This is a transcript from the live Roundtable event. Language has been slightly edited for readability, but might not flow as natural written language would. Please excuse any syntax or grammatical errors in this transcription.
Stephen Lowisz: This is the first [Automation Trends Roundtable], but certainly not the last. We plan on doing these every month or so with a different speaker, different spin, et cetera. The goal here is to have a true Roundtable, a discussion. Although I have some points and specific questions that I want to ask Ian to kick the conversation off – and I'll introduce [those] here in just a moment – by all means, we didn't do this in a webinar format for a reason. We're looking for engagement, so drill Ian with questions. Not myself, but put them all on Ian to answer. And even ask questions amongst yourselves… No pressure, Ian.
A couple of things that I will go over. So, the entire premise of this is how to delight customers and reduce unit costs, and the conversation that we've had – both with Ian, and even some of you guys offline – is that the market has changing demands. The market demands speed, but personalization. So it’s high touch, but automation. That puts you in a pickle.
With low interest rates, we have to cut costs and reduce our target cost per unit. But to do that, it requires investment in digital transformation – Mike and a few of you understand that better than anybody right now. This means a massive cash outlay and, in a time like this, that's not something that anyone wants to do.
That's the frame for the conversation. Now I'll throw a couple of stats, and then I want to introduce Ian here, but I'll throw a couple of stats out your way.
In 2019, there was a study done by MX Group. And I'll send it to everybody after this Roundtable, but it found that 56% of Millennials in the B2B sector hold director-level positions.
Now, I think we've all been to a conference (before this COVID stuff happened) where people are complaining about Millennials and their need for instant gratification, et cetera. It's been a buzzword for two decades, seeing as Millennials are now years 40 old.
The reason I bring it up is [because Millennials] are the decision makers that we have to sell into, whether it's myself at Roots, whether it's one of you guys at a TPA premium audit… we're all selling into Millennials. These Millennials require that combination of expediency, but personalization, but then we also have to lower costs, which means capital investment.
So that’s stat number one. The same study found that 44% of the decision-makers in business are Millennials, and 33% are at least influencers in the purchasing process. And again, with rates being low, it puts us in a pickle.
With that being said, for those of you guys that don't know who Ian is; Ian is now a FinTech investor at an organization, a fund called Vestigo. Before that, he was the CMO at Mass Mutual. He has run several insurance organizations, primarily – Ian, correct me if I'm wrong – but on the carrier side. So, we figured nobody better than somebody like Ian, who is now dealing in FinTech and automation to give an unbiased perspective on these things. First of all, Ian, in marketing, and especially at your time at Mass Mutual, how do you go about delighting customers? What does that even mean to you?
Ian Sheridan: Let me come back to that question.
First, my background: like all of you, I'm 35 years in industry in this space between asset management, insurance, and technology. I'm on Life 2.0 where, after a 35-year career, I decided to move into venture. And so, I offer this because the lens of information I'll give you is really from this new venture lens.
We are an early stage FinTech investor. So, that means we use terminology like the C to A Round (but essentially what it means is companies that are pre-revenue), but we can see a line of sight to about a million [dollars] ARR within six to eight months. And so, as operators, we come out to bring our experience and lens, like partners who are entrepreneurs, operators and data folks.
What makes us unique – and again, this is important for the lens of information I'll give you – is we do work with big data, about 16 petabytes of consumer clickstream data and some advanced algorithms that allow us to predict the probability of any URL going viral. From there, we use machine learning techniques to try to identify these early-stage companies to start a conversation that might lead to investment.
So, we are using big data to identify, underwrite and empower portfolio companies. And, under the operator's lens, I've been a CMO, I have been a line CEO in different aspects and have always worked with data. As it relates to automation, throughout my career -- whether it was at ADP back in the day of processing massive amounts, which at one point was 60-70% of the US payroll, think of the largest supplier revenue to the US government – or whether it was companies like Mass Mutual, processing Insurance and retirement plan-type automation. At DST, I worked with one of the largest automation systems and – Dennis, you know this, the AWD that moves, I don't even [know], is it 70% or 80% of asset management cashflows?
I've been around the space for some time. I've had success. I've had failure. What's different today is we are getting to the spot where computer power and data fluidity is now present. And so, when we were looking into our FinTech pieces, artificial intelligence, machine learning, cognitive processing rose to the top of our portfolio thought process.
