen & subscribe: Spotify | Apple Podcasts | Amazon Music | YouTube

Why data practitioners tend to over-engineer

For Ric, working harder felt like a badge of honor, until it wasn’t.

“I thought working a lot meant that I was good at my job… I would spend my evenings going through all these different solutions rather than just tapping on the board and being like, ‘What are we actually trying to solve here?’”

Grace shared a similar story from her days as a practitioner, when she wrote a massive regex script for the Codegen package.

“It’s so gross and stupid… I probably would’ve been faster just copying and pasting SQL. But I wanted to make it reusable, I wanted it to be perfect and it totally didn’t need to be.”

Perfectionism is a powerful force in data work, especially when the job is to make sense of messy systems. But that instinct to fix everything can often get in the way of making progress.

How to define “good enough” in data work

One of the biggest lessons the group shared is the importance of scoping tightly and building toward a minimum viable product (MVP).

“If you find yourself tunneling, step back and ask: what are we really trying to do here?” Ric said. “What’s the smallest version of this that actually works?”

Having a shared definition of “done” helps, too. Faith talked about the pull request templates and model documentation standards she used as a practitioner:

“We had three questions every model had to answer: who built it, who asked for it, and are we planning to upgrade it in the future? It gave us a shared sense of when something was good enough to ship.”

When automation is helpful, and when it isn’t

The team also shared stories about how they’ve learned to automate wisely, focusing on repetitive tasks and known processes.

“Automate the boring stuff,” said Ric. “But if it’s a high-risk workflow or something ambiguous, you probably still need a human in the loop.”

Grace brought up her work on CI features that compare changes between dev and production environments:

“It’s one of those things that felt so repetitive when I was a practitioner—doing manual checks on every PR. Automating that saved a ton of time.”

Faith noted that she’s started using voice-to-AI chat to track progress across her different responsibilities:

“It’s helped lessen the mental load, because I can say, ‘Hey, remember when I told you about those four projects? I finished two. What’s left?’ And it remembers.”

The key theme? Use automation to reduce friction, not add complexity.

Managing context switching in data jobs

The hosts all agreed: constant context switching is part of the job, but it doesn’t have to be painful.

Ric described how she sets up her workweek using a stream deck and structured Notion workspace:

“Every Monday, I press a button, and all the docs I need pop up. It’s like a daily routine I can follow, instead of having to figure it out every day.”

Grace relies on time blocking to stay focused:

“I try to do more time-boxed deep work. Like, all of Wednesday is for one thing. And when I switch, it’s a conscious choice, not just reacting to every Slack ping.”

Faith shared how she uses a daily work journal to reflect and prioritize:

“At the end of each day, I write down three things: what I did, what I’m proud of, and what I need to do tomorrow. It helps me remember where I left off and what matters most.”

Creating space for life outside of data

While working smarter matters inside the office, the team also talked about what it means to fully log off.

Grace finds balance through acting:

“I do theater, which means I have rehearsal at 6 p.m. And that forces me to stop working because I have to physically go somewhere else.”

Faith echoed the need for built-in rituals:

“My husband and I go to a weekly yoga class. I also do improv at a theater near my house, which is great because it activates a totally different part of my brain.”

For Ric, the lesson came after hitting burnout:

“At the end of 2023, I took a three-month leave. I realized that just because I love my job doesn’t mean it should take over my life.”

Now, she holds herself to a simple rule when deciding whether to keep working late:

“Does this need to be done right now? Does it need to be done by me? What would happen if I waited until tomorrow?”

Episode takeaways

This episode is a practical and personal reminder that doing good work in data isn't about solving every problem. It's about solving the right problems, at the right time, in a way that's sustainable.

Here are a few takeaways from the conversation:

• Over-engineering often comes from perfectionism, not necessity
• MVP thinking creates clarity and momentum
• Smart automation frees up focus, but should be intentional
• Time-blocking and rituals can ease context switching
• Work-life balance requires clear boundaries and regular check-ins