Measuring AI ROI Through Expertise Compounding With Kantata

[00:00:00] Scale your business with agentic AI with limited risk. With Denodo's AI data layer, your agents are provided with real-time company data and guardrails for company protection. Create the business you always dreamed of with Denodo. And you can do that by simply visiting denodo.com to learn more.

[00:00:26] What if companies are measuring AI success by the wrong scoreboard? Saving employees a few hours, automating routine tasks and cutting costs might look good on their dashboard. But do we really need another dashboard? Well, my guest today is the CEO of a company called Cantata. And he's going to join me today in a conversation around a much bigger opportunity

[00:00:53] that is turning decades of experience in your enterprise, past projects, client conversations and hard-won lessons and put them all into intelligence that the entire business can use today. So we'll also look at the rise of the expertise economy, why professional services firms risk becoming indistinguishable if everyone uses AI simply to work faster.

[00:01:20] But most importantly, how companies can make every customer engagement smarter than the last. We've got a lot to talk about today. So enough for me. Let me introduce you to Michael right now. So thank you for joining me on the show today, Michael. Can you tell everyone listening a little about who you are and what you do? Hey, Neil. Great. Thanks for having me. So I'm the CEO of Kantata.

[00:01:46] We are a leading provider of professional services automation software. I've been leading this company for the last four years. But prior to that, I've had a number of roles leading various types of software companies, mainly for private equity-backed investment firms. And prior to that, I actually started my career as an electrical engineer working at NASA's Jet Propulsion Laboratory in California, but then quickly found my way into consulting, working at Deloitte and other firms.

[00:02:15] So privileged to have the leadership role I have today and excited to help transform this industry. Wow. That's an incredibly cool backstory you have there. And one of the things I was excited to talk with you about today is when we start thinking about AI in the workplace, many companies are feeling under increasing pressure to show return on investment for some of those expensive AI projects. But you've argued that measuring productivity gains and time saved almost misses the bigger opportunity here.

[00:02:45] So what are business leaders getting wrong about how they're measuring AI success? Because you can only improve what you measure, right? Yeah, 100%. And look, I think everybody's struggling to measure the impact of ROI today from AI. I think you look at the firms that are out there using it today. Everybody's using it at kind of what I consider to be this kind of level one use case, which is the chat, the assistant that all of us experience every single day.

[00:03:12] You know, and most folks are trying to graduate to kind of a level two use case where it's automating work and conducting tasks for them. And, you know, I think, you know, for us in this industry, I think that's going to be kind of the race to zero, right? Everybody's going to rapidly automate things are going to really, really struggle to define kind of what ROI they're getting. Or am I making an actual tangible business change because, you know, I'm using these chat assistants?

[00:03:36] And I think most customers we talk to, and even like our own firm, I think really struggle to reach that level of critical business decisions with it. And what I think really the next opportunity is, particularly for professional services, is, you know, using the technology that's available now to actually unlock a totally different use case for them.

[00:03:55] And if you look at the way, you know, services firms have run for years, it's all built upon, you know, tribal knowledge that is, you know, stuck inside these organizations, you know, a web of interconnected decisions that need to be made. It's almost like a dance that needs to be kind of, you know, choreographed in a certain way. And I think they need to think about how this tool is going to help them dance a better dance and not just kind of automate the existing one.

[00:04:22] And that's what we really see as the big opportunity for our customers out there today. And when I was doing a little research on you before you joined me today, I was reading how you described the real opportunity as the expertise economy. Tell me a little bit more about what you meant by that and why this ability to capture and scale expertise is becoming so valuable right now. Yeah. Yeah. So if you really, you really have to understand how a services firm operates and I'll give you kind of one like very simple example.

[00:04:50] You know, you, these firms that are running, you know, projects, whether it's an IT services firm or marketing agency and accounting and auditing agency, they create volumes and reams and reams of information and data and knowledge. And that is their intellectual property. That is their, that is their true product. It is not driving a billable hour or improving utilization, which I think it once was. And those metrics are still important.

