Podcast: Play in new window | Download
Subscribe: Apple Podcasts | Spotify | Amazon Music | Android | Youtube Music | RSS
Mara Inglezakis Owens brings a human-centered focus to her work as an enterprise architect at a major US airline.
Drawing on her background in the humanities and her pragmatic approach to business, she has developed a practice that embodies both “digital anthropology” and product thinking.
The result is a knowledge architecture that works for its users and consistently demonstrates its value to key stakeholders.
We talked about:
- her role as an enterprise architect at a major US airline
- how her background as a humanities scholar, and especially as a rhetoric teacher, prepared her for her current work as a trusted business advisor
- some important mentoring she received early in her career
- how “digital anthropology” and product thinking fit into her enterprise architecture practice
- how she demonstrates the financial value of her work to executives and other stakeholders
- her thoughtful approach to the digitalization process and systems design
- the importance of documentation in knowledge engineering work
- how to sort out and document stakeholders’ self-reports versus their actual behavior
- the scope of her knowledge modeling work, not just physical objects in the world, but also processes and procedures
- two important lessons she’s learned over her career: don’t be afraid to justify financial investment in your work, and “don’t be so attached to an ideal outcome that you miss the best possible”
Mara’s bio
Mara Inglezakis Owens is an enterprise architect who specializes in digitalization and knowledge management. She has deep experience in end-to-end supply chain as well as in planning, product, and program management.
Mara’s background is in epistemology (history and philosophy of science, information science, and literature), which gives a unique, humanistic flavor to her practice. When she is not working, Mara enjoys aviation, creative writing, gardening, and raising her children. She lives in Minneapolis.
Connect with Mara online
- email: mara dot inglezakis dot owens at gmail dot com
Video
Here’s the video version of our conversation:
Podcast intro transcript
This is the Knowledge Graph Insights podcast, episode number 34. When think about architecting knowledge systems for a giant business like a global airline, you might picture huge databases and complex spaghetti diagrams of enterprise architectures. These do in fact exist, but the thing that actually makes these systems work is an understanding of the needs of the people who use, manage, and finance them. That’s the important, human-focused work that Mara Inglezakis Owens does as an enterprise architect at a major US airline.
Interview transcript
Larry:
Hi, everyone. Welcome to episode 34 of the Knowledge Graph Insights Podcast. I am really delighted today to welcome to the show, Mara, I’m going to get this right, Inglezakis Owens. She’s an enterprise architect at a major US airline. So, welcome, Mara. Tell the folks a little bit more about what you’re up to these days.
Mara:
Hi, everybody. My name’s Mara. And these days I am achieving my childhood dream of working in aviation, not as a pilot, but that’ll happen, but as an enterprise architect. I’ve been doing EA, also data and information architecture, across the whole scope of supply chain for about 10 years, everything from commodity sourcing to SaaS, software as a service, to now logistics. And a lot of my days, I spend interviewing subject matter experts, convincing business leaders they should do stuff, and on my best days, I get to crawl around on my hands and knees in an airplane hangar.
Larry:
Oh, fun. That is … Yeah. I didn’t know … I knew that there’s that great picture of you sitting in the jet engine, but I didn’t realize this was the fulfillment of a childhood dream. That’s awesome. But everything you’ve just said ties in so well to the tagline on your LinkedIn profile. You’re like, “I’m a people-loving architect, and data leader.” And one of the things I love about that, we talked a fair amount at the knowledge graph conference about your background in the humanities-
Mara:
We did.
Larry:
… and your transition into your current role. I would love to hear … And what you just said, like, the end of what you were just saying about so much of your job is about interacting with people, and convincing business leaders to fund you, and stuff.
Can you talk a little bit about that? Like, what drew you into the humanities in the first place, your transition out of it, and here we are today.
Mara:
100%. Before I talk about being in the humanities, I love to read, I was an epistemologist, and a 19th Century scholar. But before that, when I was a little girl, I was writing my own websites in HTML, XML, and some of the technologies that eventually got to be used in the semantic web, which is how I entered the knowledge graph space way later as an adult. So, that got put on hold.
Mara:
I love to read. So, I became a humanities scholar, and I was for about five years, the lowest of the low adjuncts at an R1. My teaching experience, not my scholarship, although, I did a lot of thinking about how people interact with written media, and how they enter internal argumentation with those media, and come to know the world differently. That’s what most of my work was about. It was interdisciplinary with literature, history, philosophy of science, which is why I say epistemology.
