Data Leader & 4x Author: Why Stakeholders Ignore Your Work
- Gilbert Eijkelenboom
- Apr 14
- 7 min read
MindSpeaking Podcast Episode 28 - Bill Schmarzo
🎙️Listen on your favorite channel:
🎧 Spotify
▶️ Watch the podcast with video on YouTube:
Timestamps:
00:00 – Introduction
01:19 – Gilbert introduces Bill as a lifelong learner
01:49 – Learning vs. knowledge
02:44 – Example of unlearning at Yahoo
04:51 – Shift from BI to value creation
06:53 – Why data pros love tech
08:56 – Rise of ChatGPT and technical skills
09:24 – Integrating GenAI in his process
12:00 – Data culture and AI literacy
14:38 – Empowering frontline innovation
16:49 – Design thinking over data skepticism
17:36 – Asking better questions with curiosity
19:38 – How to question without annoying
22:50 – Business must guide the journey
23:15 – The IKEA effect in data science
24:39 – Thinking like a data scientist
26:09 – Measuring risk and unintended consequences
29:09 – AI helps with trade-off decisions
29:36 – Journey maps and personas
31:55 – Thoughts on the analytics translator
34:14 – Empowerment means taking ownership
35:36 – Why human behavior matters
37:35 – Sharing knowledge, one person at a time
38:19 – How Bill writes and teaches
40:40 – Where to follow Bill
41:59 – Final takeaway: embrace learning
43:14 – Closing remarks and gratitude
🎧 Episode Summary: From Insight to Impact with the Dean of Big Data
In this episode, 🎙️Gilbert Eijkelenboom sits down with 🗣️Bill Schmarzo—known widely as the "Dean of Big Data." With 30+ years of experience across business intelligence, analytics, and data science, Bill unpacks the evolution from data-driven to value-driven decision-making. They dive deep into the mindset shifts required to build a strong data culture, the role of generative AI in data science, and how data storytelling techniques can empower teams to create business-focused data insights that truly drive impact.
You'll hear practical advice on:
The difference between data-driven and value-driven approaches
Building trust with stakeholders by understanding their decisions
Why presenting data to stakeholders requires curiosity and empathy
How to gain stakeholder buy-in through collaboration and design thinking
The IKEA effect, unintended consequences, and the power of asking “why?”
🔍 Introduction
🎙️Gilbert Eijkelenboom:
Bill, you’re known as the Dean of Big Data. But beyond your decades of experience, you’re also a lifelong learner. I’ve seen you constantly taking notes in every conversation. Have you always been like this?
🗣️Bill Schmarzo:
Yes—and I’d say learning is even more important than knowledge. If you cling to old beliefs, you limit your ability to grow. I’ve had to unlearn a lot, and sometimes that’s the hardest part. One example? When I moved from the structured world of BI into data science.
💡 From Business Intelligence to Value-Driven Data Science
🗣️Bill Schmarzo:
Back at Yahoo, I was VP of Advertiser Analytics, coming straight from a structured BI background—think dimensional models, clear questions, rigid processes. But in data science, it’s not about defining questions. It’s about discovering them. That shift from BI to data science meant embracing experimentation, failure, and ambiguity.
🎙️Gilbert Eijkelenboom:
And that’s when you began emphasizing value over data, right?
🗣️Bill Schmarzo:
Exactly. In this field, it’s easy to fall in love with technology—especially with tools like generative AI. We get caught up in what it can do—make graphics, rap songs, or research topics. But the real question is why we’re doing it.
:We need to ask: What decisions are they trying to make? What KPIs matter to them? Only then can we decide which tools, data, and models will help create business-focused data insights.
🎙️Gilbert Eijkelenboom:
Why do you think so many data professionals struggle with that mindset shift?
🗣️Bill Schmarzo:
Because technology is dazzling. We’re gadget heads by nature. But unless we understand the human journey—what stakeholders are trying to accomplish—we’re just throwing hammers around with no plan.
🤖 Generative AI and Confidence-Building for Analysts
🎙️Gilbert Eijkelenboom:
With the rise of tools like ChatGPT, do you think technical skills are becoming less important?
🗣️Bill Schmarzo:
Totally. Let me give you an example. I’ve been integrating GenAI into my Thinking Like a Data Scientist methodology. It's like having a research assistant who never sleeps and works for free. It helps me identify use cases, suggest algorithms, and even generate pseudocode—all based on the decisions we want to support.
It’s pushed me to ask better questions—third, fourth-level questions. GenAI can handle the first two. But when you start asking why certain algorithms are better, or what unintended consequences might arise, that’s when it gets interesting.
🎙️Gilbert Eijkelenboom:
So it's not about replacing data scientists, but empowering stakeholders to ask better questions and be more involved?
🗣️Bill Schmarzo:
Exactly. Stakeholders don’t need to write code, but they need to understand how their input shapes the outcomes. Confidence-building for analysts comes from knowing they’re solving meaningful problems—not just playing with models.
🤝 Building Trust and Managing Stakeholder Expectations
🎙️Gilbert Eijkelenboom:
How does this mindset shift relate to building a data culture?
🗣️Bill Schmarzo:
After publishing my previous book on data economics, a CIO held it up and said, “This is our Bible. But how do we get the culture part right?” That’s when I realized we need to empower everyone in the organization—not just the data team—with basic data and AI literacy.
People need to understand how presenting data to stakeholders works, even if they never write a line of code. Decision frameworks, basic stats, understanding value creation—it’s all essential to making better choices across the organization.
🎙️Gilbert Eijkelenboom:
And this includes helping frontline staff envision what’s possible with data, right?