[00:05:15] But I think the opportunity here is actually to use this technology to harness insights, information and context from all of the expertise that they're creating every single day on a daily basis. And really that, that, that, what we've, that's what we feel is the next rise and what we've kind of coined as this kind of expertise era and the expertise economy.

[00:05:36] And if you were to think of it in practice, it would be, you know, instead of walking into a project called or walking into new interaction, actually looking at this information that exists in your business today, which, you know, that information has been siloed. It has been stagnant and really not been able to be harnessed across their business to improve how they're operating. And now you can with this technology, you can look and find, you know, the last projects that were similar.

[00:06:00] You can connect yourselves to colleagues that might be located around the globe that have done projects just like this in the past. You can look at different ways to optimize the outcomes and work with your clients to deliver a better project for them. And really at the end of the day, what this helps them do is drive to another level of efficiency, but also differentiate their proposition. I think right now everybody's going to race to zero and try and become more efficient. But how are you actually delivering a better, more predictable outcome to your customer?

[00:06:28] And I think everybody's focused on how you actually measure outcomes. And this is actually a technology now that can help our customers do that. And if we were to take a look inside an organization on either side of the pond, I suspect that every company has institutional knowledge buried in experienced employees, heads, past projects, client conversations, and people that have since gone on to other things and lessons learned exercises.

[00:06:54] So how can AI better turn that knowledge into something that the wider business can actually use? Anything you're seeing here? Yeah. So I guess what we're seeing is really trying to focus on solving, you know, as you pointed out, Neil, the industry that is plagued by, you know, tons of turnover, tribal knowledge, pockets of knowledge, knowledge, you know, things laying kind of idly in the business.

[00:07:19] And really the way that we've leveraged the technology at Cantata to try and help our customers is to actually build a knowledge graph and a capability into our product that is able to consume all of this information and build context that is specific to their business.

[00:07:36] So conversations that might happen on Zoom calls, transcripts from interactions with customers, presentations, documents, knowledge, relationships in the firm that our product can see and exist. And it builds a context and a map of all that information so that it can actually present to them opportunities for them to improve the way in which they're actually conducting their work.

[00:08:01] And, you know, I think I reflect upon my really my own use of some of these AI tools and how instead of just being kind of a task or a tool to complete a task more quickly, it's actually something that has strategic context for what you're doing. And that is how we've seen folks use these tools. And you really have to experience it to really understand what it can actually do for you. And, you know, it could be a simple, like there's a tool in our product called an expertise agent, a super agent,

[00:08:29] where the agent can actually see and tell you what you should be doing for your business. So it doesn't prescribe an outcome. You can actually ask it, hey, what are the ways in which you can improve my business? And it has context for your business and will actually suggest things to you based on the knowledge and the information that it sees in your business, as well as and not just the qualitative information like documents and transcripts and recordings, but the quantitative information, the business metrics, your utilization rates, your hourly billing rate,

[00:08:59] how staffed you are, how idle some of your team members are, which team members are on the bench, which clients you're working with, what industries are in, and how you can actually connect to counterparts in your business. And it really is just another paradigm of the way that folks can make intelligent decisions about not just how they do their work, but how well they do their work and what sort of outcome they're delivering to their client.

[00:09:22] And one of the reasons I was excited to get you on here today is you introduced the idea of the expertise compounding rate. It's incredibly cool, but just to bring that to life, how would an organization measure that in practice? And what would tell a CEO that their company's expertise is genuinely becoming more valuable over time? Because again, it feels like a real opportunity here. Yeah, I think firms have to look beyond the traditional productivity metrics.

[00:09:50] And historically, these are things like margin, utilization rate, billable hours. And those are obviously still very fundamental to driving the metrics of the business. And we're not suggesting that those are unimportant. What we are suggesting is that there's a new level of metric that is important. And really to look about how you're driving revenue growth, how you're creating new opportunity from these tools.

