Skip to content

Episode 13: Photography, Philosophy, and Physics: Shaping Perspective on Data Strategy

In this episode, Jeff Klagenberg, a data and application architecture thought leader, shares how his love of photography, philosophy, and physics have shaped his perspective on data strategy, and the role of confirmation bias.

Listen now


Tina Tang

VP of Product Marketing at Privitar

Jeff Klagenberg

Data and Application Architecture Leader



Intro: Welcome to InConfidence, the podcast for data ops leaders. In each episode, we ask thought leaders and futurists to break down the topics and trends concerning IT and data professionals today, and to give us their take on what the data landscape will look like tomorrow. Let’s join the data conversation.


Tina: Welcome to InConfidence, the podcast for data ops leaders. My name is Tina Tang, your host and I’m here today with Jeff Klagenberg with me. He’s a technologist and data strategy leader based in Silicon Valley. He currently leads data architecture and governance at Lucid Motors, an American luxury electric car manufacturer. Jeff has led product teams in telecom, finance, aerospace from companies as early stage startups all the way to large enterprises. You should also know he’s an avid darts player and Wordle player and a self taught photographer. Welcome to InConfidence Jeff.

Jeff: Oh, thank you, Tina. Glad to be here.

Tina: So first, we’re gonna get warmed up because these are facts that people in our audience need to know. IPA or lager?

Jeff: IPA, definitely. Although I would actually tend a little bit more towards stout.

Tina: Oh, yes. I like that. I like that choice. Neat or on the rocks? 

Jeff:  Definitely neat. 

Tina: Definitely neat, okay. You don’t like any kind of dilution, right?

Jeff: Uh, I also don’t like to cool the drink down because then it, then it kind of quenches the flavors for me.

Tina: That’s a good point.

Jeff: I want to taste the flavors of my, of my preferred whiskies. I’m still

Tina:I’m still growing in that realm. So I gotta get there. Its baby steps for me. I’ve moved away from mixed to on the rocks. Next step is neat. Okay. But more importantly, chocolate or cheese?

Jeff: Uh,  cheese, cheese has always been my weakness.

Tina: Mountains or ocean?

Jeff: That’s a much, much more complicated question for me. But I have to say ocean. At the end of the day, I can live a lot longer away from the mountains than I can live away from the ocean.

Tina: But luckily, you don’t have to do…do away, I mean you don’t have to do without either because you have boat right?

Jeff:That’s true. That, that is the best of both worlds.

Tina: Right! Okay, last one, true crime or sci fi?

Jeff:Sci-Fi. I grew up in the era of space and sci fi. So you know.

Tina: Yup, I’m with you on that one. So, you know, you and I, we met for coffee. I guess it’s been a few months now. And we got into this discussion about the word “the”. What is so dangerous about the word “the”?


Jeff: You know, it is funny because data and language, they go hand in hand, right? And the word “the” –  I pronounce it the – is a very, very dangerous word because it does two things that are very damaging. So one is, think of the phrase, “I know the solution to the problem”. And you’ll hear that and immediately, the first thing that hits my mind is there’s very few problems that have one solution, and the implies a singular solution. But even just as bad as that, there are very few times that you have only one problem. So it actually makes the singular out of both the problem and the solution. And what that does is it keeps people from looking further, looking for more solutions or trying to understand what’s the biggest problem or what’s the most impactful problem to solve. And that’s only one problem with the word “the”. Probably the worst problem with it is that it creates a false sense of homogeneity, right? So if I take the phrase “the customer”, right? Well, unless I say the customer is somebody who buys my product, I’m probably inaccurate because I’m going to be grouping all customers and assuming a type of behavior. The worst place that this is used is actually in political speech, of course, to purposefully limit or focus or categorize a group of individuals as one thing. And it’s just used often in this, even in business, when we’re talking about customers, or we’re talking about employees, or we’re talking about products, or our competitors. And so it’s very, I try to be very careful about highlighting when “the” is used either to create an artificial singularity, a singular thing, or to group a set of unlike things together.

Tina: That is really powerful, surprising, surprisingly powerful stemming from a three level, three letter word, which I had to refresh my grammatical knowledge, is an article in the English language. Now, I think that this is like really resonating with me that people, including myself, have a tendency to group things and stereotype them. But in many situations, that is not a good thing to do, because you miss all the nuances and the differences that are important to know about them.

