Podcasts

Podcast with Matt Langione, Boston Consulting Group

11
May
,
2022

My guest today is Matt Langione, a partner at the Boston Consulting Group, and the North American lead in their deep tech mission. We spoke about the advice he gives to enterprises, VCs and quantum vendors, how BCG’s thinking about popular quantum computing applications has evolved recently, and much more.

Listen to additional podcasts here

THE FULL TRANSCRIPT IS BELOW

Yuval: Hello, Matt, and thanks for joining me today.

Matt: Hi, Yuval, how are you doing?

Yuval: Living the dream. So, who are you and what do you do?

Matt: Well, I'm Matt Langione, I'm a Partner at Boston Consulting Group, and I'm the North American lead in our deep tech mission. And, the deep tech mission is really around helping research-inflected, science-based technologies emerge in the market. Typically, these are maybe three to five years away from disrupting important industries. Quantum computing is the example, par excellence, of deep tech for us, so a lot of my work focuses on quantum and of course that's how you and I met.

Yuval: What percentage of your time roughly do you spend on quantum?

Matt: That's a good question, I would probably say... And, I'm going to exclude non-work time, because I have to admit I've developed a personal interest that stretches beyond BCG, but I would say of my BCG time, I don't know, maybe 60% or so now.

Yuval: Wow, and you mentioned that these kind of technologies that you focus on are three to five years away. Is that what you feel about quantum? Some people say it's 30 to 50 years away, what's your take?

Matt: Well, it depends on what you mean, so yes, I think so. I mean, quantum in sense, it's here now and it has been since 2017 or '18 in terms of cloud-accessible, quantum computing. I think, we are just on the precipice of cloud available quantum computers that are powerful and excessive of simulators. So actually, I think in the next, even less than three years, so in the next year or two, I think that we'll really be there. So, quantum computing is in a sense here, now when we think about quantum computing as the broadly transformative technology that a lot of us believe that it can be. I do still think that there's time for that to emerge, right? I mean, you'll need a number of not just engineering achievements, but even some scientific breakthroughs, depending on what physical and implementation we're talking about in order to make that kind of extraordinary value creation that we're all very excited about possible.

So, in a sense, it's a little bit of an evasion, but really it's here, in kind of in early form. And, I think that companies need to act fast in order to make sure that they grow at the pace that the technology's going to improve.

Yuval: At what stage does BCG come in? And, for those that are unfamiliar with the Boston Consulting Group, do you do strategy, do you do implementation, do you do after sales service? Where do you come in?

Matt: Well, I mean, I guess I would first say that there are really kind of three constituents that we work with primarily in quantum computing. The first is providers of the technology, and historically that's been sort of the biggest segment, maybe 50% of our work has been with folks that are actually producing quantum computers. Then, there's another segment on investors, so folks that are investing, you've obviously seen a lot of capital. I think, 90% of all capital that's been spent on quantum computing in its history has happened in the last two years. So obviously, it's been a fertile and active area, and that's around understanding the technology, doing diligence work. That, maybe is say 30% or so of our work, and then the final say 15 to 20% is with potential end users of the technology.

So, these are the folks in industries that have the potential to be disrupted by quantum, which is to say drug discoveries, so pharma, financial institutions, aerospace automotive for say computational fluid dynamics. Those are the kind of potential end users. And actually, that last bit of work is, I said, "15 to 20%." It's actually growing the fastest, which I regard as an important kind of bellwether of the maturity of quantum computing. In other words, you'd want to see our mix shift towards end users, folks that are actually implementing and experimenting with the technology. And that, seems to be the case, a lot of our kind of inbound interest is now coming from that end user community, which is really great to see. And then, the question is, okay, so within each of these, the tech providers, the investors and the end users, sort of, where do we come in and what do we do. You're right to say, BCG classically, it's a strategy firm.

And, I think that a lot of our work is on strategy, so for tech providers, it's around things like what is our value creation roadmap, what business model should we employ to maximize the value capture over time? What industries and use cases should we prioritize? How do we partner and develop an ecosystem to get complimentary layers of the stack? You've seen obviously a lot of mergers in terms of software and hardware providers recently. So, there's a lot of strategy work there for the technology providers. For investors, of course, it's mostly diligence work, which I think falls in the broad category of strategy. Now, for end users, what we do is we have this thing called an impact of quantum or an IQ assessment where we help dimensionalize the value of quantum computing in their industry and specifically with their own workflows.

