The Data Leader’s Guide to Crucial Conversations

If you’re like me, you’d rather avoid conflict. Maybe it started back in third grade when we figured out it was easier to be agreeable than ask for what we actually wanted.

The problem? Avoiding conflict and uncomfortable conversations—especially in leadership—doesn’t make problems disappear. In fact, it often makes them worse.

For data leaders, this is especially true. Whether it’s pushing back against a VP’s unrealistic deadline or addressing an underperforming team member, difficult conversations are part of the job. And the more we dodge them, the more we stall our own growth.

In this article, I want to explore conflict avoidance in the data world—why it happens, why it’s a problem, and, more importantly, how to handle crucial conversations effectively.

Let’s start with a common challenge: data experts and their tendency to avoid conflict.

Data Teams and the Conflict Avoidance Problem

If we ran a survey, I’d bet most data professionals would lean toward being conflict-avoidant. It’s a pattern I see all the time.

Just the other day, I was talking to another data leader about this exact issue. They shared a story about having to step in and have a difficult conversation on behalf of their ICs because a stakeholder had started pushing boundaries—bordering on bullying.

I chuckled and said, “Let me guess. Your ICs told you it was no big deal, and they’d just do what the stakeholder asked, right?”

They nodded.

If we plotted conflict avoidance on a spectrum, with obnoxious jerks on one end and conflict-avoidant pushovers on the other, I bet most data folks would skew toward avoidance. But plenty of us are overly dogmatic—going on the offensive the moment we see a data axiom being broken.

Neither approach is great.

The real skill lies in striking a balance: being direct without being combative, standing firm without being inflexible.

So, let’s talk about how to do that—starting with the art of having crucial conversions.

Why You Need To Be Able To Have Difficult Conversations

“A person's success in life can usually be measured by the number of uncomfortable conversations he or she is willing to have.” ―Tim Ferriss

I love this quote from Tim Ferriss because it hits on an uncomfortable truth: avoiding difficult conversations won’t save you from problems—it will only make them worse.

Sure, we avoid conflict because we worry about damaging relationships or creating a quagmire of a situation. (Guess what: you might.) But if you never learn to push back, you’ll find yourself at the mercy of others’ decisions—often at your own expense.

• Is a client micromanaging you, but you’re afraid to push back because you don’t want to risk the project?

• Is a VP pressuring you to cut corners, even though you know it’ll cause bigger delays later?

• Is a team member underperforming and slowing the whole project down, but no one wants to address it?

Many data professionals would just let things slide. The problem is that these issues rarely fix themselves—they usually escalate.

Imagine this: You notice a project is going off the rails. Someone on your team who’s slightly senior to you is responsible. You hesitate to speak up because you don’t want to overstep. The project derails completely, deadlines are missed, and leadership is frustrated.

All because you avoided a difficult but necessary conversation.

That’s the conflict snowball—ignoring small issues until they spiral into bigger ones.

This problem exists in every industry, but its impact varies. In tech and data, missed deadlines and bad decisions cost time and money. In industries like healthcare, avoiding crucial conversations can literally cost lives.

Before diving into how to have these tough conversations effectively, let's talk about what not to do.

Don’t Just Say No—And Don’t Be the Boulder

One of the most common difficult conversations data professionals face—hundreds of times in their careers—is when they need to say no or redirect a decision.

Take this classic example:

“We should get rid of data engineers and let analysts and SMEs have direct access—it’ll be faster.”

Sure, there’s a shred of truth to that. Modern tools make data more accessible than ever. But realistically? Most of that work would still end up in Excel, which is a whole different topic.

Going back to the original premise, as a data leader, your first instinct might be to say, “No, that’s a bad idea,” and you’d be right. But if the person proposing this—especially a senior leader—has already convinced themselves or others it’s a great idea, a flat-out no won’t cut it.

In fact, you’ll be viewed as an obstacle.

You can’t just be the boulder that blocks progress.

Instead, you need to provide a path around the boulder.

This isn’t about caving to bad ideas. It’s about guiding the conversation so the other person can recognize the pitfalls themselves—without feeling like you’re shutting them down.

Because here’s the thing: even at the highest levels, leaders have plenty of stressors. Sometimes, we forget that the people above us also have people above them. What seems like a major issue to you might not even register for them in the bigger picture.

What does register? The feeling that someone is undermining them. That’s why trust is key.

Taha Hussain

article on surviving toxic workplaces covers this in the context of building trust. It’s a solid read worth checking out.

Some Tactical Mindsets And Frameworks

There are plenty of methods for handling difficult conversations, but one I often hear people say to avoid is the sandwich method—where you cushion criticism between two positives.