Jeff: That yeah, that’s okay. That’s that that is very, very, very true.

Tina: And when we’re talking about data strategy and practices, this is a very important concept and sort of I’m getting, I’m thinking it’s the philosophy of “the”.

Jeff: Yeah, just don’t call it “the philosophy”. A Philosophy of “the”


Tina: So I know, because I know you, that you have a degree in physics, and that, you know, as mentioned, that you’re a technologist, and have had this very successful career in Silicon Valley. And that you’re a photographer, but can you share with our audience, and it seems like philosophy should have been in there, too, but it’s not. You don’t have a degree in philosophy, though, you clearly are a philosopher. But how do physics, photography and tech go together? Where are the threads? What’s the common thread there? Could you share that with the audience?

Jeff: Yeah, I mean, for me, you know, photography is about observation, right? It’s about finding, you know, finding something interesting and being careful about capturing it. And then, and it’s also about not just capturing a moment, or capturing a single image, it’s, it’s also about presenting it and showing people and that’s true of science. And that’s true of technology as well. I mean, you’re, you’re really trying to find an interesting solution, or an interesting fact, or an interesting piece of the scene that gets overlooked. So I find it actually very, very synergistic. And to be honest, you know, one of the funny things is my wife and I are both photographers. And as we’ve gone in and met other photographers, it’s not an uncommon pairing, science or engineering and, and photography, in particular. Other artistic forms,  music, is not uncommon as well. But yeah, it’s, it’s for me it’s, it’s also how I keep my sanity to be honest or what little sanity I have left.

Tina: I’m laughing and crying at the same time. Feel the same! Yeah, I was going to agree with you that also on the music part, I find that these three themes are recurrent in a lot of you know, friends, colleagues, family, who are also in these spaces that they have these things in common and i, i can, listening to you talk now I understand that this is also related to the philosophy of “the” – not missing the details.

Jeff: Trying not to miss the details, but I’m a physicist, so I gloss over details as well.



Tina: Okay, so there’s that duality, all right, all right. So, you know what, what inspired this career journey for you? What.. what was that journey, like with all these seemingly divergent interests, but actually have this thing in common? What was that journey like?



Jeff:Well, the first journey, the first step for me out of college. So, you know, I grew up in the space age, right? I mean, everything was, you know about space. And that made science fascinating to me. And when I went to college, you know, for whatever reason, I ended up gravitating towards physics. And that, I just found that, that, really trying to unlock the mysteries of the universe, so to speak. And I was fortunate that after I graduated, I was able to work at Lockheed on, on space shuttle payload support. So there was a number of missions involving the space shuttle to launch different spacecraft like Magellan, Galileo going to space. So that was clearly a jump because of interest in space. But as we were launching these vehicles, there’s this incredible amount of software and hardware that’s needed to support and get telemetry. And it’s funny because that now this all relates to IoT, but you have the system. And, and so I got trained as a systems engineer at Lockheed as part part of my job. And as I worked on these systems, I became more enamored of that. And I started working at a startup that was aimed at using, using the software to support satellites and applying it to other industries. And that was fascinating, right, so I got to work with, you know, the New York Stock Exchange and to work with CERN, and just this wide variety of mass general hospitals supporting brain cancer treatment, all these use these different command and control systems. But I also found that these systems were often limited by the data they needed to do exchange. And so then I moved into this area called Master Data Management, and data governance, which is where I live today. Because if you can’t make the data work together, you can’t make systems work together. So that’s how I ended up here. And I found that it both was something that people needed, or I shouldn’t say both, there I going creating a limitation here, that’s going to be a problem. I, I see that it’s something that’s needed. That different people, different groups really are seeking out and something that I enjoy as well as seem to have a unique understanding, or at least an unusual understanding, certainly not unique. There are other people that work in the same space, but I just, I just enjoy it. So…


Tina: So So, you know, now that I met you during your master data management and governance phase. And, but now I can see, you know, why the challenges and opportunities presented by an electric car would be so fascinating for you. It’s kind of like the coming together of your worlds.

Jeff: Very much so. Which is it to me is very, very cool. Yes.