So, we get pretty granular with them so they can make kind of an ROI influence decision on not just if, but when to onboard a quantum capability, and then how. The how thing is where we get into some implementation, because it's not just around vendor selection. Of course, many of these end users need to partner with folks like Classiq, with, say IBM. But, there's also developing a talent strategy, there's incorporating quantum into their transformation roadmap. When we get into implementation, it's usually in that broader kind of digital transformation context. I'd say most of our work right now, Yuval, is on the strategy side, but we're certainly building out for end users kind of implementation focus as well.

Yuval: So, let me start diving into each of these segments, if I may.

Matt: Sure.

Yuval: When you talk about vendor, or creators of computers, or software, and so on, and also a little bit touching on investors, how do you feel about the vertical integration strategy? I mean, we've seen some M&A work that seems to form sort of a monolithic company that can do everything from hardware, all the way to the application. What do you say to investors or companies, and what do you say to end users about them? Should you look for a monolithic vendor or should you prefer more of a best of breed approach?

Matt: It's a really, really good and kind of deep question, and it's one that we're being asked a lot kind of responsively to the verticalization that you reference Yuval. So, I think that it kind of depends on where an end user is in their quantum journey, if they are trying to partner in order to understand the technology better in order to kind of make a decision about whether they really want to invest in building a quantum capability. Sometimes, it's helpful to have a vertically integrated... effectively, a full stack provider that just sort of simplifies speed to action and things like that. And so, I think for a certain set of users that are more in this kind of dipping their toes in the water, maybe doing a short project ahead of a bigger kind of executive level investment decision.

I think, sometimes that kind of vertically integrated provider can be really useful for folks that have committed to quantum, that already onboarding talent, maybe they have a small team. You think it kind of eventually a best of breed approach, that's akin to what's happened in classical technology. Classical compute technologies is likely to kind of win in the long term. And of course, that's exactly why one of the things that drives my enthusiasm and interest in classic of course, is the ability to be agnostic and to kind of deploy anywhere. Because, I do think that there's risk associated with any physical implementation in any company that's operating within the context of a single physical implementation.

Yuval: And, related to this, what do the companies think about IP? Because, they could say, oh, we want to develop the IP internally, we want to onboard people that can develop IP, and then we want to reuse it later. Or, alternatives, they could go and say, we want a joint developer agreement with company XYZ, and we understand that the IP will partially be owned by that third party provider, or there will be some other IP arrangement. And then, there may be a risk that some of that IP, I may find in a competitor's product one day.

Matt: It's a good question, I mean, I think that most companies that we talk to want to have ownership over their IP, right? So, they want to be able to build a library of models, or create custom circuits that are kind of theirs alone. And so, they're looking for enablers of that, I think that's very difficult to do in-house and without a partner, right? But, I think that at companies that allow end users to own their IP will be really advantaged in this space. In some cases, of course, that's not going to be possible and sometimes the need for talent, like deep research talent in order to create the IP, are ways the impetus to have it, kind of, to own it. But, I think that the general trend that we're seeing is that folks want to own the IP as much as possible.

Yuval: When you look at the C-suite, I could imagine that the CTO or CIO heard about quantum, they want to explore quantum. But, just the other day, I met a group of CISOs, chief information security officers, and we were talking about, of course, breaking RSA-2048, but also about the users of quantum computing for optimization, for risk analysis, for machine learning and so on. Which is the C level executive in your experience that gets most interest it and excited about quantum?

Matt: Well, it's a good question, and really the reason that I wanted to come to BCG and to do deep tech was because I am really interested in simplifying explaining landing the value of deep technologies, really with the key decision maker. And oftentimes, outside of government, those key decision makers are C-suites, at kind of fortune 500 companies. And that, really is our bread and butter. I would say to kind of answer the question, it's sort of a bimodal answer. There are the folks that have heard about quantum from the perspective of security risk, that's like a bucket of folks, right? And then, there are folks that have heard of it from the perspective of opportunity. So from the security risk angle, it's typically your CISO and your CIO.

From the opportunity angle, it tends to be more like a general manager of a business unit, a chief revenue officer, and the call conversation is just extraordinarily different for each one. So, there's the cryptographic conversation that you can have that I think is really motivating for executives because it's real, there's a lot of downside. I mean, it's typically taken 15 to 20 years to implement a new cryptographic protocol, at least in the US. And so, even if you think that quantum computing's a little ways away, when you just take a quick look at the timeline, you realize a time to act was sort of yesterday. And so, I think that can be a motivating conversation for CISOs and CIOs. When I think about the upside conversation, I mean that one can go right to the CEO or the head of a business unit at a big company.

