The Next Horizon

Galloping Toward Digital Transformation

Episode Summary

Achieving digital transformation is ultimately about asking the right questions to determine which business processes you want to tackle first to achieve optimal outcomes.

Episode Notes

Digital transformation is data transformation, and right now, most businesses have too much data coming at them and not enough of an understanding of what to do with it.

Join Jon this week as he meets with Matt Baker, Senior Vice President of Strategy and Planning at Dell Technologies, to discuss best practices for starting or continuing the journey to becoming a data-fueled business. You’ll learn why delivering real-time digital experiences requires a new approach to data management, why having a focused digital transformation strategy will help you stay ahead of the pace of innovation, and ultimately why there is no single technology that can solve all your problems.


What you’ll hear in this episode:

Using data to create a real-time digital experience (01:19)

Data scientist is to technology as a doctor is to patients (02:13)

Are we really experiencing a big data problem? (04:20)

How to create a systemic approach to data processing and data management (05:05)

Why all your data cannot exist in one place (06:51) 

Using data pipelines to feed complex, distributed systems (08:58)

Does digital transformation play a role in a company’s resiliency in the face of global crises? (10:33)

It is never too late to start digitally transforming (11:55)

Is it possible to stay ahead of the pace of innovation? (13:15)

Partnering with a company that has a breadth of capabilities is the right choice for evolving to become a work-from-anywhere company (16:15)

Lessons in supply chain operations (17:30)

What NOT to do to solve your data problems (19:01)

Is there one technology you should prioritize over others? (hint: no) (20:37)

Episode Transcription

Jon Hyde: Hello and welcome back to The Next Horizon, a Dell Technologies podcast. I'm Jon Hyde and together we'll explore the implications of several major emerging technologies for business, society and most importantly, for you.

Jon Hyde: Today, we're joined by Matt Baker, senior vice president of strategy and planning for Dell Technologies. Matt, thanks so much for the time. It's a real pleasure to have you on the show.

Matt Baker: Well, it's awesome to be here. I'm really looking forward to the discussion.

Jon Hyde: Yeah, it seems like the discussion that we always want to have, we have in different ways, it's objectively about data and all the things that come along with that. And data, it's always there. It's always a big part of the problem, but it's also is a big part of the solution. And I think that really encapsulates a lot of what we as a company, try and get across to our customers.

Matt Baker: Well, I think it's absolutely true and sort of data is either a source of value or a source of potential loss or it could be reputational loss. A lot of what we talk about is sort of balancing investments in protecting your data and protecting the integrity of your operations. That's something you have to do. It's sort of the insurance policy, but the exciting thing is what you can do with data to transform your business. And when we say digitally transform businesses, they're really data fueled businesses and data fueled digital business processes. Things that are automated by the use of data to deliver a new type of real time digital experience for a customer or a process in a factory or anything.

Jon Hyde: Yeah. We hear a lot of customers that come to us and say, "We need to understand our data better." And my first question is always, "Why? What are you trying to do with it? It's because it's not just because you want this insight, it's because you want to do something with it." There's something that follows that.

Matt Baker: Absolutely. And the interesting thing is sometimes there are a set of customers who are, I don't know what my data is and therefore, I don't know what questions to ask. There's another class of folks who know what questions they want to ask, but what they're faced with is just this giant scale problem. And the technologies of today for storing data, don't do a good job of cataloging that data, assigning some sort of a value score and believability score, so on and so forth. And so a lot of modern data management is about enabling a data scientist, the person who understands what the question is and just simply wants to have access to the data that's relevant for the question and therefore to feed into an algorithm. And today data scientists spend more than 80% of their time just trying to find data and build a pipeline for that data.

Matt Baker: And the way that I described it is, a data scientist is sort of this super professional guy in a white coat. He's the doctor, but what they're doing and not to be demeaning to digging ditches, it's important work, but we're asking the data scientist who has a skillset that's really all about building algorithms, but we're asking them to do a bunch of things in the trades where there are better people who are much better at doing that. They're having to do this. Each question is another ditch dug versus let's get an engineer and engineer a system so you no longer have to go out there. You just turn on your faucet and the data you want arrives. That's the real challenge in today's sort of world of digital transformation is that there's a sea of data. There's a sea of data. A lot of it is salty and therefore we can't drink it so we've got to refine it into something that's a resource that won't kill us if we drink it but instead nourishes our lives or our businesses.