As it related to Roots, when it was first surfaced to us, I actually came at it with a bias of been there, done that, doesn't really work. [The automation industry] had some breakages, and there's breakage because of culture, there's breakage because the data wasn't there, there was breakage because the technology wasn't there. So, [prior to Roots], we would get systems started, but we'd also run into lots of friction points. So when first presented with the opportunity to do due diligence on Roots, my view was, “This probably isn't going to work out.” And then that's the lens of the operator.
The lens of the data that we use tells a very different story, that more and more we're seeing artificial intelligence [and] machine learning propagate through all sectors of the industry. We can talk more about the future work as it relates to that.
What turned me on to Roots, not only the founders who are executives and operating people, is the [business] model addressed all the failure points of previous systems. And I don't want to turn this into a Roots sales pitch, you can stop me there.
That's the lens in which I look at this conversation, is one of being an operator, like all of you, but also now spending... we started our first fund about four years ago, five years ago [and] raised 60 million, we put about 5 million of our own money into it. Last year we raised our second fund, a hundred million, it'll close up about 120 million. We've invested in 24 early-stage companies, Roots being one of them. We've looked at 2500 companies through our data and our operators lens and our actual machine learning has looked at over… I think it's 10 million URL sites to get to … it's actually three companies that report purely sourced by data. So that's the lens. So hopefully that's helpful for background.
Now, Stephen, what's the first question you want us to talk about?
Stephen Lowisz: First question. What exactly do you mean by ‘delighting customers’? And I will attribute the title of this Roundtable to you. ‘Delight’ was your word. So, what do you mean?
Ian Sheridan: Yeah, it started a long time ago as we started to embrace technology and take out friction points. My space was asset management and insurance, but really, this is true across the board. There was a Harvard business case maybe 10-15 years ago that talked about designing and delighting systems for customers. And essentially what that really meant was looking at the customer value chain from the creation of the product or service, and ultimately to the delivery to the marketplace place.
Then we would look at all those particular handoffs in the operations and say, how do we make this frictionless and design it in a way that it's easy to use? Then behavioral economics, behavioral finance really came into play for me about, 15-20 years ago. In fact, I have some patents in that space in technology and the understanding of human decision-making and cognitive thinking.
And so, for the very first time, we were able to bring in behavioral finance techniques and understanding into the ‘design and delight’ conversation with customers. But again, it starts with understanding the customer value chain, the creation of a product or service, all the elements down to the delivery, and then the consistent reminder of the customer. What has transformed, then, is that we think about the technology delivery systems through that value chain.
What's new and different is there's all these data points that essentially need to be collected [and] put together in a way to deliver the product or service. But we didn't necessarily think about it in the way that we can today, which is combining the behavioral element, making things easy, frictionless, and designing in a way that, as we build the product or service through that value chain and the collection of data, we can actually not only design systems that are easy to use -- I mean, we’re all using an iPhone and sliding a finger back and forth to make things happen – but we can go all the way to improving the outcome and the experience. And so, when we talk about that design-delight aspect, it really is foundationally at the operation level. Then it’s looking for the opportunities to reduce friction and then the new items that come out; I call that the investible moment as an investor today. As these new components come together, [all of this] new data output, we ultimately can improve an outcome. Does that make sense?
Stephen Lowisz: So again, I want to be clear. This is a Roundtable. It's really a three-by-three square on my screen, but please ask questions.
Go ahead, Dennis.
Participant: Yeah, I was going to ask Ian [this]. So he's talking about the client experience, the journey, [how] we're digitizing, we're putting automation in, but we're – and I work with mainly the wealth management side of [this], though I tap into the asset management – I hear a lot of it, but where are we in this really streamlining and in this delight phase?
I look outside of financial services and I think there's a lot of innovation. You get the financial services and, I’ve mentioned this in the past… The Death of a Salesman, right? It's a very much sales culture still, right? This ‘client first’ is still really new to this industry. So are we … is it still first inning? Is it second inning for wealth management?
Ian Sheridan: Dennis, from what I see, I'm at the tip of the spear as it relates. I call myself a farmer of the in the soil of the ecosystem of a startup – Michael, you're dealing with startup companies. So, I think you can understand what I'm saying.
This is where ideas are being formed and cultivated. And I get to say, “Wow,” three to four times a day, generally, because I'm seeing this new innovation take place. And so I don't know if it's the first inning or the third inning – but the race has started, the computing power is in place, [and] the data is in place for transformation.