And, that one is of course exciting because, because I think at this point, quantum computing is... When, I started looking into it in 2016 to 2018, it was really a deep tech and it was deep. Now, I feel that it's emerged quite a bit, so most folks, when I talk to them have had a bit of a briefing on the technology, maybe they've listened to your podcasts. And, we can get into a deeper conversation about the potential disruption in an industry. The interesting thing that I find Yuval is that there's something about whether quantum will solve problems in your industry. And then, there's something about the actual... There's almost an autotelic benefit in a given industry to just look at it in a different way. Suppose you have had vastly superior compute capacity, what things could you do?

And, what we find when we do these impact of quantum assessments, is that we find that people tend to take a very narrow view of the value that compute can take, because they're avoiding a lot of problems because they are sort of intractable with classic computers, right? So, they're not using compute to do optimization of their fleet, right? They're, not doing compute to do a lot of different things. And, they're thinking of compute in the narrow way that they use it now, I think actually just the specter of quantum computing is itself a benefit to these companies as they think about ways that they can re-engineer their entire way of working.

Yuval: One of the ways that we had success is people ask about ROI, and there's an “I” in the ROI and that's the investment. And, we're basically saying, look, the investment in quantum computing right now is not that large. You don't need to buy a quantum computer, you need to hire a couple folks and buy some cloud time. So, you almost could position it like an insurance policy, you're paying something to ensure that you're going to have the technology that you need in the future. And, if you don't, who knows you may be exposed to catastrophic damage. Do you see executives that think about it this way as an insurance policy?

Matt: Yes, I do, I think that there's a couple interesting things about the ROI as well, because we are working both with end users and tech providers, we're kind of very concerned with both the R and the I, right? Because, in a sense you kind of want the I to be reasonably high because you think that, I mean, this is an extraordinarily difficult technology to manufacture, ideate, and produce, and there's a kind of fair share of the value creation that should go to tech providers up and down the stack, but the returns are also massive, right? So, no matter how we articulate ROI in the long term, it's almost impossible to imagine that should the technology come to good and work in the way that we think it will, that the ROI will be very favorable for end users.

But, it's also very case dependent, Yuval, as you know because in some cases, quantum computing will actually save a lot of money on very compute heavy... Or, in all likelihood will save of money on very compute heavy tasks, right? That right now, are very costly on classical computers. And in others, the way to articulate savings will be actually in terms of cost outlays that are avoided. So, for example, in computational fluid dynamics, it wouldn't be about the spend that they have on HPC now. The right way to think about ROI would be avoiding things like wing flex tip tests, and really expensive tunnel tests, those cost aerospace and automotive companies 35 billion dollars a year, right? So, it's really what kind of non-com computational activities you'll be able to void that really... I think, that the ROI complication requires you to take a very broad and comprehensive view.

Yuval: When people talk about quantum, they talk about computing, sensing, and communications. I forgot to ask, do you guys deal with all three, and do your customers have the same level of interest in all three?

Matt: So, to answer your first question, yes, we do deal with all three, I would say that our expertise is freighted toward quantum computing, we certainly do diligences and have done some case work on communications and sensing. But, the reason that our expertise is tilted toward quantum computing is because the bulk of inquiries that come in for BCG are around quantum computing. The nice thing about sensing is that it's more or less available today, right? As opposed to being a roadmap technology, quantum sensing is sort of here. And certainly, there's a lot of government defense kind of interest in sensing. Quantum communication is also fairly near term, right? And, there's a lot of work that's going into quantum communication, but really the bulk, in our view, the site of largest value creation, even if it's not the nearest term. And, the thing that we get the most questions about from our clients is quantum computing.

Yuval: And, when you dive into quantum computing, people have talked about quantum machine learning about optimization, about chemical analysis and so on. Is there one application that comes to the forefront and others that say, oh, no, this is a couple years away so let's talk about something else first?

Matt: Well, it's interesting that you mentioned this, so I'll just say in our view, it all kind of comes down to large sparse matrix math. That's sort of the one quantum advantage mathematical function that this whole industry is sort of based on. Now, it turns out that there are four sort of computational problem types that are based on this large sparse matrix math. Simulation is one, optimization, machine learning, and cryptography, and people have different hierarchies and things like that. But, those sort of four problems actually open out onto an enormous number of very, very high value problems in industry, things like drug discovery and pharma, or as I mentioned, "CFD for aerospace and automotive." That's in simulation, things like risk management in insurance, portfolio optimization in finance. Then, in machine learning, of course, there's a whole host of use cases, in tech, in finance, et-cetera.