Mara:
But my best teacher for coming to where I am today was being a teacher. So, lowest of the low, first year, although, I spent most of the time teaching applied rhetoric, I was teaching freshman comp. So, this is a super diverse group of students who are showing up for a required class. To be successful, I needed to do two things. One, I had to listen carefully to what these students cared about to actually get them to get something out of the between $5000 and $8000 they were paying for this course. And then I had to generalize what I wanted them to learn about enough to make it accessible to them. Okay? So, my goal throughout my teaching career, similar to my goal now, is to inculcate effective communication through fit for purpose argumentation.
Mara:
So, while a lot of my colleagues were being like, “Here’s an essay. Write something about it. Make it sound smart,” what I did, because I needed my students to hook in, to be engaged, because the vast … I maybe, I don’t know, taught, like, four English majors over my career … No HSPS [humanities, social, and political science] people. I told my students, “Okay, guys. Get into groups. So, you’re set up to do some argumentation amongst yourselves, pick a little part of this essay,” this was the first year, “And something you react strongly to. What about the sentences are doing this? Grammar, syntax, semantics. What’s the whole universe of your group reactions? How are they related, or not related?”
Mara:
This evolved into a directed research curriculum in my applied rhetoric courses. So, I said, “Okay. Okay, guys. Go find something out in the world that needs to change. We need a pedestrian bridge over the street. We need better accessibility for people with disabilities in our gym. We need better gym hours. Figure out how it’s working, frame up a case for someone who can make a change, do your argumentation, go present it.” Some of my students actually argued well enough that they got a stoplight installed on a really busy street corner. So, it worked.
Mara:
So, fast-forward, lots of life drama that brought me out of the humanities into what is a much better place for me, in corporate. I’m in a trusted advisor role, not so dissimilar to being a teacher. As a trusted advisor, I have to be attuned to what the business says that they want. So, if they’re saying … And then what they demonstrate that they want through their behaviors, and through their artifacts, often times, their processes and information system. And then I have to think about why and how those things align, or don’t align.
Mara:
Because I’m full-time employed, and this is in this role, and all of my corporate roles, but I’m, effectively, providing a boutique service. It’s not enough for me to come up with something that sounds smart, or cool. I have to come up with a solution that accommodates process data, technology, and, most importantly, people, and that actually fulfills a business need. And I used to think about the connection from my academic career to my corporate career as like, “Oh, I became a good EA, because I taught my students to do this,” but with about a decade of reflection, I’m realizing that teaching was really mutual.
Mara:
Like, I asked my students to show me what they were thinking. I evaluated what they were doing. I was very critical, but I was generous. And I was with them as their efforts bore fruit, or didn’t. But how I demonstrated, elicited, and critiqued them evolved with constant, and often very, very vulnerable feedback. Like, I do with my clients now, I constantly asked, “How am I doing? Am I giving you what you need? Do you need something else?’
Mara:
For a student, it’s really hard to say that to someone who’s got the power of a grade over you. It’s not as perhaps scary when we’re all adults in corporate, but I still think many adults … I was always the stupidest person in the room as a scholar. So, I don’t have this problem, but a lot of us are worried about appearing, “I’m not smart enough. I am not creative enough.” So, I still have to flex that good, compassionate, people-loving, listening muscle all the time in corporate just like my wonderful undergrads at Indiana University taught me how to do.
Larry:
That’s so awesome. I’ve thought a lot about rhetoric. In fact, I don’t know if we talked about this in New York, but my first career was in college textbook publishing, and I spent my first year and a half on the road with St. Martin’s Press selling all those-
Mara:
Oh my God. I love St. Martin’s Press.
Larry:
Yeah. Exactly. Everybody does.
Mara:
Beautiful novels. Yeah.
Larry:
All the English teachers. But I always thought of rhetoric as this … Not a one-way street, but a persuasion, and communication. Like, an old broadcast mentality, not pushy, but the way you just describe it, it seems like the foundation of your work is this –
Mara:
expressive
Larry:
… rhetoric, this two-way street, listen-y, interactive rhetoric approach. Am I reading that right?
Mara:
100%. I’m a Greek-American person. It’s the Socratic method.
Larry:
Ah, okay.
Mara:
I didn’t put words to that until my mentor and friend, Steve Etzell, if you’re listening to this, this is about you, at Cargill started doing this with me. This individual trained me to be an enterprise architect, and I realized one day, he was using the techniques with me that I was using with my students, and, although, I don’t have that part of philosophy, I’m deficient there, I’ve also had the benefit of really good training from him, and other teachers throughout my life to be discursive, to teach argumentation, to get what I want, and to help the people around me, my community get what they want too.
Larry:
Yeah.
Mara:
Like, it’s 100% needed.