Tina: So, you know, Lucid has been in the news a bit lately, they have not only had some pretty stellar reviews, you know, by Car and Driver, and I think I read one in Bloomberg Businessweek as well. They’ve created this beautiful car. And it was reported, I think back in April, that Lucid signed an agreement with the government of Saudi Arabia, for the purchase of a large number of vehicles, I think 100,000 over 10 years. And that as part of that purchase, the Kingdom has committed as part of their vision 2030 initiative and their green initiative, has committed to this relationship with Lucid, which is amazing, not just for the earth, for Lucid, and the kingdom as well because they’re diversifying out of fossil fuels. Now what Is it that, you know, in your role for Data Architecture and Strategy at Lucid? What do you hope to achieve for the company? 

Jeff: Yeah, well, one, you know, first, I’m very proud to be a part of the Lucid team. And these objectives are more impactful, not only to the company, to me personally, but also to the world, which is, which is great. And just like any other company, the, my main goal is to always to get the most out of data to make the company and teams successful going forward. I mean, that’s, you know, the point of data is to create success to create new things, right? It’s not to, you know, have the data. And it’s not to, you know, too many times people focus on collecting data and not using it. And my goal is always to make the data the most useful.

Tina: Is there any – What kinds of challenges do companies like yours and maybe even in your past, what kind of challenges do organizations face in using that data? Because you mentioned that sometimes they just collect it, and they sit on it 

Jeff: Yeah. Well, let’s go. 


Tina: What’s part of those challenges?

Jeff: Yeah, so some of those challenges are very, very common, you know, as you go from, from group to group. Probably the first primary challenge that I see is that people have this idea of,  “I’m just going to collect data, and it’s magically going to answer questions”. So as I think of many of the organizations I helped out in Master Data Management, that was always a key barrier. So it’s working with the business leadership and understanding what questions they really need to answer and, and, you know, if you understand the questions, if you understand what’s going to create success, then you can understand not just what data you’d need, because you may be missing data, but also how to best present it how to best organizing it. As you know, in the master data management world, what you’re often doing is you’re taking these disparate datasets, your CRM system and your product system, you know. So, for example, I have a spike in orders, what suppliers do I care about? And to connect that data, you have to understand that that’s the type of question you’re going to answer. Because then you’re going to have to make sure those two datasets can talk to each other, which often involves cleansing and managing, fixing up reference data. And if you didn’t know, that was one of the questions, you might not do that correctly. So it might be, it might make it more difficult. And then making the data accessible. So you know, there’s a whole movement around data cataloging and metadata, so that people who come up to the dataset, so my favorite example is SAP. SAP has a tremendous amount of value for analytics and supply chain and, and for finance, and for even manufacturing or other parts of an organization. But it’s not exactly the most transparent data. It often needs support or additional explanation by people that know what the data is. It’s not that SAP data is great. It’s not it’s not bad, but it’s opaque to the average analyst. And so having ways to make that easier to understand, especially in context of how does this SAP data relate to this data in Salesforce? So how do I try, track all the way from an order to SAP and to my manufacturing system? How do I chase, you know, what vendor parts did well in my manufacturing product, things like that. So that’s always, that’s always a barrier. And then you have the challenges of scale. And I’ve worked at so many different scales. And often you have challenges with small datasets where you need so much information. You know, you’re looking for every bit of how a customer connects to my organization, to others that have a tremendous amount of data and so their challenge is how to make that cost effective. I actually wrote an article at one point in time talking about the density of data. If you think about the value of data per bit, basically, you know, your ERP system, your CRM systems have an incredible, incredibly dense data. And so there’s a lot of value because they’re directly tied to your business transaction to your customer. And then you have the Clickstream data, which is got a tremendous amount of value, but it’s not very dense, you have to have a lot of it to get the data out. And processing data, that sort of scale has different costs. And so how you manage that and how you filter it, how you choose what pieces of it to process is definitely very important.

Tina: What about the, you know, in, in today’s world of entities, not only collecting data, but also generating data, perhaps from the services, systems, products, that they manufacture and sell. Let’s say, you know, you make a some kind of wearable device, and you are doing a great business in, you know, the country that you started it in, and then you’re doing so well that you want to expand your operations outside of this country. What are some of the challenges that you might face because of differing data, cultures, regulations, different rules, different ways of handling things? Does that? Is that like, something that is a common challenge? Or have people figured that out and it’s no big deal?