[00:10:18] I talked to one client recently where they've started to obviously leverage these tools to drive revenue growth. And they're certainly driving revenue growth, but they're doing it without actually adding a whole lot of headcount. And I think in the past, this business model in this industry has been all about adding an individual or a team member, driving a certain level of margin or utilization times a billable hourly rate. And the framework of growth for the services business has really fundamentally changed. It is no longer that.

[00:10:48] And I think right now the base level firms are trying to drive growth without adding headcount. And many of them are doing that. They've figured out ways to get more efficient on a daily basis using these tools at that kind of level one surface kind of assistant level. And the next horizon is really how they're actually using it to unlock new business potential. And I can give you some examples.

[00:11:09] So this one customer in particular was using it for ways to find a moment of price elasticity. So AI has allowed them to deliver projects more efficiently. There's no doubt about that, whether it be writing code or using it to automate document creation or things like that. Very basic use cases. But what they've discovered is that they can price projects differently. And a project that in the past, a client would have not proceeded with.

[00:11:38] The client might have said, you know what, there isn't a business case for this investment. It's too expensive. It's going to cost me too much to do this. And I'm going to defer this piece of work to the future and live with the current system. It has allowed this one client in particular to find these points of elasticity where they are actually able to price projects differently and to achieve revenue growth. And now their ability to actually track and measure this is questionable, but they know it is happening. They are pricing these projects differently.

[00:12:06] They are achieving revenue growth by being able to take on projects that they wouldn't have in the past. So what's the business metric there? It could be opportunity close time. It could be how many opportunities, sales opportunities or project opportunities are you moving from something that was closed loss to now a closed one opportunity from a services standpoint. So these are the metrics that I think people are starting to think about. The other one that we've talked to customers about is really if you think about the idea of a resource.

[00:12:36] I think in the past, a resource was a person for the most part. And now it's a person plus a piece of technology. Or if you really want to get to a very granular level, it could be the price of a token that you consume from one of these AI models. And how do you actually start to implement their own hybrid pricing model?

[00:12:59] And I think this is an area where I've not seen a client yet be able to do this, where they're looking intelligently about how they're pricing these projects. What's the composition of people versus technology and resources and tokens? And what does that mean for the overall evolution of their business? Could that be a metric of success for them? Like how much of a project is actually processed and done in a hybrid model?

[00:13:27] And we can get away from perhaps actually prescribing a certain level at this point. But they can't even measure that today. And those are the sorts of problems that we're focused on trying to help them solve. And right now, I think every customer that I've met with is they're using these tools. They're sweeping all the notional costs of them into a bucket and then deciding what they do with them later. And there has to be a better way. We know there is a better way. And we're determined to help them solve that. 100% with you there.

[00:13:54] And if we look at the global trends, professional services firms are increasingly adopting AI at a rapid rate. Boom! Revenue growth has slowed and confidence in real-time visibility has declined in recent months. So what does that disconnect tell us about the way that companies have been implementing AI in the workplace? Any trends you're seeing here? Yeah. So I think I just shared one that I saw with a particular customer.

[00:14:22] That was certainly kind of a trend that is not isolated to that one customer. We've certainly seen that elsewhere. I think the other, you know, if you were to kind of zoom out from this over even a – and it's hard to zoom out, right? You know, people can't give roadmaps now that are more than kind of six or 12 months because things are changing so quickly.

[00:14:41] But if you were to actually zoom out and look over, you know, a multi-year period, you know, I believe passionately that this is just the next wave of technology that's going to unlock the next wave of economic growth for everybody, right? Not just for, you know, professional services firms and software firms alike. But you just have to embrace it. And I think there's this moment of paranoia that folks have when they're so close to it and they're looking at the page from just a few inches away.

[00:15:09] And I think if you'd really back up, this is a moment. And it's a moment I think that all of us have to capture and to be balanced about and to invest in and make sure that we're looking at this not as some kind of cataclysmic event. No, this is a major opportunity for, I think, for everybody to truly invest and wrap their arms around this technology and figure out how it creates economic value for everybody. And it undoubtedly will.