What I would argue today – and I participate with MIT and the World Congress talking about the future of work – the last two Davos as I was there, and closed-door sessions talking about AI on a global basis and what is the impact of the global workforce.
I think we're much farther along than most people understand. And here's how I'll make the argument. GPT-3. The open AI sourcecode, was really GPT-2, was released a couple of years back. GPT-3 last year released. On our way to Davos, I’ll use this example, using one of my partners or advisors to the fund. She worked at DARPA. She actually developed the listening device that the NSA uses, that we all use now on our iPhones.
She was playing around with the source code, had just been one of the early folks to use it. And [she] said, “Ian, watch this”. As we're sitting in the Boston Logan Airport on our way to Zurich, she says, “I'm going to write a book using GPT-3.”
So, she launches the code. She sets a few parameters. And what the story is, is some kind of romantic space novel. I don't remember all of the details, but by the time we landed in Zurich, the book was written. It's now published and she's selling it on Amazon. It’s just as a kind of a kicks and giggles thing.
But my point is, is that compute power that restricted us previously is now there. Dennis, in our world, we couldn't make money on trust and custody systems on balances less than $50,000 back in the day. And that was one of my unit cost challenges.
We’re talking about the democratization of financial services. The barriers 20 years ago, 30 years ago, where the systems could not get the profitability where we needed to them be. So we went where the money was and a lot of people were disenfranchised on them.
Today, because of technology, we can actually work with folks who are falling below the saving and investing loan and using artificial intelligence machine learning to create efficiencies and fluidity to bring these people back into the system through pure chat. Chatbots versus humans. I just threw a lot at you, but this is a place that I spend a lot of time in.
And so GPT-3 is now out into the atmosphere. People are beginning to use it. That's one aspect. Artificial intelligence and what makes it unique now is not the algorithms. In fact, the mathematical models that we use at Vestigo, six different models. They’re models that were developed in the sixties and seventies. We fine tune them, but ultimately they're now capable of doing a lot more because the compute power is there and the biggest transformational component of AI is the dataset is there.
GPT-3 has read the entire internet, which is what it claims it has done. And now, privately, it's being set against the STEM databases – Microsoft, Amazon, and engineering and science databases.
In fact, the reason the vaccine moved as fast as it has is [because] it's a combination of machine learning and AI with doctors, with real humans working forward. In fact, that's the segue into why I'm excited as an investor about Roots and cognitive processing, because it's… we're entering this phase where artificial intelligence is very real. Machine learning is very real. And now you can enhance the human in ways we couldn't do before.
And so, as an investor, this space of empowering human thought process and execution exists when it didn't before. And I'll stop here in a second. Twenty years ago, I couldn't even do anything in the cloud as a financial services executive for fear of [government regulations]. Bob, you'll appreciate this, I've worked with government regulators and thank you for your service, but I couldn't do anything with the government, with a government contract, if I was on the cloud because of PPI concerns and everything else. I'll stop there. I threw a lot out there.
Participant: I wanted to pick up – this might get really into the weeds right away – but the combination of your comment about writing the book and the cognitive learning capabilities of it.
So, in thinking of variables in insurance policy and doing it in a way to enhance a claims experience, are you suggesting that the knowledge or capability is there, that the machine could read policies and understand the differences between deductibles or coverage types or things like that, and be able to make initial recommendations on claims? Is that … is that a practical application of what you were describing?
Ian Sheridan: It is. It comes down to the training, set the data set and giving the computers the time to do what they need to do, and the human oversight.
So even in the case of the book, the output needed to be edited. There were some weird kind of tangents went on. And I'm not arguing GPT-3 is the future of AI. Don't misinterpret. That's just one aspect.
Machine learning and the algorithms that we use against, more importantly, the datasets that are available to us. And it used to take massive amounts of data to train. Like I work with 16 petabytes of data – clickstream data and unique panel – but really, the training set over the years for what we need has gotten smaller and smaller.
What I've learned at each cycle – and we run these algorithms every two weeks against the North American data set to find URLs that are going to be interesting to us. Then, what I've learned is each period, we take them down like Indy cars, we get the output, and then we score [them]. More of this, less of this, [this] was wrong, this was right. We use five different models so that we get different scoring.