Jeff: Believe me, it’s a common challenge. And, and it’s not necessarily well figured out, in general, and…

Tina: I mean it’s not a switch that you flip, right

Jeff: it and sadly, you know, I’ve certainly run into organizations that think, “Oh, I’m using this software package. So therefore, when I scale out, you know, and I go into the EU, or I go into, into Germany, I go into Latin America, that I’m not going to have a problem”. And of course, you know, you do. I like the fact you use the term culture, because there’s one, there’s the compliance right there. There’s different regulations, and all these different jurisdictions about privacy. There’s also the culture, there’s the customer expectation of how their data will be handled, and what their expectations are, in terms of, of insight into that, that data, how you handle consents, and, of course, language around consensus, not just from a legal standpoint, but in making your customer comfortable and understanding exactly what you’re going to do with it. And there’s a whole movement around privacy by design. And, you know, as part of that, it’s really filling out more metadata around your data and knowing exactly what fields are, you know, sensitive or have special meaning to your customers, or proprietary information to your company, right, so that you want to protect more. And then making sure that’s available and understandable and usable in terms of automation. And there’s a tremendous need for more and more automation to make it easier to give the right people access, and to protect the data more robustly, because all of us want to protect our customers data. We don’t want our customers data to inadvertently get to a place that shouldn’t. And, you know, we also want to make sure we’re managing consent so that if we have a phone or a Fitbit or whatever, we have that if we’d said “no, I don’t want it to collect my data”, that that actually gets enforced immediately so that you know, you don’t have a problem with that data somehow leaking. So it’s automating consent, automating enforcement, automating policies, it is all something that everybody I know is working on right now.


Tina: And so in, in these situations where it’s it’s a subtlety, but you know, most people associate privacy with compliance, which is like a word that makes people think of kind of shrinking and becoming, you know, stepping away. But another way to think about it is growth, I want to expand, and these places have a different way of doing things than what I might be accustomed to, but because of growth, I’m going to, you know, abide by the local laws and rules and practices. So it’s, it’s like a growth mindset for protecting data, and respecting our customers and users.

Jeff: Yeah, I would agree with that. And understanding that, even though there’s a lot of similarities, there’s subtle differences. So making your choices that can adapt easily just you know, that’s going to be, that’s always a key part of data. In general, whether you’re talking about analytics, or you’re talking about privacy. Just for me, in terms of a privacy philosophy, I always want to make sure that we’re, that any organization is privacy forward. And if I feel like the compliance needs are constraining me, I feel like I’m doing something wrong. Because for me, most of the compliance rules are the table stakes, they’re the, the bare minimums of protecting people’s data. And most organizations that are customer forward, are better served at exceeding those compliance needs. So obviously, monitoring for them and making sure you are not out of compliance is important. But usually, the rules themselves are not particularly onerous, in my opinion.


Tina: Right Um, and just I mean, from, from a customer’s perspective, there are a lot of regulated industries, which, you know, protecting or transforming the data, you know, the sensitive data, so that, you know, reduces the risk of re identification, or, you know, exposure, or exploitation is one thing, but to actually get those datasets into the hands of, you know, downstream users who need to use it for like value added initiatives, like, you know, in data science, or data analytics is quite something else. And in those cases, would, would you agree that the barrier is not the technology? It’s the process or lack of process? In those situations? Have you experienced anything like that?

Jeff: Yeah, it’s it. If you have, even if you have a manual process, it’ll, it’ll help as long as people understand, hey, A, B, and C have to happen before I can use this data. And, and to be honest, it doesn’t matter if it’s for ensuring that data is de-identified, or it’s I have to cleanse or standardize my reference data, there’s still a process, I need to know how to prep that data and how to make sure the appropriate people are getting the appropriate data. It’s just the appropriate people usually is about compliance, the appropriate data is just as often about cleansing or improving the data as it is about, you know, de-identifying and making it more private. Or protecting privacy for it. So that’s, that’s important. And then, of course, doing this at scale, doing it, making it faster and getting the business value as quickly as possible. You know, that typically requires not just the process, but some level of automation. And like I said, a key part of enabling that automation is making sure you have the metadata available to enable to help enable the automation as well. So all of those things are, are necessary. But you have to think about the process.