Why? Because once you're in a leadership role, people see right through it. It comes off as formulaic, and worse, it dilutes the clarity of your message. So, just cut it out.

Instead, I really enjoyed the framework Celina Wong shared during one of our recent chats on the TFA. It’s simple, direct, and effective:

1. Find Common Ground: Start by aligning on shared priorities.
   "We both know that accurate and reliable data reporting is critical to our decision-making and overall business strategy."

2. Call Out the Problem: Clearly state the issue and why it matters.
   "That said, I've seen some discrepancies in our recent reports that could undermine their credibility. The errors, though small, are creating confusion and could lead to misguided decisions."

3. Provide a Solution: Offer a constructive path forward. "I recommend we implement a peer review process before finalizing any reports. This extra step should help us catch errors and ensure our data is robust and trustworthy."

Radical Candor

Maybe it’s just required reading for Facebook, but it’s hard to talk about difficult conversations without referencing Radical Candor. After all, if you look at the chart above where I have conflict-avoidant vs. aggressive jerks, the only thing missing is the Care Personally axis.

But unlike a strict framework, Radical Candor is more of a mindset test.

Think about your last few difficult interactions. Where did you fall?

• Did you care deeply but hesitate to state your opinion or challenge directly because you didn’t want to upset anyone?

• Did you wrap your point in so much fluff that your real message got lost?

Here’s an example of what that looks like in practice:

Senior Data Leader:
"I went over the updated set of goals for 2025 Q2. The direction is promising, though there’s a minor disconnect between the proposed milestones and our current resource levels. I believe our team can probably figure it out."

VP:
"Can you be more specific about the disconnect?"

Senior Data Leader:
"It’s just that the staffing levels might not perfectly match the timeline. I don’t see it as an immediate issue, but it’s worth keeping in mind."

This response is vague, hedged, and ultimately unhelpful. A better approach? Just say what you mean:

“I don’t believe our current staff level will be able to hit these milestones because of X, Y, and Z. To stay on track, we should either hire a contractor or extend the timeline by two months—either approach will help us meet the goal.”

That’s being direct while still showing you care about the outcome.

Final Thoughts

Difficult conversations aren’t always predictable. Sometimes, they blindside you, and other times, you’re the one who needs to initiate them. Either way, learning how to navigate them is a skill that will serve you well—both as a data leader and in life.

It’s a fine balance between being direct and being a jerk. It takes practice, but the more you lean into uncomfortable conversations, the more you can refine your approach. And the more you refine your approach, the more you realize that things tend to work out on the other side.

So don’t shy away. Start practicing. The more you engage in crucial conversations, the better you’ll get at them.

As always, thanks for reading.

---

VAST at NVIDIA GTC - Don’t Miss Out!

Experience the Future of AI Data Infrastructure

Join VAST Data at NVIDIA GTC 2025 and discover the data platform powering today’s AI breakthroughs. Visit our booth #733 to see the future of AI infrastructure in action, attend our exclusive events, and hear from industry leaders in our speaking sessions. Whether you’re building next-gen AI models or optimizing your data pipeline, we’ve got something for you.

🚀 Explore the agenda and book a meeting here.

Let’s shape the future of AI together! See you at GTC.

Get 20% off on your pass, on us.

---

Join My Data Engineering And Data Science Discord

If you’re looking to talk more about data engineering, data science, breaking into your first job, and finding other like minded data specialists. Then you should join the Seattle Data Guy discord!

We are now over 8000 members!

---

Join My Technical Consultants Community

If you’re a data consultant or considering becoming one then you should join the Technical Freelancer Community! We have over 1500 members!

You’ll find plenty of free resources you can access to expedite your journey as a technical consultant as well as be able to talk to other consultants about questions you may have!

---

Articles Worth Reading

There are thousands of new articles posted daily all over the web! I have spent a lot of time sifting through some of these articles as well as TechCrunch and companies tech blog and wanted to share some of my favorites!

---

Introducing Netflix’s Key-Value Data Abstraction Layer

At Netflix our ability to deliver seamless, high-quality, streaming experiences to millions of users hinges on robust, global backend infrastructure. Central to this infrastructure is our use of multiple online distributed databases such as Apache Cassandra, a NoSQL database known for its high availability and scalability. Cassandra serves as the backbone for a diverse array of use cases within Netflix, ranging from user sign-ups and storing viewing histories to supporting real-time analytics and live streaming.

Read More Here

---

End Of Day 168

Thanks for checking out our community. We put out 4-5 Newsletters a month discussing data, tech, and start-ups.