So, if all five models say ‘this was right,’ it scores higher. If one model or three models score yes, then it falls into a different category. So I've lived it personally, in the venture world, using consumer clickstream data in North America. (25:55)
As an employee when I was in corporate, I also saw it. But we had limitations because, one, our iron was antiquated, our mainframes restricted us, and the fluidity of data just across -- I'll use an example of Mass Mutual. When I was the CFO in the retirement business, I went to my counterparts across divisions and said I'd like to mash all our data together: life insurance, disability, we owned Oppenheimer funds at the time bearing asset management… everything under this tent, I want to mash together, and I want to learn from it.
In those days, we used Siebel analytics. It's now called Oracle Level. And what came out of that mashup was that we quickly understood that 60% of clients across divisions had three or four products or services from Mass Mutual or its affiliated companies. And yet the experience – to come back to the design and delight – there was no connective tissue whatsoever. There was no interrelationship that allowed us to leverage them.
So, there wasn't a delight experience. There were these myopic approaches to each customer, and bifurcated from what, ultimately we were trying to do, which is grow our business. And you're growing your business with your customers by delighting them and reminding them on a regular cadence that they made a smart decision to do business with you.
So when I see AI today, whether it's in accounting – and Bob, I see a lot of accounting applications as well – what has been amazing is that the AI can complement the human accountant or human finance person to say, “This is the moment in time when you now need to talk to this client.” The client’s son or daughter just turned 16 and is driving a car. What are you doing? Talking to them about the automobile insurance policy, right? Time- or event-driven AI is becoming really good.
Stephen Lowisz: Doesn't AI and these large tech implementations that we're talking about – and this isn't a leading question, let's take Roots out of it completely, I want this to be unbiased. Doesn't digital transformation and implementing AI and ML mean that you have a massive cash outlay, which is very difficult in a time like this? When interest rates are low, costs are rising. What's your take, Ian?
Ian Sheridan: Listen, we're all driven by it, and I was always a unit-cost trained manager back at ADP. We measured every unit of work before we actually delivered a product or service and if we couldn't prove consistent, steady profit… we didn't like spikes. If we can consistently create profitability than we were happy.
The idea of ‘it's too expensive to modernize’ is, I think, the wakeup call. [What] COVID taught us is that really there's a convergence of a couple of things. We have low interest rates and we can expect low interest rates for a while. So that puts tremendous pressure on the general investment accounts of these insurance carriers. And so, they're going to have to find unit cost reduction.
And so, when I was doing due diligence on Roots – and again, I've seen this playbook from other workflow systems and where they break and where they're not consistent – but what was different through COVID. And remember, I was fundraising a hundred million dollar fund in January, and I was also invested.
So, I was following C-suite of insurance, asset management, brokerage banking, et cetera. And all of them reported a couple of things early in through the COVID period, which is, “Hey, our disaster recovery programs worked brilliantly. We were able to get everybody working, distributed, and everything works great. Secondly, we've got to reduce expense, so I guarantee you, I'm not doing that real estate deal that I thought I was doing as it related to office expansion.”
And then the other part was real unit talk, real targets for unit cost reduction. So whether it was a Fortune 100, 500 or down into the smaller regional sectors of insurance, we heard consistently different bogeys of expense reduction. And so how do you do that?
And the way you do that, and this is if we talk to LIMRA, which will release their research at the end of this quarter – Dennis, this in your wheelhouse so you can analyze, do the evaluation on it – but essentially what it has come across as a result of all this convergence, low interest rates, distributed work, is massive targets to reduce expense.
Modernization for insurance. We've been a lagger on it for a really long time and we're at a point where a couple of big forces are against us. One, the talent that we need to make the transformation and to win, the next – you mentioned millennials – to win the next set of customers. We're unable to recruit that talent. If we're asking them to come in and program COBOL for Tran in Pascal, that's the world I came from.
So that's still running at the bottom of the tech stack. So that's still there. And then you want to bring in a sophisticated data scientist engineer coming out of school and say, by the way, we want you to work on COBOL? Really difficult.
I worked for a firm. I won't mention the name, but it was a heavy COBOL-driven company as well as its own proprietary code and when I came in to figure out how to bring these, I was looking at one aspect of the company, as we were preparing for an IPO, my job was to figure out in this portfolio what do we keep? What do we shut down? What do we modernize? How do we package this for the IPO?