And cryptography, as you now obviously there's the kind of encryption and decryption use cases. Our view, and this has changed a little bit since 2018 when we wrote our first Value Creation report. But, our view is that the first use case is going to be simulation. That part actually hasn't changed, right? The idea that you can precisely and efficiently simulate the dynamics of quantum systems. And so, the biggest use cases there would be in say chemistry, designing catalysts, or in pharma, drug discovery, beyond small molecules. We used to think that optimization was coming next because machine learning would be encumbered by data ingestion bottlenecks. Our thinking has changed on that a bit, and we think that machine learning actually may be next in line.

And, with machine learning, of course, there are a lot of use cases, whether it's anti-money laundering or anti-fraud, whether it's search or ad optimization in tech. So, any kind of machine learning use case that you have now we think would be kind of on the wing. And then, next after that would be the traditional optimization use cases like supply chain optimization, portfolio optimization, and these kind of things, logistics. That's sort of our sense of the timeline, so simulation first, then machine learning, then optimization.

Yuval: As we get close to the end of our conversation today, I want to talk to you about the process. So, let's say I'm a CTO, I find quantum very compelling, I've listened to this podcast, you've been super fluent and engaging, I want to bring in BCG to help me figure out my digital transformation and quantum computing strategy. And, I'm assuming that you're going to work for a period of time, help me identify the use cases, and then potentially hand it off to a company like Classiq to actually implement a proof of concept. How long does that process take? How long on average, should I budget, for BCG to be working with me, to identify and provide me with this roadmap?

Matt: So, we're pretty flexible, and there are obviously a lot of dependencies and factors there, but honestly it often converges in this sort of maybe 10 to 16 week kind of timeframe. One of the virtues of working with BCG is... And actually, one of the challenges of working at BCG is that we work typically very fast. The idea is that we'll get deeply, deeply embedded with your team, we'll bring in experts really quickly. And now that we've got a lot of experience in quantum, we can move fast. So, from first phone call to me to the handoff at the end of the project could be 10, 12 weeks. And then, I think you're absolutely right though, that for implementation, it's typically... I mean, of these projects end with some kind of recommendation around talent onboarding, and talent onboarding tends to be something that is far more complicated than most CIOs, CTOs, et-cetera, tend to imagine within quantum, there's just a real talent scarcity.

And so, typically what ends up happening is that for implementation, for enablement, BCG sometimes plays a supporting role there, and oftentimes you're right, it'd be a company like Classiq that kind of takes up the mantle thereafter.

Yuval: And, the trigger for the initial call, is it typically I've read about quantum, I want to figure out what that does for me? Or, is it more, I've seen my competitor across the street publish something and I don't want to be left behind?

Matt: Yeah, so I've got kind of the Ted talk, and so for me, a lot of it is, "Hey, I saw your Ted talk where I read one of the BCG articles that have been circulating on Value Creation. I'm sitting here in... I'm at." I don't know what it is, Pfizer, or I'm at Dow, or something like that. "I'm in one of these industries that you highlight, let's have a phone call." And, we'll give them a little bit of an overview and tell them what the work would entail. So, it's mostly that for now, but I don't deny that FOMO might play an increasingly big role as you see companies that are partnering with Classiq or partnering with IBM and producing really interesting earlier results. I think, right now the lovely thing is that those are being shared and really seen as an indicator of sophistication and being at the vanguard of research and development. And so, those are all fairly public now, and I don't doubt that's going to be kind of a driver of interest in quantum computing going forward in the C-suite.

Yuval: So Matt, how can people get in touch with you to learn more about your work?

Matt: Well, I think, they should email me, langione.matt@bcg.com. Certainly, download the white papers, you can sort of Google them quite easily, watch the Ted talk, but really get in touch with me. I'm the north American lead of our deep technician, Jean-François Bobier, so bobier.jeanfrançois@bcg.com, is sort of the European lead. Together we can absolutely, we can absolutely support and just really look forward to meeting folks that are interested in implementing really kind of the change agents, right from classical to quantum are going to be these end users, right? And so, we love working at the interface of end users, tech providers, and of course, folks like yourself.

Yuval: Fantastic, well, Matt, thank you so much for joining me today.

Matt: Thank you.

My guest today is Matt Langione, a partner at the Boston Consulting Group, and the North American lead in their deep tech mission. We spoke about the advice he gives to enterprises, VCs and quantum vendors, how BCG’s thinking about popular quantum computing applications has evolved recently, and much more.