🗣️Bill Schmarzo:
Yes! That’s where the real corporate innovation happens. Not at the C-level, but at the front lines—where people are solving real problems. With the right literacy and tools, they can generate incredible ideas. That’s why I’m such a big fan of design thinking.
🧠 Using Stories to Drive Data Impact
🎙️Gilbert Eijkelenboom:
So if we teach stakeholders how to ask the right questions, and teach data scientists to delay skepticism, we create a shared language?
🗣️Bill Schmarzo:
Absolutely. In fact, when I managed data science teams, I spent more time teaching design thinking than algorithms. They learned to empathize, build personas, map journeys—because those tools gave them context for their models.
🎙️Gilbert Eijkelenboom:
That reminds me of something from our training: the IKEA effect. When people help build something—even a dashboard—they value it more.
🗣️Bill Schmarzo:
Exactly! When stakeholders co-create solutions, they’re more invested. They gain stakeholder buy-in and become advocates for the final result. Plus, they can explain the solution to others—something that’s often hard for analysts to do alone.
🗨️ Engaging Non-Technical Audiences with Data
🎙️Gilbert Eijkelenboom:
Some analysts worry they’re annoying stakeholders by asking too many questions. How can they dig deeper without damaging relationships?
🗣️Bill Schmarzo:
It’s all about setup. Let stakeholders know upfront that you’re here to solve their problems—not quiz them. Frame your questions as necessary for identifying the right data features or transformations. Emphasize that this is a collaborative journey, not a handoff.
🎙️Gilbert Eijkelenboom:
And that journey helps define what “good enough” looks like—not from the model’s perspective, but from the business one?
🗣️Bill Schmarzo:
Yes. Whether 75% accuracy is acceptable depends on the context. For marketing, maybe yes. For healthcare, probably not. It’s the stakeholder who understands the cost of being wrong—and they need to be part of that evaluation.
🧭 The Process of Thinking Like a Data Scientist
🎙️Gilbert Eijkelenboom:
We touched on your book The Art of Thinking Like a Data Scientist. Can you walk us through that process?
🗣️Bill Schmarzo:
Sure. The first—and most critical—step is deeply understanding the problem. What are the KPIs? What does success look like? What are the risks if we fail? And now, especially with AI, we must consider unintended consequences. That part scares me. These models move fast, and we need guardrails.
Step two is bringing stakeholders together. Ask them: What are your outcomes? What decisions do you need to make? What KPIs matter to you? You’ll get a more diverse—and often conflicting—set of inputs. That’s a good thing. Innovation comes from friction.
🎙️Gilbert Eijkelenboom:
And this is where design thinking comes in?
🗣️Bill Schmarzo:
Absolutely. Tools like journey maps and personas help us empathize with different user types. The journey map, for example, reveals what customers go through across five stages—what decisions they make, what pains they feel. Personas then add depth: What are their backgrounds? Aspirations? The more we understand that, the better we can align business-focused data insights to their needs.
🔄 Should We Rely on Analytics Translators?
🎙️Gilbert Eijkelenboom:
McKinsey once wrote about the role of the analytics translator—someone who sits between business and data science. What’s your view?
🗣️Bill Schmarzo:
I’m not a fan. Let me explain. When my mother-in-law moved to a care facility, they started doing everything for her. Over time, she stopped gardening, cooking—her skills atrophied. The same happens when we insert a translator between business and data teams. Everyone becomes less capable. Blame gets passed around.
🎙️Gilbert Eijkelenboom:
So we should encourage direct interaction?
🗣️Bill Schmarzo:
Exactly. If stakeholders and data teams collaborate directly, they share accountability, develop empathy, and create more valuable outcomes. You can’t outsource that. Maybe GenAI can support as an aid, but not as a replacement.
📚 Learning, Legacy, and Leadership
🎙️Gilbert Eijkelenboom:
I love how you tie data science back to human behavior. But why don’t we see more of this in data education?
🗣️Bill Schmarzo:
Because we haven’t done a good enough job sharing it. I write books, teach, speak—because I feel a responsibility to give back. We can change the culture one person at a time. Offer the book. Share the message. Let people decide for themselves what resonates.
🎙️Gilbert Eijkelenboom:
And your latest book, AI and Data Literacy, reflects that mission.
🗣️Bill Schmarzo:
Yes, and it’s about simplifying complex ideas so everyone can participate. I wake up at 4:30 AM, write at my coffee shop, and keep working between flights and meetings—not because I have to, but because it’s fulfilling. And honestly, a lot of that drive comes from my wife. She never lets me settle.
🎙️Gilbert Eijkelenboom:
That’s beautiful. We all need someone who pushes us to be better.
🗣️Bill Schmarzo:
Exactly. Whether you’re into data science, art, or even stamp collecting—you need that support system. And you need curiosity.
📣 Final Thoughts and Takeaways
🎙️Gilbert Eijkelenboom:
Where can people connect with you?
🗣️Bill Schmarzo:
LinkedIn, hands down. It’s not just about me—it’s the community. We challenge each other, respectfully disagree, and grow. No one’s an expert anymore. Not in this fast-moving world. We’re all learning together.
🎙️Gilbert Eijkelenboom:
And your big takeaway for listeners?
🗣️Bill Schmarzo:
Develop a passion for learning. Embrace diverse perspectives. And treat every conversation as an opportunity to grow.
🎙️Gilbert Eijkelenboom:
Bill, thank you so much. You’ve helped me—and many others—rethink what data can really do. From asking better questions to building trust and enabling impact, you’ve shown us the path. Thank you for being so generous.
🗣️Bill Schmarzo:
Thank you, Gilbert. It’s mutual. We’re truly better together.
Comments