[00:15:34] So, you know, I think we do see services firms under pressure, right? And I think a lot of that has to do with this moment where things are evolving so quickly that folks are just being a little bit more patient to make some decisions, right? And I think, you know, there's probably not a firm or a customer that we've seen that isn't taking some extra time to consider really what the true long-term implications might be. They can't necessarily predict them.

[00:16:00] But with a space that's evolving so quickly, I think they're taking a little bit more time to evaluate how they take on these projects, what is truly important. Am I investing in the right technology long-term? Who are the winners going to be? So I do think that that's creating some more complicated decision-making for firms to make. But certainly I'm super optimistic about the opportunity. I think there's a world of opportunity there. And I think if this industry has proven anything, it is the industry that knows how to adapt.

[00:16:30] And without a doubt, they're going to adapt. And I also think there's a huge opportunity for change right now because many organizations will have workflows and processes that were just built for an entirely different time, a more analog time, not an AI world where we find ourselves now. And, of course, there's this risk that businesses could simply use AI to automate existing processes and scale inefficient ways of working, which is not ideal.

[00:16:58] So how should leaders maybe identify where AI can improve decisions, project outcomes, and customer values rather than just making people work faster or exacerbating the old ways of working? Quite a balancing act, I would imagine. Quite overwhelming for a lot of leaders. Yeah. And I think what firms have to do is really focus on what the next level of AI maturity means to their business. Right?

[00:17:26] And make it something tailored and specific to the business. What are their actual goals? What do they hope to achieve? And, you know, level one is probably that level of, you know, improvement from automation of existing tasks. That is table stakes. I think you have to think beyond that and you have to get beyond that and really define what that level of interconnected systems mean to your business. So that sharing intelligence, streamlining workflows, you know, operating in a unified way.

[00:17:56] So it's not necessarily kind of having more agents that just automate existing work, but actually having things that help you be more intelligent. And certainly our point of view on this is kind of this idea of this super agent that we've launched where it's not kind of prescribing to you a level of fixed agents and a catalog of agents. It's actually giving you an intelligent companion that really doesn't have a limit in terms of what it can do.

[00:18:21] We're not coming out of the box with, hey, here's a dozen agents or two dozen agents that do specific tasks for you. It's actually a strategic companion that understands the context in your business that you can interact with to create your own agents. Right? To do that within our product. To have our agents actually connect to ones that you're building with in other places in your corporate enterprise. I think every customer has said this loud and clear to us. They have a corporate strategy related to AI.

[00:18:48] And as a provider of software to these firms, we have to work with it. Right? We have to be able to connect to it, to communicate with it, to reduce the, as much as we possibly can, kind of the intricacies of how we hand off, you know, these requests from one system to another. So that we're, you know, offering a level of context to their business.

[00:19:11] And, you know, I think we believe passionately that like the deep expertise and this idea of kind of industry specific context is fundamentally important as opposed to kind of a generic horizontal AI use case. And I think, and I don't think that's a controversial point. I think everybody believes that.

[00:19:28] And folks can experience it, you know, when they go out and they use these tools and see kind of what it can do for them generically compared to kind of a vertically focused provider that is actually tailoring the tool in a certain way, training the tool in a certain way, you know, giving it the right nomenclature. And also having kind of a, you know, a veritable, you know, encyclopedia of past history to help train it. Right?

[00:19:51] I think this is an area where, you know, being an established provider, I think, is a major advantage to have that sort of information and context to help build and train these models.

[00:20:03] And for any professional services leaders that could be listening to our conversation today, any practical steps that you'd advise that they take to start capturing that institutional knowledge and making expertise available across the business and also measuring whether AI is creating a sustainable commercial value? Any tips or advice just for anyone wanting to get started there? Yeah, absolutely. So I think step one is to really define the metrics of success for your business.