Listen to additional podcasts here

THE FULL TRANSCRIPT IS BELOW

Yuval: Hello, Matt, and thanks for joining me today.

Matt: Hi, Yuval, how are you doing?

Yuval: Living the dream. So, who are you and what do you do?

Matt: Well, I'm Matt Langione, I'm a Partner at Boston Consulting Group, and I'm the North American lead in our deep tech mission. And, the deep tech mission is really around helping research-inflected, science-based technologies emerge in the market. Typically, these are maybe three to five years away from disrupting important industries. Quantum computing is the example, par excellence, of deep tech for us, so a lot of my work focuses on quantum and of course that's how you and I met.

Yuval: What percentage of your time roughly do you spend on quantum?

Matt: That's a good question, I would probably say... And, I'm going to exclude non-work time, because I have to admit I've developed a personal interest that stretches beyond BCG, but I would say of my BCG time, I don't know, maybe 60% or so now.

Yuval: Wow, and you mentioned that these kind of technologies that you focus on are three to five years away. Is that what you feel about quantum? Some people say it's 30 to 50 years away, what's your take?

Matt: Well, it depends on what you mean, so yes, I think so. I mean, quantum in sense, it's here now and it has been since 2017 or '18 in terms of cloud-accessible, quantum computing. I think, we are just on the precipice of cloud available quantum computers that are powerful and excessive of simulators. So actually, I think in the next, even less than three years, so in the next year or two, I think that we'll really be there. So, quantum computing is in a sense here, now when we think about quantum computing as the broadly transformative technology that a lot of us believe that it can be. I do still think that there's time for that to emerge, right? I mean, you'll need a number of not just engineering achievements, but even some scientific breakthroughs, depending on what physical and implementation we're talking about in order to make that kind of extraordinary value creation that we're all very excited about possible.

So, in a sense, it's a little bit of an evasion, but really it's here, in kind of in early form. And, I think that companies need to act fast in order to make sure that they grow at the pace that the technology's going to improve.

Yuval: At what stage does BCG come in? And, for those that are unfamiliar with the Boston Consulting Group, do you do strategy, do you do implementation, do you do after sales service? Where do you come in?

Matt: Well, I mean, I guess I would first say that there are really kind of three constituents that we work with primarily in quantum computing. The first is providers of the technology, and historically that's been sort of the biggest segment, maybe 50% of our work has been with folks that are actually producing quantum computers. Then, there's another segment on investors, so folks that are investing, you've obviously seen a lot of capital. I think, 90% of all capital that's been spent on quantum computing in its history has happened in the last two years. So obviously, it's been a fertile and active area, and that's around understanding the technology, doing diligence work. That, maybe is say 30% or so of our work, and then the final say 15 to 20% is with potential end users of the technology.

So, these are the folks in industries that have the potential to be disrupted by quantum, which is to say drug discoveries, so pharma, financial institutions, aerospace automotive for say computational fluid dynamics. Those are the kind of potential end users. And actually, that last bit of work is, I said, "15 to 20%." It's actually growing the fastest, which I regard as an important kind of bellwether of the maturity of quantum computing. In other words, you'd want to see our mix shift towards end users, folks that are actually implementing and experimenting with the technology. And that, seems to be the case, a lot of our kind of inbound interest is now coming from that end user community, which is really great to see. And then, the question is, okay, so within each of these, the tech providers, the investors and the end users, sort of, where do we come in and what do we do. You're right to say, BCG classically, it's a strategy firm.

And, I think that a lot of our work is on strategy, so for tech providers, it's around things like what is our value creation roadmap, what business model should we employ to maximize the value capture over time? What industries and use cases should we prioritize? How do we partner and develop an ecosystem to get complimentary layers of the stack? You've seen obviously a lot of mergers in terms of software and hardware providers recently. So, there's a lot of strategy work there for the technology providers. For investors, of course, it's mostly diligence work, which I think falls in the broad category of strategy. Now, for end users, what we do is we have this thing called an impact of quantum or an IQ assessment where we help dimensionalize the value of quantum computing in their industry and specifically with their own workflows.

So, we get pretty granular with them so they can make kind of an ROI influence decision on not just if, but when to onboard a quantum capability, and then how. The how thing is where we get into some implementation, because it's not just around vendor selection. Of course, many of these end users need to partner with folks like Classiq, with, say IBM. But, there's also developing a talent strategy, there's incorporating quantum into their transformation roadmap. When we get into implementation, it's usually in that broader kind of digital transformation context. I'd say most of our work right now, Yuval, is on the strategy side, but we're certainly building out for end users kind of implementation focus as well.