Tina: Right? I mean, I agree that you know, even if it’s a manual process with paper forms it’s  better than nothing. But I will say in favor of technology is that if you need lineage and auditing capabilities, you need technology to help.

Jeff:Yeah, I will definitely agree with them. As you look at the ability to audit, obviously, writing something down in a paper log is not, is not particularly searchable. So if you’re looking at enabling and speeding up and, and all the things we want to do, our technology is a first, is a necessary and important piece of that.


Tina: Right, right. So, um, what – what are some of the things that you’ve faced in your many roles? You know, you’re, you’re a practitioner, you’re in the trenches, and you’re facing a lot of different situations in different industries at different scales and speeds, right? Can you talk about some of the, like, maybe something that was really challenging, but that you were able to overcome and, you know, had a unique solution to, like, what’s something that you’re proud of? Show and tell time.

Jeff: Well, I mean, it, it’s funny, because in data, you know, a lot of times, I feel like it’s pretty banal, but you know, there’s a case where I was helping out a regulator, who had a paper form process. So we’re, we’re going through digital transformation with them. And if there was several groups that just weren’t managing to get the data to work, and they’re trying to understand the rest of the supply chain, the drug supply chain. And one of the problems is you had these forms, and so off of a form, they would create an entity. And what happens is, I was looking at one of these forms, and I realized, well, there’s two entities right there, there’s the actual physical facility that’s being used for production and there’s the company that manages that facility. And that simple breaking that up into two separate entities, then you have the things that are associated with the process management of how things are operated at that facility. And then you had things that were, I inspected, the, the facility was inspected, and had defects or had problems. And then as you looked at the data over time with the supply chain, the business owners, you know, they, they sell facilities, or they buy facilities, so you can see all this facilities moving around, and you have a set of problems that would chase that. And then you have, you know, businesses that are buying, facilities are moving and you can see all different set of problems are associated with that. And so it’s a, you know, it’s really decomposing entities into something that’s more appropriate for the ground truth, what you’re really trying to understand. And that’s, so that’s something I’m, in particular, proud of, because that’s still am being used today in terms of breaking that up and getting more granularity in terms of the supply chain. And often, language is a key driver and understanding that as you listened to groups talk about what they’re trying to do, you just, you ironically, listen for articles, right? When people talk about A or talk about the,  that’s a that’s a clue that hey, that’s an entity. And, you know, so that’s a key part of, of, of doing that type of analysis.


Tina: I was, I was talking to someone earlier today and as he was speaking, I, I had this, like kind of epiphany, just like we, you were just talking about being a really good listener, and listening for the clues, or the hints of the words that people use. And I just, it suddenly kind of came into my mind. It’s like the data practices or the practice of data strategy management is kinda like Buddhist. It’s very, like you have to have this compassion and empathy for what the business wants to do, and what they’re actually doing and where they want to be. You know, it’s like this kind of very self reflective process that, you know, you’re here today, this is the present moment. And you have, like, this journey that you go on to become better. And more, you know, you, there’s this whole reflection thing that has to happen in order for you to move forward. So anyway, okay, that’s my philosophy. The Zen Buddhism of Data.

Jeff: I’ll take that. 

Tina: It’s Friday. Okay. This is where my head’s at. 

Jeff: Well, we all need more Zen, so that’s a good thing.

Tina: Yeah. Yeah. I think that that’s very true. And well, and that kind of leads me to your Twitter feed. Let’s talk about your Twitter feed. Did you know this was coming? 

Jeff:My My showdy. 


Tina: Yeah. Yeah. I mean, speaking of philosophy, and data, hot takes and perspectives on this Twitter feed. So some people may or may not be familiar with Dilbert. But I, you know, you know, growing up, so to speak, in the tech world, back in the day, this is like, must read, the morning must read every morning, had to read Dilbert to get me going in the morning. It was like coffee. And there are quite a bit of Dilbert references in here. And my favorite Dilbert’s boss, the second option feels right, let’s go with that. Should we always ignore what the data says? Or is this more of a one time thing? It’s called intuition. It’s a slippery slope to witchcraft. That’s hilarious. Because it’s so true, right? We see that every day, in, in a lot of departments and divisions. In companies, it seems like this kind of overlook, or, or overly looked at processes or lack of processes can are allowed to proliferate, where the light doesn’t shine on the can kind of conduct themselves in the secret.