[00:20:33] And I can tell you what it isn't. Right? It is not simply doing the same work faster. Right? That is the fastest way to create uniformly and a lack of differentiation amongst your competitors. Right? It doesn't mean that that's not important, but that is not the metric of success. That is kind of the price of admission in this economy right now. So that is something you have to do. But that's not the metric that you should focus on.

[00:20:59] What you should really focus on, and this is what we believe as a provider to thousands of customers throughout the world, it's about how you actually make every engagement smarter and more intelligent. Right? At the end of the day, that is what your customer is going to care about. Your customer doesn't care that you're doing a piece of work faster and it allows you to give them a better price. Tell them how you're actually delivering more predictably. How are you actually improving the outcome? Can you improve time to value?

[00:21:26] Can you showcase knowledge within your business so that it's more visible to me as a customer so I can see that, yes, this is truly what differentiates you. And how you actually use the tool to unlock that knowledge and that potential that's in your business. Right? And, you know, the client doesn't see the invisible task that's done automatically. They see the outcome and the deliverables that you provide them, which at the end of the day is hopefully a successful project. And at some point, that is what you need to showcase to your customer.

[00:21:56] And, you know, I think we believe that the way to do that is to actually activate the intelligent that's within your business. You know, learn from the expertise, compound it, and potentially show your customer how you're actually compounding for them. Right? Much of this business is repeat business in this world. You get a client and hopefully that is, you know, a relationship that lasts for, you know, a decade or more when it comes to the work that you're doing for them. And you're building strategic context for their business.

[00:22:25] Show them that context. Right? And that's what our product can do. It can show them the context of their business and the deliverables that you create for them, the interactions you're creating for them, the outputs, the documents you create for them. And our tool can do all of that. And that is what's going to truly differentiate them. Not the lowest price or not the best agent that has automated a piece of work inside your business. And that's my advice to everybody. Right?

[00:22:50] And I think, you know, I have the privilege of meeting with lots and lots of customers on a repeated basis. And, you know, this is, I think the comfort in this, I think, is that every customer is facing similar sorts of challenges and is in a similar place. You know, I think, you know, we haven't seen somebody who's infinitely ahead and, you know, we haven't seen somebody who's infinitely behind. Everybody's kind of in the same place.

[00:23:13] And, you know, I just keep, if I could do one thing, it's to share that information, share the insights with every customer and meet them where they are on their journey. Right? And try and show them what the person who's ahead of them has done. Right? And that's the best we can do and try and bring every single client and every single customer forward on their journey. And for anybody listening that's discovering Cantata for the very first time, maybe just expand on that.

[00:23:41] Tell them a little bit more about the company, the kind of problems that you're hearing and helping professional service firms solve there. And also how you're seeing AI changing the role that your platform plays for customers as well. What are you seeing? Yeah. So, yeah, I got to tell you another story here. The question sparked a, you know, like an anecdote from memory. And it was, you know, it was the CEO of a firm that I met with probably within the last three or four months.

[00:24:09] And, you know, and we were talking about the bit and it was a current customer of ours. And they were facing a lot of challenges. And this is a very, very successful firm that focused on security services. And he said, he said, look, I'm running this business today. And he gestured to the folks that are outside his office. I've got thousands and thousands of employees that are out there. And he goes, look, my biggest customer, right?

[00:24:36] I can't tell you if the work we're doing for them is even profitable. Right? And he said, do you know what? He said, I don't care. He said, I know when I start using your tool, you're going to help me fix that. Right? And that will get fixed. He said, what I care about is that this technology is going to fundamentally change the way I operate my business. The metrics that I use today are going to be different. People's jobs are going to be different.

[00:25:03] And he looked and he said, I need you to help me transform the way I operate my business. Right? And, you know, for me, that was like a crystallizing moment where, you know, it was no clearer to me in that conversation that, you know, the metrics of the past are, you know, they're not unimportant, but they're not the ones that are going to dictate success or failure in the future.