Yuval: So, let me start diving into each of these segments, if I may.

Matt: Sure.

Yuval: When you talk about vendor, or creators of computers, or software, and so on, and also a little bit touching on investors, how do you feel about the vertical integration strategy? I mean, we've seen some M&A work that seems to form sort of a monolithic company that can do everything from hardware, all the way to the application. What do you say to investors or companies, and what do you say to end users about them? Should you look for a monolithic vendor or should you prefer more of a best of breed approach?

Matt: It's a really, really good and kind of deep question, and it's one that we're being asked a lot kind of responsively to the verticalization that you reference Yuval. So, I think that it kind of depends on where an end user is in their quantum journey, if they are trying to partner in order to understand the technology better in order to kind of make a decision about whether they really want to invest in building a quantum capability. Sometimes, it's helpful to have a vertically integrated... effectively, a full stack provider that just sort of simplifies speed to action and things like that. And so, I think for a certain set of users that are more in this kind of dipping their toes in the water, maybe doing a short project ahead of a bigger kind of executive level investment decision.

I think, sometimes that kind of vertically integrated provider can be really useful for folks that have committed to quantum, that already onboarding talent, maybe they have a small team. You think it kind of eventually a best of breed approach, that's akin to what's happened in classical technology. Classical compute technologies is likely to kind of win in the long term. And of course, that's exactly why one of the things that drives my enthusiasm and interest in classic of course, is the ability to be agnostic and to kind of deploy anywhere. Because, I do think that there's risk associated with any physical implementation in any company that's operating within the context of a single physical implementation.

Yuval: And, related to this, what do the companies think about IP? Because, they could say, oh, we want to develop the IP internally, we want to onboard people that can develop IP, and then we want to reuse it later. Or, alternatives, they could go and say, we want a joint developer agreement with company XYZ, and we understand that the IP will partially be owned by that third party provider, or there will be some other IP arrangement. And then, there may be a risk that some of that IP, I may find in a competitor's product one day.

Matt: It's a good question, I mean, I think that most companies that we talk to want to have ownership over their IP, right? So, they want to be able to build a library of models, or create custom circuits that are kind of theirs alone. And so, they're looking for enablers of that, I think that's very difficult to do in-house and without a partner, right? But, I think that at companies that allow end users to own their IP will be really advantaged in this space. In some cases, of course, that's not going to be possible and sometimes the need for talent, like deep research talent in order to create the IP, are ways the impetus to have it, kind of, to own it. But, I think that the general trend that we're seeing is that folks want to own the IP as much as possible.

Yuval: When you look at the C-suite, I could imagine that the CTO or CIO heard about quantum, they want to explore quantum. But, just the other day, I met a group of CISOs, chief information security officers, and we were talking about, of course, breaking RSA-2048, but also about the users of quantum computing for optimization, for risk analysis, for machine learning and so on. Which is the C level executive in your experience that gets most interest it and excited about quantum?

Matt: Well, it's a good question, and really the reason that I wanted to come to BCG and to do deep tech was because I am really interested in simplifying explaining landing the value of deep technologies, really with the key decision maker. And oftentimes, outside of government, those key decision makers are C-suites, at kind of fortune 500 companies. And that, really is our bread and butter. I would say to kind of answer the question, it's sort of a bimodal answer. There are the folks that have heard about quantum from the perspective of security risk, that's like a bucket of folks, right? And then, there are folks that have heard of it from the perspective of opportunity. So from the security risk angle, it's typically your CISO and your CIO.

From the opportunity angle, it tends to be more like a general manager of a business unit, a chief revenue officer, and the call conversation is just extraordinarily different for each one. So, there's the cryptographic conversation that you can have that I think is really motivating for executives because it's real, there's a lot of downside. I mean, it's typically taken 15 to 20 years to implement a new cryptographic protocol, at least in the US. And so, even if you think that quantum computing's a little ways away, when you just take a quick look at the timeline, you realize a time to act was sort of yesterday. And so, I think that can be a motivating conversation for CISOs and CIOs. When I think about the upside conversation, I mean that one can go right to the CEO or the head of a business unit at a big company.

