Data Hurdles

Data Hurdles: Matthew Cox
===

Chris Detzel: [00:00:00] Hello, data enthusiasts. This is Chris Detzel. And I'm Michael Burke. Welcome to Data Hurdles.~ We are your gateway into the intricate world of data for ai, machine learning. Big data and social justice intersect. So expect thought provoking discussions, captivating stories and insights from experts all across the industries as we explore the unexpected ways data.~

~Impacts our lives. So get ready to be informed, inspired, and excited about the future of data. So~ let's conquer these data hurdles together.

All welcome to another data hurdles. I'm Chris Detzel.

Mike Burke: I'm Michael Burke. How you doing, Chris?

Chris Detzel: Pretty good. How about you?

Mike Burke: Good. I know this is a data talk, but I wanna say that I was checking out your DFW Running group website and it is pretty amazing what you've been able to do with chat GPT over the course of the last couple weeks.

Like I feel every week it completely changes.

Chris Detzel: It's Claude by the way. Before, we'll go into that in a minute, but before we do, I wanna to introduce, Matthew Cox, he's the Chief Data Officer at Fusable. Matthew, how are you?

Matthew Cox: I'm great, thank you. Happy to be here today. Good to see you guys again.

Chris Detzel: Yeah, you too.

And Michael it's using Claude 3.7 now, so I don't use chat. GBT Claude is just known to help you code basic stuff. I'm now a coder, Matthew what do they call it now? A vibe Coder, is that right?

Mike Burke: Twitter? Yeah. Yeah. That's a new thing. Yeah. So

Chris Detzel: I don't know if that's a good thing. And [00:01:00] coders don't like that.

I posted something on one of these forums and they just tore me up. Anyways.

Matthew Cox: Hey, listen. The day you have a cycling group, Chris, we should talk 'cause that's, yeah, there you go. I thought about it. If bikes involved, then I'm as good as they.

Chris Detzel: That's interesting. So maybe we'll do that.

And I know you're a big cyclist, so today instead of talking about my website, let's talk about you, Matthew. Last time we spoke was all, it feels like a long time ago, especially on data hurdles. You've now moved positions. You're now the chief data officer. I. Fusable. How's that going?

Matthew Cox: It's going great.

It's been six months now and I'm really enjoying the journey that we're on. It's about a lot of transformation and enabling a set of capabilities for our customers that I think will really deliver a set of value that they've not seen before. So I'm really with the capabilities we have with the technology plan that we have in place, I think we're gonna see this company really do something amazing in the marketplace, and I'm excited to be here for the ride.

Chris Detzel: Tell us a little bit about Fusable, what they do before we just dive right in.

Matthew Cox: So Fusable Fusable actually [00:02:00] is a, we were formed from the divestiture from a company called Ramble Riley and a series of acquisitions that were made, product acquisitions, a company acquisitions.

We've really stood up the multitude of actually multiple independent product lines that we're seizing together, bringing together to deliver what we call vital data to services when in vital industries. And those vital industries would be agriculture, structure, trucking industrial but really we're trying to deliver a set of services that spanned.

Data analytics and digital, and I think that's one of the things that really drove my interest in Fusable was it wasn't just a data company, right? It's, we're actually bringing together full digital, con, core media brands. We have core media brands for agriculture, construction, and trucking.

We have digital services. So we can actually run campaigns or digital capabilities against all of our data. So the breadth that this company represents within the industry is really unique and amazing [00:03:00] as we look to deliver, solutions across what we term as these vital industries.

And so that's what's exciting to me is taking some really amazing independent products that have been brought together. And then looking at how do we create a unifying connected data strategy. That harnesses all of that media and digital knowledge together in one package, and then serve that up for amazing

Mike Burke: solutions to our customers.

So Cool. And I know that we've, we talked about this a little bit, these industries don't, it's not like the industries that you've worked in with Google and that I'm in right now at Databricks or ZoomInfo. These guys have very limited data sources, and this is a new frontier in many ways with some of these, the I think we call it like hard tech industries, but areas that don't have access to the same kind of data sources.

They're not using like the live ramps and the DMVs the same way that the type of data sources that feasible will provide. Is that right?

Matthew Cox: That, that is correct, and I think that's where there's really a need in these industries for a data partner and not really just a data partner, but a solution partner.

I think that's really what we're [00:04:00] serving up from a capability standpoint, is to be able to leverage our knowledge and skill with data and digital. And now of course, as you guys know, one of my favorite topics is AI to help really establish a technological presence with these industries and help them become more efficient with it as customers in the processes that they're undertaking.

And

Mike Burke: so in these verticals, what kind of questions are folks asking? What are they hiring you and Fusable for today?

Matthew Cox: There's really there's. Two. So I, what I do is when you think about all those different verticals, right? Then you have customer segments that go into that, but then we narrow it down to use cases.

So two, I actually say we probably have four specific use cases I'll call out. And number one is sales and marketing, right? So we have a sales and marketing use case. We have a risk management use case. Then we have the digital, right? So I'll call out digital as a use case as well, and that, getting to our digital audience, whether it's through media or through, it's through digital campaigns.

And then fourth we've really put a lot of energy around professional services now because we're really trying to, not only, it's one thing, as you guys know, to deliver product and then, [00:05:00] and create a solution or create a partnership with a customer, but it's another one to actually help them.