What I quickly learned were the engineers who were focused on the front end systems and a little bit of the middleware had no idea what was going on at the bottom of the tech stack. And so I had to call 65, 70 year old retirees on the beach in Florida who knew COBOL to come back in. And two things I learned. One, they were not happy people because they were let go. So, they weren't treated well on the way out. And the only ones that would comeback were human resource risks. And so I had [HR] screaming, “I can't tell you why, but this guy can't be allowed in the building!”
So we have a modernization issue for real, that has to be addressed. And if we want to recruit talent from our best schools, we need to give them a reason for that. And I will tell you very differently, and I lecture at MIT and I work with students at WPI, Northeastern as part of our ecosystem development. These young students are very different than when we all went to school. They are not only informed. They are equipped to do the math, to do the work. That, in my day, when I got out of business school, I worked for Goldman Sachs. If there wasn't a copying machine and a gopher list of stuff, I probably wouldn't have had a job. It was my first six months to a year. I was beaten around and run around and making copies and delivering file balances that were, through these massive printers and politics… Those days were over.
And you've got to come into work today because of technology ready and able to go. And I think our universities, from what my experience has been, they're doing a good job. The insurance industry has to catch up through modernization of digital strategies and adopting these new technologies that will address the next worker. And the next worker is someone who is going to be skillset oriented, hyper-curious and forever learning.
That's how they'll survive in the AI world. It's not – my dad worked for one company his whole career, he was a Bell Labs engineer and I thought, getting patents was the way to go. It didn't make him wealthy, but he had a great career. One job. I took a different approach, focused on strategy and growth but the next generation of workers are going to have to consistently reengineer themselves against these changes that occur. And I would argue that machine learning and AI, the cognitive processing empowers them. So, it's an incredible time to enter the workforce and be part of this transformation that's happening
Stephen Lowisz: Any comments or questions? Oh, go ahead, Dennis.
Participant: Again, my observations have been, going into COVID… I think digitalization was on the mind, but [companies thought], “We can deal with that next year, or in two years.” And then, boom, everyone's having to play catch up. We are seeing those firms that really did make those investments early on to really be more nimble, to really become remote or to really have better client engagement capabilities. I think [those companies] are going to really accelerate coming out of it.
For instance, just prospecting in general, one of the things that we're seeing on the wealth management [side] is this real effort to say, we've really got to go out and go after new clients, and new business. And it seems to make sense, you've got people, either seeking advice for the first time or looking to switch advisors where it may be.
But the industry has been very rusty on this whole process. And when I talk to firms about how they are going to go out and do their prospecting, it's… It's this mix of, “Yeah, we've got some neat tools and maybe we can utilize Miro socialized program and grapevine six to help build out and build those interactions.” But for the most part, the advisors are [still talking about] “…when we get back face to face…” In my opinion, even if [face to face] comes back, it's never going to be the way it was. So, I see firms trying to figure out to exactly how far should they ramp this up.
It's the same with a wholesaler. I look at both those areas and the adviser in that face-to-face – I should say in-person – I just don't see that as being a staple, moving forward.
Stephen, you mentioned talking about Millennials in the positions, but I think every Boomer has been digitized in some degree. So, they're coming in with different expectations and there's no need to meet this often. And so you've got an industry that's trying to really do business development and start engaging clients at the same time, they've got to do it with thinner margins because they're not just engaging clients more proactively.
And by the way, fee-based business – while it's great, in many ways it has made a lot of advisors in firms complacent when it comes to new business. Everyone's saying we're going to try to capture clients at the early stages of the development, right? So they're further down-market to the workplace, at Fidelity or where you could be Robinhood. Everyone's trying new things out, [asking] “How do we get them hooked in?” And then we can start upselling them. But again I hear firms say, yeah, it all makes sense, but then [they] fall back to these old ways, [like] “Well, once the wholesalers get back out there and do things …”
I don't think it's going back to that. Or, again, firms are talking about [how] they may want to digitize, but their data is held in six different places at the bank. And this one [bank] doesn't talk to the other, and legal won't [help], there's a lots of firewalls, but then it sounds really great and everyone's moved this direction. There are a lot of roadblocks and I think part of it is just the way they're structured. They've got those underlying legacy tools and part of it is still culture. I still see that prevailing culture in some of that management.
I know, Stephen, you mentioned Millennials that need more managing director positions, but I still see pushback when we're talking about [digital transformation] and head shaking, so I'd be curious to see what other people think who've been in this business longer than I have. Are they seeing a change or resistance to change?