Jon Hyde: Yeah. Small scale data is one thing but when you start to achieve scale of data, which everybody's hitting this deluge of data, it's just inefficient. The only way to really overcome that is to extend IT back out into the operational world, into what we call the physical world. How do we do that? What are the challenges that a customer face there and what is Dell really doing to help change that?

Matt Baker: What's really fascinating to me is what you just said, which is we have this big data problem. The reality is that we have a big data problem, but we have too much data coming at us and we're not able to understand what portion of it's important. If we knew what portion of it was important, then we could dispose of everything except the unique, let's say excursions, if a sensor goes from green to red. We want to know that snapshot, maybe 10 minutes between it going from green to red and then we need, that's what you want to save. And then you can send that back to a central location and train an algorithm against a big set of historical data. But when you get into the inferencing, the processing of it, you've got the sea of data coming at you and what you don't want to do is send it all back to a central location.

Matt Baker: You want to process it locally and run the algorithm. When there are exceptions, you ship them back and you create this tight inner loop of inferencing. You have this sort of near real time loop for other optimizations that you might do. And then you have this big non real time loop, which is where you do the training of your machine learning or AI algorithms. You need to do all three of those so it's not like something's going to occur only at the edge and then that means the core is no longer relevant. What you're going to create is a systemic approach to data processing and data management. And you have to understand that in this world, we're going to have different loops, different locations. Those locations are going to be challenged in ways. It's an end to end view of system architecture, if you will, that we're approaching both the data problem and the edge problem, not as two separate issues, but as issues that are combined together.

Matt Baker: And that frankly pertains to the cloud as well. You can't really separate these various phenomena from one another. You have to think about them in the context of there's processing and there's data storage. And so you have to pull them together in a meaningful way and optimize them for the type of work that they're going to do. There is no edge without data. There is no cloud without data. And unfortunately we've spent I think a few decades really focused on sort of the operational side of infrastructure. I would say that cloud is really about infrastructure operations and a way of doing infrastructure operations. But nowhere in the cloud conversation to date has there been really a strong attempt made at data management. That's what we're tackling. But today the only story that most have is come to my cloud because all your data will be in one place.

Matt Baker: And I just explained why that's impossible. You can't have all your data in one place because of what everyone ultimately is going to try to achieve, which is these real time digital experiences out in the world, which means we are absolutely sort of coming from a point of extreme contraction, into a centralized approach to swinging back as the pendulum always swings, to highly decentralized systems. And we need to have operational platforms like the Dell Technologies cloud platform that allows you to manage edge, to core, to cloud. And then we need to bring storage and data management solutions that allow you then to manage data in that same context. There's data operations and there's IT operations and we sort of have to marry the two up.

Jon Hyde: Yeah. That paints a really clear picture of what we're trying to achieve, which I think at the end of the day is you talked about it in an interesting way. It sounds like a data pipeline design conversation to think about how we optimize, classify and deliver data and then how you deal with what stays extended at the edge versus what you might bring back to a centralized cloud for the different types of things like inferencing and training.

Matt Baker: Yeah. And I think you point on a really important phrase that many in the industry aren't necessarily familiar with, which is data pipelines. There've been data pipelines around forever, but most often they were batch oriented. And in fact, they were called something very specific, they were ETL tools, extract, transform and load tools. And they were done in a batch process. And so every night or whatever, data would be moved from distributed systems into a centralized, usually data warehouse and then people would do look back analysis on that data. That world is still important by the way, but it's not at all what the future looks like on the whole. On the whole, the future are these what we call data pipelines and they tend to operate in more real time and they tend to feed these complex distributed systems.

Matt Baker: You aren't going to have a fully optimized transport system until you're able to link up a bunch of cars into a network and have them communicating with one another, but also influencing onboard so you don't crash into one another. Which means you're going to have to have both the in vehicle, the far edge and then you're going to have to have some sort of, let's say metro level management of that data to create an outcome. And then you're going to have a ton of data that you then look at centrally to optimize the algorithms that both the car uses to operate individually and the transportation system uses sort of at the metro level to optimize the flow of traffic.

