This episode features an interview with Biba Helou, the Senior Vice President of Enterprise Data at Capital One. In this episode, Biba talks about the agility the cloud provides, how data can help personalize customers’ experiences, and much more.
This episode features an interview with Biba Helou, the Senior Vice President of Enterprise Data at Capital One.
In this episode, Biba talks about the agility the cloud provides, how data can help personalize customers’ experiences, and much more.
--------
How you approach data will define what’s possible for your organization. Data engineers, data scientists, application developers, and a host of other data professionals who depend on the Snowflake Data Cloud continue to thrive thanks to a decade of technology breakthroughs. But that journey is only the beginning.
Attend Snowflake Summit 2023 in Las Vegas June 26-29 to learn how to access, build, and monetize data, tools, models, and applications in ways that were previously unimaginable. Enable seamless alignment and collaboration across these crucial functions in the Data Cloud to transform nearly every aspect of your organization.
Learn more and register at www.snowflake.com/summit
[00:00:00] Steve Hamm: good morning, Bebo.
Biba Helou: Good morning.
Steve Hamm: It's great to have you here. You know, our listeners are probably aware of the capital. One is one of the top credit card issuers in the country, but it would be great if you could describe the other main dimensions of the company's business.
Biba Helou: Sure. So, uh, capital one is , uh, one of the top 10 banks and we offer checking and savings products for consumers, small businesses and commercial companies. Uh, what. Most people probably don't know is that we've also built a massive in house tech workforce of nearly 11,000 people and 85% of those are engineers.
So, uh, these teams , um, build fantastic software products, leveraging machine learning, AI, and data to solve. Important problems in banking and to help people manage their finances. So we also consider ourselves to be a [00:01:00] big tech company as well. And , uh, as a leading bank , uh, digital bank , um, machine learning and AI are really central to how we build our products.
Steve Hamm: yeah, it's interesting years ago, I lived next to a guy who was a vice-president at JP Morgan chase. And he told me during one of our talks out on the lawn, the JP Morgan chase was a tech company that happened to have a banking license. And so this was 15 years ago. So already banks were really seeing themselves as tech companies and tech innovator.
So I think that's one of the little known, but very important trends in banking and financial services these days.
Biba Helou: Yup. That's exactly how we feel.
Steve Hamm: Yeah, no, the industry is still, I would say in the midst of a big long-term transition, I mean really financial services. Um, you know, there's been a lot that's happened because of the internet. There's been a lot that's happened because of mobile computing. Um, what would you say is the transition [00:02:00] now what's going on and, and how does data help address these, you know, the changes or adapting to the change?
Biba Helou: Well, as you said, you know, most people are going digital in their everyday lives. They use their devices , uh, to make decisions on a minute by minute basis. Uh, seems like technology is very integrated into everything that they do. And because , uh, there's so much , um, power out there , um, in computing and the ability to process data, we can continue to go down the past and personalizing all of those interactions.
Uh, and get closer and closer to our end customers and our consumers. So, you know, technology has changed the way we communicate, how we shop, how we travel, how we manage health and money. And so that, that digital transformation, that digital journey. I just keeps changing the way that we live our lives day by day, we saw that in [00:03:00] academic too , uh, you know, where people started relying on your technology even more.
So I've just seen that change year over year , uh, at , uh, an incredibly fast pace.
Steve Hamm: Yeah, it's just interesting. Even thinking about currency and coins. I realized recently looking back that I haven't touched a dollar bill or a coin in more than a year, and that's a huge change. And I'm sure for a lot of people that is happening and, you know, reckoning with that, even reckoning with that isn't of, of the big, the big change that banks have to make.
Biba Helou: Yeah, that's very true. I mean, that, I would say you're right statement. Like even a year ago, I always felt like I had to have cash in my wallet. Uh, just in case I don't even think, I think about cash anymore or having cash in my wallet because of that, just in case seems to have changed.
Steve Hamm: What's your role now? And what was your career path that took you there?