[00:25:24] And, you know, and this is a CEO of a really successful firm saying, you know, that, yes, those are important, but I don't care about those because I've got bigger things to worry about. And that they're all struggling and they all don't know the answer. And that we, Cantata, is in this position of privilege where we have the benefit of working with thousands of customers and that our role in this space is not just to deliver a piece of software to them.

[00:25:50] It's to deliver a piece of software and share our expertise, share our knowledge with them to put that in the software and to actually help them transform the way they're operating to a future business model. And I'm not going to sit here and tell you, I know the, you know, what the 18 or 24 month or 36 month answer is to that. And, you know, I think this is like a living, breathing thing that evolves every single day and every single week.

[00:26:15] But what I can tell you is that as a firm, like that is what we were spending all of our time on and to do it in a way that is honorable with high integrity and transparency with our customers and to share these insights as soon as we get them. Because I think everybody's in this journey together. And I think that is the role that we're trying to play as Cantata.

[00:26:36] And along the way, we're going to deliver you an awesome product that takes all those insights and gives them to you in a package way that we see is the way to move forward. And that's based on not one interaction, but thousands and thousands of interactions that we have every single day. Well, as companies scramble to justify those AI investments, improve return on investment. I love the light bulb moment you've delivered today that they could be measuring the wrong thing there.

[00:27:04] But for anyone listening wants to dig a little bit deeper on this, find out more about you, your work, et cetera. Where would you like me to point them? Yeah. So, you know, we'd love every conversation we could have with a prospective, you know, customer services firm. I think, you know, obviously the first place is to go to our website, cantata.com. We've got tons of material and services there. We've got an insightful blog and podcast that we do on a repeated basis that shares these insights.

[00:27:31] And for us, I think it's to learn a little bit more about what we are doing for our customers. So we've recently launched a tool called the Cantata Expertise Engine. And this is, you know, all about understanding how each firm operates, how we build contacts to help them with, you know, challenges like staffing, delivery, forecasting, financial management, and all the things we've talked about today. And it's real and it's here. And I'd encourage everybody who wants to learn a little bit more to double click on that, to read it.

[00:28:00] I think what we're doing is truly, truly different. We're taking a slightly different approach from what you might read in the headlines of the newspapers. And we know firsthand from our interactions with customers that it's resonating. So that would be great if folks could take a tour through our website and reach out if you feel like we can help you.

[00:28:19] And for anyone listening that have been listening to everything you said there and thinking about how AI success doesn't just hinge on productivity gains and wants to explore that ability to tap into their own expertise economy across the enterprise. I will add links to everything that you mentioned there, Michael. So please go check those out, everyone. They'll be in the show notes and over on the blog post associated with this episode at Tech Talks Network.

[00:28:44] But more than anything, Michael, thank you for taking the time to come on here and bring all this to life in a language that everyone can understand. Really appreciate you, Tom. Neil, thank you so much for making the time. And it was great to connect and look forward to the future. One of the things I loved about today's conversation is it challenged that idea that the return on investment on every AI project should be measured by how many hours we save or how much money we save.

[00:29:11] Michael explained the real opportunity here is to capture what a business learns, make that expertise available to everyone and then use it to deliver better and more predictable outcomes for the customers. And I also loved his warning that using AI just to do the same work faster could set entire industries racing towards zero differentiation and everyone just looks exactly the same.

[00:29:37] Whereas the winners, they'll be the ones that turn experience into an asset, an asset that compounds every time they complete another project. So many big takeaways on this one. But I'd love to hear your thoughts. Is your company using AI in the workplace to make people faster or make the entire business smarter? Let me know. TechTalksNetwork.com. We have 4,000 interviews there.

[00:30:03] You can learn more about how to work with me, meet me at an event, or just send me a quick audio message. Whatever it is, I'm here for you. But that's it for today's episode. But I will be back real soon, and I look forward to speaking with you all again on the AI at Work podcast.