Matt Baker: That pipeline, and it's not just a single pipeline, it's many pipelines that are feeding data across the different processes in a broad distributed system. And that's why it's so exciting, but it's also so daunting because building and operating large scale distributed systems is difficult. It's one reason why people have centralized over time, but we're at a point in time where centralization just is no longer tenable and on whole. And therefore we're swinging back to large scale decentralized systems.

Jon Hyde: All of this points to the importance of data being the real potential to help companies unlock new insights and opportunities and really move the needle forward. And I think this year, 2020 has been this year of unprecedented change that everyone just realized, wow, we need to change the way that we do things. I think we found some companies that were in a really good position to change things and some who were at a disadvantage to change things.

Matt Baker: Yeah. And I think that the digital transformation index that we've been mapping for, I think the past four years or so.

Jon Hyde: Six years, yeah.

Matt Baker: Six years, there you go. I think is a really important element of helping us and our customers understand where they stand in this notion of digital transformation. And again, remember digital transformation to me is data transformation. There are a set of customers that self identify as digital leaders and those digital leaders, they've executed more swiftly on their digital transformations. They've created these strong data fueled business processes that have helped them weather this COVID storm much better. And so they're actually investing and growing in this environment. There's a set of digital laggards who find themselves in dire straits. And we're certainly jumping into help. It's never too late for folks to start building out and getting started.

Matt Baker: I love that cartoon that that said something about, "Hey, when are we going to get to our digital transformation initiative?: And then there was this wrecking ball coming towards them that said COVID-19. And the answer was is that you're going to get to it now. And what we found is that in fact, people, as you said, who have gone through the journey, gone and done the hard work, are seeing benefits in this unprecedented time. And the folks that didn't, are hurting.

Matt Baker: In reality that the middle is the majority and there sort of, it's a bit of a mixed bag. And they're the ones who are really looking and saying, "All right, we need to make some investments that are just sort of no brainer tactical decisions about how to keep the business moving in a highly disrupted time." But they're also saying, "All right, what elements of our digital transformation should we accelerate? Because those actually will help us get out of this fixed we're in." And it's really fascinating to watch the mix of people who self identify in these different ways and then watch through time as their experiences change. And unfortunately, the ones that have invested are the ones that have definitely weathered the storm better.

Jon Hyde: How do you keep ahead of the pace of innovation? Because it's moving really fast. And we've seen a lot of information come in from the index, as well as some customer conversations recently talking about the pace of change and the fact that IT in a lot of our customers are having to massively accelerate their processes of what they're trying to deliver. In some cases, what they considered to be a two year project is now becoming a six month project and the pressures that that puts on them, but also the pressures that puts on companies like Dell to help them deliver at that speed.

Matt Baker: Yeah. Well, the main thing that I think we're all facing, nobody's immune from this in a time of disruption like this, which is you need to decide what's truly important. And then you need to have clarity on what it is that you're trying to achieve from a business standpoint. All too many times, people think that, hey, digital transformation is about adopting a technology and fit your latest buzzword in. I need a cloud in order to be digitally transformed. It used to be, I need to do Hadoop to do digital transformation. It's the wrong way to think about it. It's what business or organizational problem am I trying to resolve? And then what technologies can I apply to that to solve that problem? And frankly, that's where we really come in to help our customers, which is, okay, you've, you've decided what your business priorities are. Good, because you're not going to get to all of them. Let's really focus in on the top three items that you've identified and then we'll help you match up the technology to solve those problems to create the digital initiative to fuel your digital transformation.

Matt Baker: Helping customers really winnow back what they're trying to achieve and hone in on the things that are going to truly move the needle. Once we've gotten there, what technology should we apply? And how can we do it in a way that takes that two month project and accelerates it? And that might be, we need to shift investment from one area to another. We might need to do some Herculean tasks on the Dell side to deliver certain systems in 14 days instead of two months. There's all sorts. Every situation is different, but I feel proud of the organization, particularly in this unprecedented time, the impact that we've had on customers to really help them move from a position of being in real peril, to being at least stabilized, in some cases thriving.