Biba Helou: Role now is , uh, uh, SVP of enterprise data. So I'm responsible for the enterprise data systems supporting capital one. So think. Data pipelines. Real-time streaming platforms , uh, data lakes, data warehouses, et cetera. I've been at capital one for nine years and have had a multitude of positions within capital one.
Uh, and prior , uh, to capital one have been in a number of different , uh, areas as well. Mostly technology, mostly I would call set , uh, centralized technology. Although I have worked in different lines of business , uh, developing application architectures as well, but , uh, most significantly in the last nine years, very focused on cloud cloud services.
Uh, and ultimately obviously leading to data.
Steve Hamm: now, have you been in banking almost your entire career or is that just [00:05:00] the last night?
Biba Helou: So I have been in a few financial services companies, the world bank, which is not quite your traditional bank , uh, Fannie Mae, which is , uh, the mortgage , uh, uh, consolidator and securitization company. But I've also worked for , uh, Nortel networks and , uh, a startup that , uh, specialized in data storage as well.
So done a little bit of everything.
Steve Hamm: no, that's great. So in your career you've been a manager and a leader. In our organizations that have been dealing with these transitions, these huge technology transitions and, and the cloud transition and the cloud data transition seemed like some of the biggest ones, one of the most important leadership or management lessons that you've learned along the way.
Biba Helou: Yeah , I, I , uh, tend to gravitate towards big , uh, transformational or transitional , uh, efforts. Uh, and I love doing that. What I've found is that , uh, when you're leading people and when you're leading [00:06:00] big efforts , um, people need context. So that's one thing that I've learned along the way. It's not enough just to set a North star, but they really want to understand the why and the benefits from every angle, including the benefits to the people who are actually doing the work.
So that's one big lesson that I learned is. That you really need to spend time providing context. And then also adaptability is important and making sure that the people that work with you and for you, all your stakeholders understand. That while you have targets, you need to be open to the ever changing nature of the world and technology and continually factor in that evolution as you're moving along your journey.
Steve Hamm: so part of your job is to kind of condition the people around you to deal with change and transformation.
[00:07:00] Biba Helou: Yes. I think that's probably the most important part of the job.
Steve Hamm: Yeah. That's interesting. Now you've been at capital one for nine years. And that's, I think that's exactly the period of time that the company has really been on this major, move to the cloud, a major move data to the cloud. If you could kind of look back on that period and talk about , well, why did that start and what has gone on w w what, what took you to the place you are now?
Biba Helou: Yeah, the primary driver for our moving to the cloud in general was agility and giving our software engineers and our business partners. The ability. To innovate at the pace that they wanted to. So we, we have the traditional enterprise data centers where you are limited by physical space, right? So you couldn't quite grow as quickly as you wanted to.
And just the physical bounds of adding new services, new systems, new technologies, it [00:08:00] just didn't quite go as fast as we would have liked. So we started our, our journey , uh, by. Building our engineering workforce. Uh, we became a , uh, agile shop, a hundred percent agile shop to make sure that our software engineers were developing in ways that allowed them to iterate and provide value of along the way, rather than waiting until the end of a big project.
And then it just turned into the cloud public cloud public cloud service providers. They were able to also innovate out of case that we could leverage , uh, new services that allowed our, of our engineers to build better and better products. So, uh, we just needed that modern infrastructure. We needed to be able to build quickly and we needed to be able to leverage new services as they were being made available.
Steve Hamm: So you, a lot of people, I'm sure that Capitol [00:09:00] one moved the computing to the cloud and the storage to the cloud, but really data management store, you know, storage of a massive amount of data , the, the whole process of integration and, and , uh, and management. And then finally analysis. Has that, is that something that's still in the early stages or have you really done the migration already?
Biba Helou: I think it's, it's an ever evolving thing that we're doing. So we have moved our data to the cloud. We have continually evolved our data management practices and our integration practices. So I would say we've done quite a bit, but we also know there's it's an ever , uh, ever evolving transformation. Uh, there's still much more to do just because of the sheer amount of data and the different uses for data that are being identified on a daily basis.