Ian Sheridan: You point out a couple of things out there. I'll tell you what I saw this week that just blew my mind. Cost per acquisition as an investor is a big deal, right? We want to know that you're bringing down customer acquisition costs. We want to know that you're creating a long-term value of that customer and the revenue stream.
We're looking at those things. We want to make sure that there's a true technology that empowers humans and creates the stickiness between us at the right times.
[We’re] not just thinking about wealth management, I sat with a company this week that is using the GPT-3 source code to write their own microwebsites. We were looking at this about six months ago, we call it internet tunneling. You can think of it today as maybe Google clickbait that's stuff that surrounds you when you're traveling the internet. But they were able to, over three months of development using the OPNET AI source code, to develop a program that would watch hundreds of thousands of micro websites targeted to advisors and clients. They, in fact, they launched 29,000 websites in 20 minutes, they were able to output… Are you all familiar with pay for placement on Google? You pay to be in the ranking. They were able to out-perform Google's algorithm through the launching of these micro-sites. Now the business, I'm not going to invest in right now, but that part of it was really exciting, the LTV (long-term value) was so good.
But, I think companies are just at a point where … I'll say it this way, when we started Vestigo and still today, our thesis was that anything we invest in, ultimately we'll have to partner with an institution or an incumbent. That's how I would say, right? So we're going to have to Charles Schwab Fidelity, Pershing, DST, Broadridge, senior down. We're going to have to acquire these companies and that's, what's going to make it, but we have to partner with, and [we’re] going to have to build multitenant, scalable systems. Like what Roots has created. And so, we thought that's how it's going to be. And I still believe that.
But what I'm starting to see, Dennis, to your point is particularly in banking, they're getting their asses kicked. Jamie Diamond said it great a couple of weeks ago. Could it be so they shouldn't be scared of what's happening. And I was saying that about a year and a half, two years ago, they were moving slow and I'm seeing these new FinTech companies, neo banks and others really stand up and do it smarter, faster, cheaper while designing and delighting their customers.
So, I think it's the incumbents who have big moats, and who have the capability to partner. They all have, I think, the forward ones who are thinking about tomorrow today have some development structure to get an ecosystem that I'm involved in because I see them, I engage with them on certain deals. I think more is moving in that way, but there's never been a more [applicable] time where compute power and data is available. That wasn't there a couple of years ago.
Participant: Can I ask a question? So, when I hear you talking, everything that you've said makes a lot of sense especially when you're talking about building known scale, right? I run the large BD. I've run a couple of large RAs. I can at least envision the scale. But now I'm going to ask a real practical question.
I apologize it's not as theoretical, but I'm enmeshed in startups where both financial services, one is totally going to be retail-oriented and they’re built upon these… as you can talk about Broadridge Fidelity, there's these common structures that sit underneath them. And my struggle, from a practical perspective, is I don't have the scale to do that.
And partnering with someone who did, that would be great, and [especially if they have] the underlying data. But to enhance the client experience, is it practical to project what volume would be? Or is this something that you see at this point, an introduction to AI, that you need to scale it first to be able to afford to develop the all the tools? I see I need the tools. One of ours is an option trading online and it's like a Robinhood, but wanting to put box in there makes a lot of sense. But I don't understand how to get that started from a real perspective, a practical operator’s perspective. Does that make sense?
Ian Sheridan: Yeah, I don't think there's an easy answer for that. It's going to depend on the problem you're trying to solve. And then the size of the problem. And then what's available in the data around and what the breakaway moments usually are…
When you find these entrepreneurs, I see them as artists or conductors of music in the sense that they're able to bring and weave things together in ways that the average person isn't thinking through. So, they're able to get to the scale question, [because they’re] prompted by scale all the time.
When I was in industry, then it was this overriding barrier that prevented innovation. And what I think is different today, is that the available solutions from the technology spectrum are much broader and the way you can bring data together today is also very different. And Dennis you talked about myopic systems within architecture, and I certainly lived in that, and then what was our answer? “We'll create a data lake and we'll take off with junk from that pile and we'll put it into this pile.”
We still ended up with a polluted lake. In fact, now that we've gone to AWS and cloud services, overall, a lot of the on-prem issues we had, whether it's for cyber or other interrelated or infrastructure issues are now showing up in the cloud as well. So I don't think there's an easy answer, Michael, but I do think that's the breakaway moment; is that when someone is able to see how elements can come together that not only create a better profitable margin, a lower friction environment, that's where it happens.