Jeff:Well, it’s indicative of a very, I mean, just like the word, there’s different different biases. And this is actually studied in in data and data science, in particular, lot, different biases that the human brain really needs. And tricks us when we’re working with data, or that we need to accommodate in data. And so in, one of these is, you know, confirmation bias, where I look at data, and I just use it to reaffirm what I already believe. And it, it, it’s very pernicious, it’s amazing how often I catch myself in it. And, you know, even if you know about it, you’re a human being, and so you’re going to be subject to it. And there’s definitely techniques to try to, to deal with it. And one of them is don’t look at the data first, you know, figure out what you’re going to try to affect because at the end of the day, data doesn’t do anything unless you’re using it to inform or change a decision, right? If you just do it, you’re going to do anyhow. Don’t waste the money on the data. So, so it’s a, one of the techniques is you go and you pre ask yourself the question, what data will affect this decision? What am I looking forward to make the decision one way or the other, then look at the data, then it’s harder, right? Then you know, you’re sitting there and you’ve looked at the data. But the things that will happen is you’ll go and you’ll see the data and it doesn’t agree with you and you go oh, I can’t trust that data, but I can trust the stuff that agrees with me. Right? These are all techniques our brain has developed to lead us down the path of confirming our decision based on data. But yeah, so I appreciate, Dilbert brings in a lot of that,that sort of thing in the text.


Tina:So I mean, along those lines, is data actually valuable? Or is it too risky?

Jeff: Oh, that’s a that’s a lovely question. Data is if you unlock the power of data to power your decisions, right, and I don’t like the people who say, data driven decisions, decisions are always driven by choices. They’re, they’re driven by your business strategy. They’re, they, data can power those decisions, can make those decisions more effective, but they don’t drive those decisions. And if you actually look at the data, challenge yourself in terms of what it says, If you realize that data is never perfect, and therefore don’t wait to make a decision until you’re convinced the data is perfect, then yeah, it can make incredible, driving credible decisions or making business far more effective. The challenge is, like I said, don’t wait for it to be perfect. Understand what’s good enough in terms of the data. And, you know, don’t disregard your intuition by any means, but use the data to question that decision, that intuition and refine it. And, you know, also use it as a way to mitigate that. And I will say, it’s one of the the options that people don’t think about is, often they think about, oh, I’m gonna use the data and make a decision one way or the other. But you can also use the data say, Oh, I’m gonna go this way. But I understand it may be riskier so I’ll start building a mitigation plan to say, hey, if it starts not working the way I want, I can plan a different path. And the data can lead you to what’s that, that plan B if if your plan A doesn’t work out?

Jeff:But so, it that it is risky, it is risky, from the standpoint of data has risks, data has cost. So always account for those as well. So it’s a you know, and quite literally, if you’re really really not going to use the data, then you really have to consider you know what you’re collecting and why

Tina: That sounds like solid advice. And a great takeaway for our listeners. Jeff, I want to thank you for your time, speaking to our audience today. And it’s been great catching up with you.

Jeff: Always, always a pleasure team. We’ll have to do coffee again soon.

Tina: Yes, definitely.

Outro: No matter where you are in your data journey. Privitar is here to help. Privitar empowers organizations to leverage their data to innovate faster, while protecting the privacy of individuals at massive scale. Privitar is unique in combining technology, thought leadership, and expert services to help your data operations thrive. Want to learn more? Our team of experts is ready to answer your questions and discuss how data privacy can fuel your business. Visit Thanks for listening to InConfidence brought to you by Privitar. To hear more insights and advice on how to effectively use, manage and protect your data. Subscribe to the show and your favorite podcast player. If you liked the show, leave us a rating. Join us for the next data conversation.

Ready to learn more about Privitar?

Our experts are ready to answer your questions and discuss how Privitar’s security and privacy solutions can fuel your efficiency, innovation, and business growth.