And, that one is of course exciting because, because I think at this point, quantum computing is... When, I started looking into it in 2016 to 2018, it was really a deep tech and it was deep. Now, I feel that it's emerged quite a bit, so most folks, when I talk to them have had a bit of a briefing on the technology, maybe they've listened to your podcasts. And, we can get into a deeper conversation about the potential disruption in an industry. The interesting thing that I find Yuval is that there's something about whether quantum will solve problems in your industry. And then, there's something about the actual... There's almost an autotelic benefit in a given industry to just look at it in a different way. Suppose you have had vastly superior compute capacity, what things could you do?

And, what we find when we do these impact of quantum assessments, is that we find that people tend to take a very narrow view of the value that compute can take, because they're avoiding a lot of problems because they are sort of intractable with classic computers, right? So, they're not using compute to do optimization of their fleet, right? They're, not doing compute to do a lot of different things. And, they're thinking of compute in the narrow way that they use it now, I think actually just the specter of quantum computing is itself a benefit to these companies as they think about ways that they can re-engineer their entire way of working.

Yuval: One of the ways that we had success is people ask about ROI, and there's an “I” in the ROI and that's the investment. And, we're basically saying, look, the investment in quantum computing right now is not that large. You don't need to buy a quantum computer, you need to hire a couple folks and buy some cloud time. So, you almost could position it like an insurance policy, you're paying something to ensure that you're going to have the technology that you need in the future. And, if you don't, who knows you may be exposed to catastrophic damage. Do you see executives that think about it this way as an insurance policy?

Matt: Yes, I do, I think that there's a couple interesting things about the ROI as well, because we are working both with end users and tech providers, we're kind of very concerned with both the R and the I, right? Because, in a sense you kind of want the I to be reasonably high because you think that, I mean, this is an extraordinarily difficult technology to manufacture, ideate, and produce, and there's a kind of fair share of the value creation that should go to tech providers up and down the stack, but the returns are also massive, right? So, no matter how we articulate ROI in the long term, it's almost impossible to imagine that should the technology come to good and work in the way that we think it will, that the ROI will be very favorable for end users.

But, it's also very case dependent, Yuval, as you know because in some cases, quantum computing will actually save a lot of money on very compute heavy... Or, in all likelihood will save of money on very compute heavy tasks, right? That right now, are very costly on classical computers. And in others, the way to articulate savings will be actually in terms of cost outlays that are avoided. So, for example, in computational fluid dynamics, it wouldn't be about the spend that they have on HPC now. The right way to think about ROI would be avoiding things like wing flex tip tests, and really expensive tunnel tests, those cost aerospace and automotive companies 35 billion dollars a year, right? So, it's really what kind of non-com computational activities you'll be able to void that really... I think, that the ROI complication requires you to take a very broad and comprehensive view.

Yuval: When people talk about quantum, they talk about computing, sensing, and communications. I forgot to ask, do you guys deal with all three, and do your customers have the same level of interest in all three?

Matt: So, to answer your first question, yes, we do deal with all three, I would say that our expertise is freighted toward quantum computing, we certainly do diligences and have done some case work on communications and sensing. But, the reason that our expertise is tilted toward quantum computing is because the bulk of inquiries that come in for BCG are around quantum computing. The nice thing about sensing is that it's more or less available today, right? As opposed to being a roadmap technology, quantum sensing is sort of here. And certainly, there's a lot of government defense kind of interest in sensing. Quantum communication is also fairly near term, right? And, there's a lot of work that's going into quantum communication, but really the bulk, in our view, the site of largest value creation, even if it's not the nearest term. And, the thing that we get the most questions about from our clients is quantum computing.

Yuval: And, when you dive into quantum computing, people have talked about quantum machine learning about optimization, about chemical analysis and so on. Is there one application that comes to the forefront and others that say, oh, no, this is a couple years away so let's talk about something else first?

Matt: Well, it's interesting that you mentioned this, so I'll just say in our view, it all kind of comes down to large sparse matrix math. That's sort of the one quantum advantage mathematical function that this whole industry is sort of based on. Now, it turns out that there are four sort of computational problem types that are based on this large sparse matrix math. Simulation is one, optimization, machine learning, and cryptography, and people have different hierarchies and things like that. But, those sort of four problems actually open out onto an enormous number of very, very high value problems in industry, things like drug discovery and pharma, or as I mentioned, "CFD for aerospace and automotive." That's in simulation, things like risk management in insurance, portfolio optimization in finance. Then, in machine learning, of course, there's a whole host of use cases, in tech, in finance, et-cetera.