Participant: No, and I get it and I quite honestly, I wasn't expecting I wasn't expecting [a magic solution], but we're at the client double rate where it's starting to look like, ‘Oh, this makes a lot of sense,’ and yet I'm thinking that because of what it's based on the structure it's based on, I'm still having that argument. Anyway, I was just looking for some intuitiveness. So thank you so much.
Ian Sheridan: I just look at this screen right now, the diversity from work experience that you all have… I realized like there are elements where all this can come together and create a whole different type of outcome.
Accounting is another area where, not only does cognitive processing make a difference, but the normalization of financial data for consumption is a big thing. We've invested in that space as well. But that space alone without the processing element is limited.
So here I've got, a portfolio of companies, like twenty in the first [round of funding], four started it before and are in the second fund, that will be another 20 when we're done. A lot of people will sit in and be like wait a minute, have you ever tried to weave all these together? We're not focused on that, but if they decided to work on a break. But, you do see these opportunities in these new startups where and frankly, when I was at ADP and we would have this massive conference for all of the ADP businesses were together, ADP brought up brokerage, which is now Broadridge and all kinds of companies.
And I would look at, across the floor as this corporate strategy guy. And I'd say, man, if you could just get this stuff connected, we would just dominate in the market. We couldn't do it. We didn't do it. There was a reason structurally the company couldn’t operate at that level. But you see those opportunities now today that I don't think we saw before.
I'll use another example. When we started Vestigo, we looked into the FinTech world. We said, ‘Let me see what other VC firms are out there.’ And there's a lot of venture capital firms in here in Boston, some very well-established great venture capital firms. As we looked at where they were investing in FinTech, as we called it, it was all payments. There was very little into what we do in the enterprise. What I call work site in the wealth space. The focus funds. In fact, we did the analysis and I have a slide from our friend, Rick Cares, but how much was actually invested into FinTech payments versus those categories I mentioned was dramatically different.
And so now what we see is all those wonderful payment systems and rails that were created, there's not enough room for all of them. Now they're connecting into the bank systems or the asset management systems. Or other systems of the economy and that's proved to be really interesting to watch.
Stephen Lowisz: John, were you going to say something?
Participant: A deployment question, really, maybe to tap into your insight. Ian, when you're looking at a startup and you're looking at that whole customer value chain that no doubt has multiple friction points within it, how would you prioritize? What's the logic you’re using in prioritizing where to start? How would you consult with these startups that you work with?
Ian Sheridan: Yeah, that’s a good question.
The way I would do it as an operator versus as an investor would probably be different. Because when I wrote corporate strategy plans, it was all about driving. There were usually… I was trained in the McKinsey way. So three to five strategic options and I’d develop tactical plans under each, from current state to future state.
As an investor, we're looking to invest in companies that are going to not only just be profitable, but robust businesses that will be multitenant, and that will scale. So, it's not about profitability and burnout. It's about really building a strong infrastructure model. And so the starting point generally is driven by the total addressable market analysis and getting the concise ‘here's.’ Here’s where we can play and here's where we can win and over time. How do we think we can expand products and services to capture more market share.’ It also requires a deep understanding of the marketplace and who's winning and losing, and this is what we use data for, so that you can figure out where you want to attack.
I was a big fan of – Dennis you probably remember this book, Blue Ocean. I was a big fan of that book, I'll use that term every once in a while. I like to find blue oceans first and make sure that I can win in that space before all the fishermen show up and it turns into a bloodbath.
So as an investor, I think that way as an operator… generally it was the corporate initiatives that would come down from the C-suite. And it might be a reduction of unit costs as I picked up last year. So I have insurance carriers that I connect with, many or several are invested in both, fund one and fund two with us. So I get a really interesting line of sight there. And they'll have a big bogey of what they've got to do to get the bottom line where it needs to be.
My last days in corporate, John, were why I got out. It was, “Okay Ian, we want you to look at this, but we need you to get PFOS down to this level.” And it got to an exercise of human capital carve-outs and things like that. And it wasn't really fun. Now it's about what are the technologies that we can use to empower the customer experience and improve the margin and the outcome for both.