Steve Hamm: now you've talked a lot about agility, the focus on the engineering [00:10:00] and building new applications fast. Could you talk about a couple of examples of how using the data cloud has really made a difference in the bank operates or its ability to get quickly to market with a new service or something like that?
Biba Helou: Yeah. So, um, you know, you, you see in a lot of different products , uh, How , uh, people use , uh, data for personalization. So you can take a look at our , um, our, our, our products and, and. You can see how the information that you experience , uh, when looking at capital one accounts, for example, is very personalized and lets you get to the action that you want to take very quickly.
Uh, we needed , uh, the cloud data platform so that we could tackle , um, a lot of the, the changes that we needed to make on a real-time basis in order to keep up with our, our , uh, Customer needs and our customer behaviors. So we, we are leveraging the cloud to run several complex models, simultaneously making sure that we're protecting our customer [00:11:00] data, but allowing our customers to do what they need to do in the moment, leveraging the data in the cloud, the data cloud, essentially.
Steve Hamm: and one of the core activities of your bank is being able to deliver personalized, unique experiences for customers, whether they're. Individual customers are small businesses, especially. I think that would be the focus. Can you talk about , um, you know, how the data cloud and data analytics have really helped you deliver those kinds of experiences?
I mean, maybe to the, you know, the handheld device to the iPhone or something like that.
Biba Helou: Sure. Yeah. We rely on machine learning and other advanced capabilities to guide investment decisions. Uh, we, we use it to detect fraud and make sure that we're providing more intelligent products to our customers so they can manage their finances. Uh, we, we can account for events in the world, such as the pandemic, for example, to make sure that our decisions are, are accurate , uh, and beneficial to our customers.
And we [00:12:00] continually use a test and learn philosophy using the data that we have so that we can meet the objectives that our business , uh, put on us as well.
Steve Hamm: yeah. Yeah, no, I understand the capital one was one of the first large banks to embrace no flakes technology. And in fact, lever was even a partner in developing that technology and making sure that it was enterprise ready. If you could talk about the partnership, how it developed and, and what you've achieved through that partnership, that would be great.
Biba Helou: Yeah. Yeah. As we mentioned, Steve, wait, we were one of the early adopters of snowflake and we really rely on the elastic capabilities. Uh, to be able to , uh, leverage , uh, data as much data as we need and as much compute as we need and to be able to scale both independently , um, it lets us store and access , uh, larger, more, very data sets.
And we , um, have [00:13:00] multiple , uh, users of that data working in those systems at the same time. And it's helped us , uh, train machine learning models for experience, to such as Eno, which is our intelligent assistant. And it uses AI driven insights to notify customers about their unusual transactions. We have , uh, really appreciated the partnership with snowflake.
We have a lot of requirements for not only scale and use concurrent use by all of our different data, engineers and analysts, but we also have a lot of. Needs for reliability, resiliency. Uh, we have, I have a lot of governance requirements as well to make sure that we're protecting our customer data. And the snowflake partnership , uh, has really worked where snowflake has incorporated a lot of our needs into their roadmap and has really , uh, snowflake has really worked with capital one to [00:14:00] understand what a big complex enterprise needs and have.
Built a lot of capabilities into the product.
Steve Hamm: no, that sounds like a very good partnership. I'm sure there've been bumps. Along the way though. Can you, can you talk about any of the challenges that you've encountered moving to the cloud?
Biba Helou: sure, you know, over time organizations tend to accumulate a lot of data. And what I always say is the cloud is great because you can do anything that you want to do, but. The problem with the cloud is also, you can do anything that you want to do. So just making sure that , uh, you know, you're keeping a track of all of the different activities by the software engineers and the data scientists , um, is, is something that we've had to learn how to do in the cloud, in the data cloud, any cloud.