And cryptography, as you now obviously there's the kind of encryption and decryption use cases. Our view, and this has changed a little bit since 2018 when we wrote our first Value Creation report. But, our view is that the first use case is going to be simulation. That part actually hasn't changed, right? The idea that you can precisely and efficiently simulate the dynamics of quantum systems. And so, the biggest use cases there would be in say chemistry, designing catalysts, or in pharma, drug discovery, beyond small molecules. We used to think that optimization was coming next because machine learning would be encumbered by data ingestion bottlenecks. Our thinking has changed on that a bit, and we think that machine learning actually may be next in line.

And, with machine learning, of course, there are a lot of use cases, whether it's anti-money laundering or anti-fraud, whether it's search or ad optimization in tech. So, any kind of machine learning use case that you have now we think would be kind of on the wing. And then, next after that would be the traditional optimization use cases like supply chain optimization, portfolio optimization, and these kind of things, logistics. That's sort of our sense of the timeline, so simulation first, then machine learning, then optimization.

Yuval: As we get close to the end of our conversation today, I want to talk to you about the process. So, let's say I'm a CTO, I find quantum very compelling, I've listened to this podcast, you've been super fluent and engaging, I want to bring in BCG to help me figure out my digital transformation and quantum computing strategy. And, I'm assuming that you're going to work for a period of time, help me identify the use cases, and then potentially hand it off to a company like Classiq to actually implement a proof of concept. How long does that process take? How long on average, should I budget, for BCG to be working with me, to identify and provide me with this roadmap?

Matt: So, we're pretty flexible, and there are obviously a lot of dependencies and factors there, but honestly it often converges in this sort of maybe 10 to 16 week kind of timeframe. One of the virtues of working with BCG is... And actually, one of the challenges of working at BCG is that we work typically very fast. The idea is that we'll get deeply, deeply embedded with your team, we'll bring in experts really quickly. And now that we've got a lot of experience in quantum, we can move fast. So, from first phone call to me to the handoff at the end of the project could be 10, 12 weeks. And then, I think you're absolutely right though, that for implementation, it's typically... I mean, of these projects end with some kind of recommendation around talent onboarding, and talent onboarding tends to be something that is far more complicated than most CIOs, CTOs, et-cetera, tend to imagine within quantum, there's just a real talent scarcity.

And so, typically what ends up happening is that for implementation, for enablement, BCG sometimes plays a supporting role there, and oftentimes you're right, it'd be a company like Classiq that kind of takes up the mantle thereafter.

Yuval: And, the trigger for the initial call, is it typically I've read about quantum, I want to figure out what that does for me? Or, is it more, I've seen my competitor across the street publish something and I don't want to be left behind?

Matt: Yeah, so I've got kind of the Ted talk, and so for me, a lot of it is, "Hey, I saw your Ted talk where I read one of the BCG articles that have been circulating on Value Creation. I'm sitting here in... I'm at." I don't know what it is, Pfizer, or I'm at Dow, or something like that. "I'm in one of these industries that you highlight, let's have a phone call." And, we'll give them a little bit of an overview and tell them what the work would entail. So, it's mostly that for now, but I don't deny that FOMO might play an increasingly big role as you see companies that are partnering with Classiq or partnering with IBM and producing really interesting earlier results. I think, right now the lovely thing is that those are being shared and really seen as an indicator of sophistication and being at the vanguard of research and development. And so, those are all fairly public now, and I don't doubt that's going to be kind of a driver of interest in quantum computing going forward in the C-suite.

Yuval: So Matt, how can people get in touch with you to learn more about your work?

Matt: Well, I think, they should email me, langione.matt@bcg.com. Certainly, download the white papers, you can sort of Google them quite easily, watch the Ted talk, but really get in touch with me. I'm the north American lead of our deep technician, Jean-François Bobier, so bobier.jeanfrançois@bcg.com, is sort of the European lead. Together we can absolutely, we can absolutely support and just really look forward to meeting folks that are interested in implementing really kind of the change agents, right from classical to quantum are going to be these end users, right? And so, we love working at the interface of end users, tech providers, and of course, folks like yourself.

Yuval: Fantastic, well, Matt, thank you so much for joining me today.

Matt: Thank you.

About "The Qubit Guy's Podcast"

Hosted by The Qubit Guy (Yuval Boger, our Chief Marketing Officer), the podcast hosts thought leaders in quantum computing to discuss business and technical questions that impact the quantum computing ecosystem. Our guests provide interesting insights about quantum computer software and algorithm, quantum computer hardware, key applications for quantum computing, market studies of the quantum industry and more.

If you would like to suggest a guest for the podcast, please contact us.

Start Creating Quantum Software Without Limits

contact us