And for the first time, I would say the last six years doing this, I actually can see that what was difficult inside the company was that there's a combination of partners, and that they can bring to the table today that I couldn't back when I was inside corporate. Whether it's a company like Roots, where they can bring in a scalable bot process and we've seen in their model. The first bot wins, and it creates a return very quickly. And adoption grows to two, three, four [Bots]. I think, what's the largest customer, Stephen, eight bots at this point?
Stephen Lowisz: Something like that. Yeah.
Ian Sheridan: And when I was calling in, by the way, Chaz and John founded the company and said, these bots are like people interacting. So they start talking to their customers and they've actually named these things. And they treat them like employees.
And that was one of the issues that I had culturally. Whether I was trying to partner or innovate or drive automation, trying to keep – and Dennis, this goes to your earlier point – cultural alignment is really difficult.
One insurance carrier I worked for, I remember coming in as a corporate strategist, and I asked everyone across the lines, “Who's the customer?” And I ended up with 15 different definitions. How can that be? That's no wonder we're not at best profitability, when we're chasing different things and we're creating friction points [internally] that the market doesn't even care about. And we've created this internal function.
So, I think that's a long way of answering your question inside, it's what are the key drivers to, corporate strategy and growth that we're trying to achieve? But I think it's three big lenses. There's the worker's lens today. So, if I'm going to attract good talent, I got to have a really good story because the best talent… they're hard to get. When I was graduating business school, Goldman was a hard company to get to. But now Goldman is aggressively recruiting in the schools because nobody wants to go to Wall Street. They don't want to work with old tech.
So there's this worker's lens that I would argue bringing in modern solutions to empower the human worker to higher levels of engagement are really important.
Then there's the marketplace today. So how do I compete and win today? And that's going to drive certain technology adoption decisions. It might be, do I need to reduce unit costs so that I can invest over here, or do I need to bring in technology that reduces friction? How many times are we going to ask the customer to fill out their name and synoptic data processing experience? It's silly. [All the way] to future growth.
I looked at public information. If you look at Charles Schwab as an example pre-COVID, they were making an acquisition – Dennis, you're probably pretty familiar with that acquisition that they were making. A sort of substantial acquisition of a competitor. Then COVID hits, two things happen: trading volumes across North America, across the globe, double. So, all of a sudden their systems are pressed to the metal on American trade. And not just Reddit people, like more trading activity. Then you had your workers distributed [remotely]. So you had to manage your flows with a distributed workforce. And then by the way, you're doing an acquisition.
So I would argue, as we've talked about worker's lens marketplace, the future of growth, you've got to prepare your business for the next set of changes. And so, if you're stuck with infrastructure modernization challenges, you're not going to be ready for the opportunities that are going to present themselves. Every sector has consolidation events that happened. Right now, we've got the crazy world of SPAC going up. Everybody's getting SPAC. If you're not ready because you're not modernized, you're not taking out the things that are just driving your costs up, not delighting your employees or your customers. Get that stuff out of your way.
Stephen Lowisz: I do want to be conscious of the fact that it is the top of the hour. I know most of us have to go. So, a couple of things that I do want everybody to takeaway. First, there's going to be an email going out, if not today, then early tomorrow morning. We referenced a number of studies and there's a few more that I think you'll find interesting. My favorite of which, Gallup put out a study, I think it was about a year or so ago. They found that 85% of the workforce was disengaged and 18% of it was actively disengaged, meaning they're actively working against your company.
And interestingly enough, the number one reason [that employees are actively disengaged] was that they don't have meaningful, purposeful work. So it then starts another entire conversation, which is a conversation for another day. But as business leaders, do we have the moral and ethical responsibility of innovating and finding our people meaningful, purposeful work? Or is it only to give them gainful employment? Because those are two very different things at this point.
There's a couple of articles, things that I think you'll all find interesting. We've looked at all of your backgrounds. I think we found some commonalities that will be good conversation topics within your organization. With that being said, if you guys have any questions Ian, are we able, I didn't ask you this before. Are we able to share your email?
Ian Sheridan: Sure, sure. Cool.
Stephen Lowisz: Nothing like putting you on the spot here. But all, we're happy to answer any questions, give you thoughts. Our team has some insights that we can share. Ian certainly does, but I appreciate you all joining for the first Automation Trends Roundtable.
Thank you so much.
Ian Sheridan: Yeah. Thanks. Thanks for the insights. Take care, everybody.