Jon Hyde: It makes me proud to be an employee here at Dell as well just thinking about all of the positive impact that we've had on customers, not just through the course of the pandemic, but leading up to what we didn't understand at the time was an unprecedented event. Yet we took the time to be leaders in those places, to put our customers in a good position to be successful in the future, regardless of what that was going to be. And I think it's an interesting thing for us to consider when we look at our customers and we think about how they need to be successful in the future, what is it that really sets us apart in this idea of an ecosystem strategy approach and how we start to drive this forward with our customers?

Matt Baker: Yeah, well I think there's a number of things that make Dell Technologies stand out. Well, first of all, we are the world's largest IT infrastructure company on Earth, IT solutions company on Earth. And we do that across end user devices into the data center, into the cloud, et cetera. And it was very interesting because at a moment's notice, we were in a position where yeah, the PC business was doing well, but then all of a sudden someone's strapped rockets onto it. And we were able to really help our customers make that transition from working largely in the office or teaching largely in the classroom to shifting towards a work from home, learn from home type of environment. And now with those investments made, they're now better positioned to be when we're out the other side of this, to be the work from anywhere company. And that's really going to help them have access to talent globally because we realized that you can get the work done regardless of where you are.

Matt Baker: That's one example. Probably the biggest shock to businesses when this happened, besides the demand side of things, it was the supply chain side of things. And so people were just struggling to work through and unfortunately had inefficient non-digital supply chains. We've got some great lessons we can share with our customers on supply chain diversity and really ensuring that you're prepared. I guess it took a pandemic to prove it but the reality is, is once you develop a more efficient supply chain and a more diverse supply chain, you're better able to weather all sorts of storms. Be they geopolitical weirdness, be they pandemics, or frankly, just having access to more sources of supply that just makes you more flexible, cost efficient and better able to deliver innovative products to customers. And guess what? How do you do that? You take data and you analyze the supply chain and you apply data science to your supply chain. You create a world class digital supply chain and we've got great experience doing that. And we're going to be there for you. We're going to lean in when times are tough and we're going to support you.

Jon Hyde: Yeah, absolutely. We started with the statement that I offered around data is always a big part of the problem, but data is also a big part of the solution. That sounds super simple. It's like, oh, I just have to understand my data. Well, it's not simple. I think that's exactly why we're having this conversation. Understanding the daunting nature of that simplified statement, what's the one thing that you would suggest every customer try to do?

Matt Baker: Yeah. Well, it's unfortunately not a one thing answer. Maybe I'll answer it by saying what I wouldn't do first and then let's talk about what I would do. I wouldn't go hiring a bunch of data scientists out of the gate because the data scientists are there to create algorithms, to answer questions that you give them. They are not going to figure out what the questions to be asked. I think the one thing people need to do is they need to ask themselves, "What am I trying to do with my customer experience or my user experience? What outcomes am I trying to build?" And from there then there will come questions about, all right, well, how do I automate that? What questions am I going to ask that I'd like to automate? Then you apply the data scientists.

Matt Baker: At the end of the day, the first thing you want to do to be a data fueled business is you want to have a really strong digital transformation strategy. And that comes back down to planning, all right, what are my business processes? If I were to automate that business process, what outcome would it create? Would it help? Would it hurt? And it's really frankly, to build a digital transformation strategy. That's what you have to do first. And you can't do that without data. And you won't achieve that without data. But understanding what you're setting out to achieve is incredibly important. There's no amount of Hadoop, there's no amount of AI, there's no amount of whatever buzzword on any given year, any given decade, there's no buzzword technology that you're going to buy that's going to solve the problem. The problem is all about asking the right questions and planning the right business.

Matt Baker: Build a bite size plan and really understand what outcome you're trying to achieve. Don't go technology first, go business process first and work your way through one business process at a time. And soon enough, you'll find yourself galloping forward towards your digital transformation and you will make a tremendous amount of progress. It's about taking one step, not a giant leap across the universe.

Jon Hyde: I think that is the perfect way to encapsulate this and Matt Baker, thank you so much for your time today. It's been an awesome conversation. Really appreciate it.

Matt Baker: Yeah. Well, thank you. It's been great.

Jon Hyde: For those of you who enjoyed this podcast, you can find it at www.delltechnologies.com/nexthorizon, along with feature podcasts and other great content focused on emerging technologies. Thank you so much for listening and be sure to subscribe. Until next time, I'm Jon Hyde and this is The Next Horizon.