Uh, and we continue to work through how you manage that. At a level that continues to meet the agility needs that we have, but , uh, creating the right governance [00:15:00] processes around it. So, and then what, obviously the other , uh, challenges that anybody has with their cloud services, making sure that we are watching , uh, Out for waste and sprawl and making sure that we continually manage so we can do more with our money as opposed to redundant , uh, things with our money.
Steve Hamm: It makes total sense. Now I know that one of the features of the data cloud. Is the ability to share data both internally within an organization, through the different units that might in the past have been siloed, but also with external partners , um, and suppliers and, and, and distribution partners or, or whatever kind.
So it seems like that opens up a big new area for data. And I wondered, how are you leveraging the data cloud to share data more robustly?
Biba Helou: So internally , uh, we are able to make [00:16:00] data available to more users without having to make copies. Of data or move data around. So I would say , uh, just creating the visibility into the data that we have between our groups internally and being allowed to provide the access that they need, obviously within governed processes, without having to tap shoulders and find out what data is and where it is and how to get access to it has been a really significant.
And , uh, um, uh, in, within the company, I would say externally, we are really scratching the surface of how we leverage data that is being made available in the marketplace without having to move it around. So, uh, very interested in the snowflake data marketplace that is being stood up and , uh, and experimenting there.
Steve Hamm: Yeah. So what I mean is, is the, is the [00:17:00] opportunity with sharing with external and through the marketplace? Is that largely with. With retailers and people like that or who who's the partner that you're, that you're sharing data with. And, and what power does that provide?
Biba Helou: It can be any number of partners that you share data with or, or they share data with you, obviously , uh, data that helps augment our analytics is called shared data where you're getting data from external sources. You can think of , uh, you know, obviously in a credit card , uh, In the credit card business, you'll leverage credit Bureau data.
So you make that as a, an example of sharing data. We've used the data marketplace for the, a star schema data , uh, about COVID-19 to help us. Uh, run some scenarios to make sure that we could take care of our workforce and our customers. So that's just scratches the surface, but there are so many [00:18:00] opportunities for capital one to be able to use data, to help make decisions for our workforce and our customers in a much faster way.
Steve Hamm: No, the data marketplace, I mean, one could have separate kind of sharing relationships. One-on-one with a lot of different sources. Is w what's the advantage of the data marketplace?
Biba Helou: To me, the advantage of the data marketplaces, just being able to see the vast availability of data that's out there that could be leveraged and it's all, it also prevents the need to move data around and make multiple copies of data. So done. Right. You can have access to data. You can bring compute to data without having to make copies of it.
And not only , uh, not only , uh, pay for the data that you're using, but also you prevent having to create a whole new infrastructure just for getting access to that data.
Steve Hamm: Yeah. [00:19:00] Yeah. I remember the old days of data sharing where you actually had to often you'd make a hard drive and have somebody delivered across town. It's been a few years, but I think some organizations are probably still doing that, but
Biba Helou: Oh, yeah. I mean, sending data , uh, fire traditional final transfer D uh, is, is still very, a very big part of most companies right now.
Steve Hamm: I know. Yeah. In some ways, in terms of, when you think about a data marketplace, we're at the very beginning of that transformation, correct?
Biba Helou: that's right. Yeah. I would say that is very true.
Steve Hamm: Yeah, no. We've talked about how quickly things are changing in our, in our economy and the business world and competition and in technology. So when you look ahead, say this year or next, what are the most important trends in cloud computing and data analytics that you see coming.
Biba Helou: uh, there are so many, so I'm just going to concentrate on a few, obviously. Um, uh, because there are so many [00:20:00] trends and I'm keeping an eye on all of them, but I think one big one is , uh, connected , uh, data and the automated insights. So there is. So much data out there that's in different systems, in different companies, different lines of business within the company, but just making sure that you can connect all of that data and, and make sure that you are gleaning insights from that data.