Larry:
Yeah. And the idea of getting what you want, because that’s all … Like, especially, architects. You craft this beautiful scaffolding, or framework, or even the whole structure for a thing, with careful listening, all those skills you just have been talking about applying, but that doesn’t do any good unless you can get-
Mara:
No.
Larry:
… people to take action, and actually support you and fund you. Like, one of the things you talked about in your talk at KGC was the importance of tying … Especially when addressing the senior executive stakeholders-
Mara:
Yes.
Larry:
… like, tying things to the P&L, or other business accomplishments. Can you talk a little bit about how your rhetoric background … I’m picturing how those fit together now-
Mara:
Yes.
Larry:
… but I want to hear from you, if I’m on the right track.
Mara:
So, as an enterprise architect who is keen to deliver stuff that actually works for my community, I am all about usability and 90% of the work I do is digital anthropology, direct observation. What are you saying you want? What are you actually demonstrating that you need?
Mara:
And luckily, for me, because of product thinking that is very popular, I think first popularized by Steve Jobs for most of us, there are really strong throughlines between, “I need to be able to do this,” and, “This is the kind of money,” or the kind of return on investment, whether it’s direct cost, or indirect cost saving, that I am getting. So, my first stop when I know I am going to need money for something, or someone for whom I’m advocating, that’s the vast majority of the time, needs money, I start with, “What do you need to do?” And my job is easy. I just interview the business and walk backwards with them, “Why do you need to do this? Why is it a problem that you can’t do this?”
Mara:
And one of my previous roles, I was serving a business who had a huge, huge, huge, huge drama any time they wanted to do a big capital software deployment, because, effectively, they didn’t … They outsourced their IT management to one of the big five consulting firms, and this drove time delays, it drove cost overages. So, when talking about what would a program that actually works look like, the first thing we talked about is I don’t have to go to the vendor, I have all of this knowledge in-house that that’s what it would look like, and here are the kinds of interactions I would be able to do that I used to have to call Big Five on the phone, I’d pay an hourly rate. Wouldn’t it be nice if I had a resource internally that I could interrogate?
Mara:
So, in that case, what we did was say, “Okay. You have a usability issue. You want to be able to interrogate a resource, and not call up the third-party on the phone, and pay $200 an hour. Take the $200 an hour, and the need to discuss with a human, that’s your cost savings, and it’s an hourly cost savings, and you can multiply by how long are these program durations to get like, “This is our actual spend. That may show up in another cost center, because of how we build those relationships. This is your potential first savings,” depending upon size of program, you multiply and divide to get that.
Mara:
And that’s my formula, figure out what the problem is, what the current modality to address the program, what the real costs are directly or indirectly, and go from there. I had wonderful coaching from my former VP of finance, whose name is Molly, who sat down with me one day over coffee and very painfully with a PowerPoint helped me write this out, but thanks to the teaching of that mentor, I have figured out, as a person who wishes money would disappear, because I think it’s evil, I’ve learned how to do that. And it’s something I think anybody, if they are really willing to listen, and ask questions about, “Why do you need this? Why is it painful now?” And then if they’re like me, they need to go to finance and get coaching, maybe they don’t need to go get that coaching, just, like, put in that work, I really think anybody can do this.
Larry:
Interesting. As you talk about that, that combination of the product mindset you mentioned earlier, that pragmatic financial approach to it, you’re addressing pain, but it’s not, like, selling a SaaS product, or a consumer product-
Mara:
No.
Larry:
… something like that. It’s, like, what you consider … Especially if they’ve already engaged a consultant, you can go like, “Remember that $200,000 bill last year, we can do that with Mike, who makes $100,000 and we only need a quarter of his time,” or that’s the kind of argument you’re making. Right?
Mara:
Yes. And when I say pain, I don’t mean like … We eventually get to P&L pain, but we have to start thinking about Jevonian 19th Century political economic pain. What do you want to be doing that you can’t do, because you have to focus here instead? And if we are holistic like that, we can eventually, somewhat artificially, reduce that to the P&L, but that’s I think the step that good product thinkers do, but people who don’t have that discipline miss. And I think that is so often necessary to get to concrete and convincing statements about the P&L.
Larry:
I love that you ground that in 19th Century political economy.
Mara:
Of course, I do.
Larry:
But it makes perfect sense. That’s how you can connect the pain, and then you get to that crass old P&L thing that you need to-
Mara:
Yeah.
Larry:
… but that really matters to a lot of people.
Mara:
Yeah.
Larry:
One of the things that you’re doing, and with all of this, like, all of this rhetoric that you need to implement to get to funding yourself with funding knowledge engineering, which is really a fascinating part, or probably the most … As the host of the Knowledge Graph Insights Podcast-
Mara:
Yeah.