Uh, continuously is one big trend. Uh, you might hear it referred to as a data fabric or a data mash, but just , uh, going more and more down the, the path of being able to create those data fabrics and making sure that we can connect data and connect , uh, uh, um, how, how we use that data. I think that's one big trend that you'll
Steve Hamm: Before you depart from that I to understand a little bit better. So the mesh, how does the mesh enable automated analytics?
Biba Helou: It lets you connect data that you might not have [00:21:00] thought was related necessarily by making sure that you understand what's out there. Uh, and, and how that data is related. And then it just allows you to leverage it more real time. I think that's really the way that I would look at it. Uh, just to give you insights about the data itself and then being able to use it for any other use cases that you need to use.
Steve Hamm: yeah. Sorry. I interrupted you there, but
Biba Helou: That's
Steve Hamm: ahead. What was the second thing?
Biba Helou: And then I think the other , um, the other area is just being able to use more types of data. And again, connecting that data. So both structured and unstructured data. So traditionally we tend to think of structured data and then you have keys that you join, and then you can put data together to provide insights, but.
We want to be able to , um, look at the world, wants to be able to look at , uh, not only that structure data, [00:22:00] but also marry that with things like audio and video and do natural language processing and figuring out how to connect that information with what is traditionally structured data. And create insights from that , uh, make decisions based on that data.
I've heard it , uh, I think I've heard it called ex ex uh, analytics. Um, and I think it just means , uh, really just using , uh, a wide variety of types of data to be able to make decisions. So I think what we'll see a lot more , um, with that trend as well.
Steve Hamm: I know that smell like in the past month or six months has added a lot of new data types or is adding a new, a lot of new data types. And also also making it easier for people to program with a bunch of different languages, kind of the language of their choice. Are these, these are the kinds of things that really help your organization innovate.
Talk through [00:23:00] that. I mean, I have noticed that capital one is racked up kind of the most patents among banking industry, the banking industry for the past year, a second year in a row, actually. So was the focus of your innovations and how does it, how do you depend on vendors like snowflake to enable those.
Biba Helou: I think just having data available creates an environment where our business partners and our engineers can just. Think about the possibilities. So it's, it's sometimes hard to imagine what's possible without , uh, without knowing what's really out there that you could use and just having access to and seeing what type of information is out there has really opened up , uh, the innovative spirit , uh, within our company.
So we , um, identify new and innovative products and experiences. That helped us , uh, [00:24:00] create , uh, Tech centric banking solutions for our customers. And , uh, the growth in our portfolio is really generated by harvesting those innovations , um, from our associates across the company. And the more and more that they take a look at what's information, what information is out there, they're leveraging technologies such as artificial intelligence, machine learning , um, and, and applying that to , uh, Business problems, customer problems that they're trying to solve.
And out of that comes really great , uh, products that are patentable.
Steve Hamm: Yeah, it sounds like your company and your, your group within it. You know, rely on outside vendors for basic functionality. I mean, for, for these big infrastructure things and these, you know, the data management, things like that, but you, but you, you, you create your own applications. You're not relying on , uh, some outside application company to do the analytics for you.
We're bringing the AI to it. It's like, that's what you're cooking up yourselves. Correct.
[00:25:00] Biba Helou: Yes, that's right. We cook up , um, T to use your term , our, our customer applications, ourselves leveraging a lot of tools that external providers add. I will also add that we are also very innovative in. Creating a management tools ourselves, where there are gaps in the industry. So we're not always just focused on patenting financial products.
We do , uh, also delve into , uh, tech products as well, where we see a need for something that , uh, enables an enterprise such as ours to operate effectively.
Steve Hamm: a gap rather than waiting for somebody to fill it, you fill it yourself.
Biba Helou: That's right. That's right.
Steve Hamm: makes a lot of sense. So I want to ask you to put on your visionary cap now for a minute and look out, say five years or more, how do you see the data cloud impacting business and also society?
[00:26:00] Biba Helou: I think it's just an acceleration of what's already happening. I mean, just the continued ability to see and visualize things in the moment , uh, allowing , um, decisions to be made much faster. I anticipate , uh, that data. Um, predictions will be made with , uh, with, with the ability to just , uh, take a lot of data information and make predictions in the moment, rather than waiting for lots of analysts analytics, looking at trends over time, and then making decisions , uh, in , uh, A little bit of a longer fashion.
So just immediate recommendations, immediate decisions being made that just allows the world to evolve much more quickly.
Steve Hamm: So that suggests that you see the user of these technologies on the data being , not, not just the data scientist or data engineer or some analysis analyst [00:27:00] within a business unit, but really. Uh, the operational people, whether they be executives or people working in a branch office or something like that, do you see a lot of , uh, capabilities coming to the hands of kind of non-experts or non-tech.
Biba Helou: I do I do. I actually see data science becoming much more of a data science as a service business intelligence as a service , uh, uh, over time and the consumer or the user of the data where as you said, the operator is going to be able to use these tools that are in hand to make. Decisions , uh, rather than something that was very centralized or continues to be centralized data is, is really complex.
Right. And data sciences is amazingly complex. That's why we have so many of , uh, people out there doing this analytics with PhDs, but I just think more and more products. Are going to be con self-service and give the end users the [00:28:00] ability to do what typically those data scientists , uh, had to spend a lot of time on doing centrally.
Steve Hamm: uh, you know, the last question we always try to, we ask one that's real kind of personal and a little softer. So I noticed that your family name is Lebanese. And I just curious, when did your family immigrate and how has your middle Eastern heritage affected your life and career?
Biba Helou: Uh, my, yes, I am Lebanese. I was born in Beirut actually. So I'm both. My parents were born and raised in Lebanon as well. Uh, my, my family , uh, Immigrated in the seventies based on the unrest and the civil war in Lebanon. Uh, and I was obviously very young. I mean, I grew up outside of Lebanon. We left , uh, Lebanon when I was just a couple of months old, but , um, w you know, the way that it's impacted me, my family is very adaptable.
So my entire family essentially. Left. That means cousins, aunts, uncles, grandparents , uh, left [00:29:00] and are all over the world. And it, it just taught me, even though I grew up mostly in the United States, living with my parents just taught me the importance of. Knowledge information, education, and being curious and really understanding what is around you so that you can adapt and innovate and change the way you go about things.
Because the circumstances around you may or may not be in your control, but the way that you react. To those circumstances. And what you do with that information is something that is, was just ingrained with me. It's very important and you need to be able to adapt. So I would say from that life experience and growing up in my family with my family all over the world and seeing how everybody has pretty much done a really great job adapting to their different environments.
I think that spilled over [00:30:00] into my work life, which. Also demonstrates in my career, my ability to adapt, to changing trends, to continually want to be on those edge of those transformations that have been occurring throughout my career. So I, I attribute that very much to my Lebanese heritage.
Steve Hamm: Yeah, that's interesting. You know, at the top of the conversation, we talked about your kind of management and leadership lessons. They're the same. I mean, you, you talked about the things that you tell and you teach to the people on your teams. That's very much the approach that you've taken to life because of your immigrant experience.
So that's really very interesting.
Biba Helou: Yep. And it teaches you, the policy possibilities are endless. Uh, I am very proud to say that I'm a very , uh, very yeah. Successful family. And I think it's all , uh, through learning those lessons for sure.
Steve Hamm: Yeah , well, this has been a delightful conversation. Um, really interesting stuff , uh, about managing data, [00:31:00] about analyzing data, about running a bank on data. And I think one of the, the, the coolest lessons from this conversation really is that that businesses have to see themselves. Uh, partly as technology companies, not just as consumers of technologies, but as innovators in technology and transformers of their industry, because it seems like if you don't transform your own industry, you might be, you know, bulldozed by the transformations that come.
So I, I see that as the, that's the big message today.
Biba Helou: Yeah, that's right, Steve. And I think capital one definitely embodies exactly what you just said.
Steve Hamm: Oh , well, great. Well, thank you so much.
Biba Helou: Thank you.