Larry:
… it’s the most fascinating part of this whole thing to me. Can you talk about, specifically, around … Because you’re selling the benefits of a knowledge-engineered program-
Mara:
Yes.
Larry:
Yeah. Can you talk a little bit how you position that, and approach that project?
Mara:
I can. Yes. I can. So, this is partly luck, the organizations I’ve landed in, and partly attraction to organizations that have allowed me to work in this way. So, I’m not a humanities scholar anymore, but I’m still, clearly, as you can see, a humanist with a big HCI, human-computer interaction, focus.
Mara:
So, when I work, I think a lot about the digitalization process, and I think about how the information systems we build, and indicate how we think about the phenomena surrounding us in the world. So, how we encode, what we encode, what kind of structures we impose, or neglect to impose often times, the modalities we choose, which can be anything from a vendor product to a programming language, and free open source.
Mara:
When we think about those kinds of things that we can’t necessarily induce good answers for, or even deduce good answers for, without observation and we have a business problem … This is an embarrassing one, but it’s a real one. Like, “Okay, I spent a lot of money curating my product data, but now I can’t get good metrics about the products I sold.” Happens that, although, it was not documented this way when investigated, the $2 million data set describes SKUs, not products, which are higher level.
Mara:
So, thinking about how we actually digitalize, not how we should, not the ideal data architecture way, or enterprise architecture, information architecture, any other science-ey way, but how humans actually do it is both key to good programs, and when it’s not done well, and this is very common that it’s not done well, a fantastic argument for knowledge engineering. Because what does that knowledge engineering fix, but those perceptual and assumption mismatches between the user and the person who is doing your design and engineering labor for you-
Larry:
I want to make sure I’m understanding. I think I get that mismatch you were just talking about, because as the SKUs, for people that don’t know, stock keeping units, it’s, like, that’s instances of the product. Right? Whereas the product-
Mara:
Right.
Larry:
Like, in an ontology, the product would be the class, and any one SKU would be an instance. Is that-
Mara:
Mm-hmm. Or think about it I’ve got a cup of coffee here. Actually, let me get my glass of water. This is even better. Okay. Water product, water in glass, SKU.
Larry:
Ah. I like that. I’ve used that analogy in content model. That’s a whole other thing. Yeah. Okay. That’s really interesting. And something else you just said in the middle of that story, that you reminded me was key to your talk at KGC, if I recall correctly, was that notion that this is all documentation, that-
Mara:
It is documentation.
Larry:
Like, you do. You talk about all this people stuff, which makes sense as a foundation of it, but then you have to document both the discoveries, and the decisions, and all that stuff.
Mara:
You do.
Larry:
Yeah.
Mara:
And I think a place I have shined as someone … Shone? As someone who has done knowledge engineering and now leads a lot of activities that rely on knowledge engineering is getting into spaces where there isn’t robust documentation, and, for me, the supply chain person, that’s the physical world. So, we’re talking legacy technology, or companies that even began as pre-digital.
Mara:
I’ve done a lot of interrogation of the physical world. Like, I’m currently working on a program seeking to optimize how we manage passenger compartments in aircraft, and what that’s meant for me is getting down on my hands and knees on two by fours, and in an Airbus to see how the aircraft is put together.
Mara:
And as people doing knowledge management, knowledge engineering work, it is crucial that we … If we can observe a phenomenon, whether that’s a human interaction, whether that’s a physical object that we go, and interrogate that thing. So, we don’t end up with the ideal world, we get from the data store, or the documentation, the schematic, or the narrative, that’s what ChatGPT can do, it can interrogate that stuff, that we actually look at the real phenomena. That that’s how we get to robust systems. And it’s not easy, but it’s really, really rewarding. And it forces us out of our comfort zones as the architect in the ivory tower for those of us who are more comfortable in that space.
Larry:
Yeah. And that gets at, I don’t know, one of these, certainly, core tools in any architect’s toolkit is the modeling skillset. And what you just said, like, sorting out SKUs from products, or … And understanding whether it’s an airline cabin, or a, I don’t know, financial instrument, or whatever it is, like-
Mara:
Yeah.
Larry:
… truly understanding that, so, you can model it correctly. Yeah. That’s-
Mara:
And this can be physical object, or it can be shadowing people. It can be going side-by-side with your colleagues in the business to see not what they’re telling you they’re doing, again, the self-report versus the actual behavior, but what they’re actually doing and then asking questions.
Mara:
So, “You’re supposed to be doing this. Why are you doing it this way?” And if you ask those questions, in a non-judgemental way, if you’re really curious, suddenly, you will understand why the organization, and its digitalization program, or lack thereof is shaped the way it is.
Larry:
Yeah. No. The way you just said that through that distinction between what people say they want, and their actual behavior, and inferring from that, what they actually want, that’s probably how Henry Ford came up with cars, not horses, and Steve Jobs, I thought all that kind of stuff.
Mara:
That’s anthropology.
Larry:
Yeah. Exact. Just applied anthropology.
Mara:
Yes.
Larry:
But that al comes back to the people stuff again, and-
Mara:
Yeah.
Larry:
Like, so, that combination of crawling around in the Airbus jet, and talking to the subject matter experts, and then that just as part of this whole process culminating in a knowledge model, or a data model to describe that knowledge, that’s … How many stakeholders are we talking about? What are the time scales? Just of a typical … I don’t know how you divvy up your work-
Mara:
So, me as an individual, or me with teams that I-
Larry:
For you as a team. Like, getting from that point of understanding the cabin to having a digital twin, whatever it is you’re trying to get to, tell me a little bit about that, the macro view of that time span.
Mara:
Oh, man. Like, most other things, it depends upon the kind of time, and resources, both people’s brainpower resources, and where possible and appropriate, automation, augmentation by automation, come in. And I want to be careful when talking about models, that we aren’t just thinking about the physical world, but we’re also thinking about procedure, and process that might not be documented. That’s the going and talking to the SMEs saying, “Why are you doing it this way?”
Mara:
When we have our picture of this stuff, and we have our picture, which we can also formally model of how humans interact over time with other humans, with digital artifacts, with physical artifacts, or with their own ideas, then we can start thinking critically about where can I automate? What kinds of technologies can I automate? So, I hear we’re bringing the cost down, we’re changing the time skill, and we also start understanding once the universe of known unknowns starts opening up, what kinds of people we have to bring into the program.
Mara:
So, I’ve done programs like this to support major ERP deployments that have involved hundreds of people, and I’ve done little programs like this to support a content management system that have involved five. So, it really depends on the problem space, and the willingness of the organization to invest, and the degree to which automation is not just possible, but a responsible choice. Again, depending upon the posture of the organization, and its relative conservatism. And, for me, I’ve worked in this. You can probably hear a lot of really conservative organizations in a safety-first way.
Larry:
No. And also the span of scale is really … From a small five person CMS project to a huge ERP thing-
Mara:
And, clearly, I’m not directing 100 people at once, but I am working with leadership, stakeholders, and key SMEs to delegate.
Larry:
Yeah. To delegate all that. Well, hey, Mara, I can’t believe it, but we’re coming up close to time already. But before we wrap up is there anything last? Anything you want to revisit from the conversation, or make sure that we share before we wrap up?
Mara:
For people who … I know I talked about this at KGC, but it’s really important. For those of us who are really committed to having truth, and beauty in our systems, don’t get shut down, because you need to justify your activities to finance. You can figure out how to do that. And also don’t believe that the real world is worse, or uglier than the design you have in your head, the …
Mara:
I know nothing about Japanese philosophy, but 20 years ago, I visited a friend in Japan who took me all around, and I was struck by the prevalence of asymmetry and decay in so much Japanese art. And she explained this to me as a concept of wabi-sabi, or imperfection in beauty.
Mara:
And I think often times that idiosyncratic systems, that truly reflect the people they are serving, because, remember, the technology comes from the word for tool. These are things that serve us, and tools that look like us, that reflect us, are, ultimately, the most beautiful. There’s my humanism again. So, that’s my parting shot. Don’t be afraid if you have to justify, justify the money. You can do that. And don’t be so attached to an ideal outcome that you miss the best possible.
Larry:
That’s interesting. I want to do a workshop on wabi-sabi now, and really get my head all the way around that.
Mara:
You have to call an art historian.
Larry:
Yeah. That’s right. Hey. One very last thing, Mara, if folks want to follow you, or connect online, what’s the best place to find you and connect?
Mara:
LinkedIn. And can I share my contact details with you and show them?
Larry:
Yeah.
Mara:
Or my personal email, you can find me at M-A-R-A dot I-N-G-L-E-Z-A-K-I-S dot O-W-E-N-S at Gmail dot com. I’m not a social media person, but I love people, and if anyone hearing this would like to follow up and talk about this stuff, here I am.
Larry:
Excellent. I’ll put those in the show notes as well.
Mara:
Okay.
Larry:
Well, thank you so much, Mara. It’s always fun to talk to you, but this was a particularly awesome conversation. So, thank you.
Mara:
Thank you, Larry. The pleasure is mine. Thank you so much for interviewing me today.