Leverage your product more effectively internally. And so we're wanting to help become that glue, that accelerator. One of the things we've talked a lot in the past is time to value. And a big proponent of time to values. We believe that there's a really great opportunity for us to help our customers really accelerate that time to value and shrink that amount of time.

And so that's the four areas we're focusing on. But from a sales and marketing standpoint, whenever there's a, a, we'll call it an. Asset. So we have two terms we'll talk a little bit about, one is it's the entity, right? So you, the typical company, right? We talk about master trading management company, right?

So you have this, a company that then has an asset, and that would be, say a tractor or it could be a truck, right? So when someone finances a new truck or someone purchases a new truck or a new asset, that's information that companies in all of these different verticals wanna know about, right? It can become sales and marketing activity, it can become an insurance activity.

And there's a number of things that. That triggers. So we have, we'll say transactional data that we collect that then triggers activities that our customers wanna know about so that they can then [00:06:00] follow up on a Very

Mike Burke: interesting, so is it more geared towards the sellers of these added services and added technologies to the individual who's making a purchase?

Or is the data to the individual's company? So it's more

Matthew Cox: of the former, right? So what'll happen is. Look, let's say I go to the, I go to a dealership today and I buy a new tractor, right? I get a finance, that's one, one of the ingestion points. So what other dealers or other companies in that industry would like to know is, okay, now there's service plans.

There could be additional components. They could look at it and say, okay, hold on. They just purchased that. That kinda think about like even a lease with automobiles, right? Someone purchased it. We have that tracking. We know probably in three years they're gonna want to get outta that vehicle.

They're gonna change out or sell that tractor of that asset. So that now becomes an opportunity that they can follow up going forward from an insurance standpoint, you could say I, if I have an insurance, if I'm an insurance company and I have a relationship with that company, I'm seeing that they're buying more vehicles.

Right now I can, that I understand the broader risk that I have potentially with that today or that organization. So that helps me from an underwriting standpoint. So these [00:07:00] transactions become both. A, an action that can be followed up on, but also becomes part of a longer set of detailed history. We have years and years of all these transactions and all this stuff, so now insight becomes a very valuable part that we can provide as well.

That's

Mike Burke: so interesting. But I actually was looking at buying a tractor at one point. And I did something very similar not very similar, but I went on to, I think it's Iron Planet, one of the big used resellers of tractor equipment. Yep. And I dumped all their data into a database and then built like a mini, what are the box of whiskers of what should I be spending on a tractor?

I know nothing about farm equipment, but I, I think that when you talk about these service industries, even at larger scale, my dad was a builder and a contractor, and now he's in manufacturing. They track very little information that's and making that's right. Decisions about how to grow their business, how to market to others.

This whole, it's not just individual businesses, but the whole ecosystem is very like not data driven today.

Matthew Cox: You're [00:08:00] absolutely correct, and I think that's, again, there, if I'm in construction, I'm worried more about the construction project in front of me and not going out to market my services and everything else.

So that's where the digital side of our business comes into play, because they can leverage all those digital assets, but then also they have access to all the data. We have as well. And so one of the challenges for us has been when you think about, when you know, anytime you go through an acquisition process and you're buying additional companies the merging of that entity of that organization, their data is always one of the biggest challenges you have.

And so that's one of the areas that we're putting a lot of data transformation around today, is we're taking all of those silos of organizations, platforms, and products. And there's a number of products we have, it's. There's one called cab, we once, EDA, rig dig, equipment watch, et cetera. So we have all these really amazing products that we've put in place.

Now we're taking all of that data and bringing it together in that connected data strategy. So the idea of saying, if I understand company A, B, A, B, C in one regard, now look at all of these additional areas of [00:09:00] activity and knowledge that I can now bring together into one fold and represent that out.

To my customers. And so connected data is a really big part of our priority initiatives that we have underway.

Chris Detzel: I have a quick question. My uncle owns and his brothers, it used to be called in Saint Roy City, Texas. Roy City, yes. No. Yep. I'm sure you do. You know all those little cities. Or it's not little anymore, but they own New Holland Tractor.

So they've owned it used to be called Ford Brothers Ford, but they have two. One is in Terro and then one is in Roy City. Roy City being the, would this company use your DA data or would it be more of the finance companies that would use that data? So I'm just trying to understand

Matthew Cox: actually. Could be both, but Absolutely.

The new Hollands, the Kubotas, the John Deers, the cases, all of them would be, would absolutely be customers of our data. That is correct. Okay. And it could be dealership. Or it could be the OEMs themselves, right? It's really both because the information we have can really enable an OEM. So say, an original equipment manufacturer, but also then the dealerships that are now trying to look at the various [00:10:00] types of products they're trying to sell now, just like you talked about.

Mike, from a use standpoint, we actually have a product called Iron Guides, which gives prices, right? It's like a price guide. Think of KBB or nada.org or those other organizations where actually gives you a pricing. We actually can give pricing on those type of those type of assets or units. Hey Matt, I think we lost you there for a moment.

I'm not

Mike Burke: sure if anyone

Matthew Cox: else. You look great guys. You look great. I see. Rolling. Okay, good. Okay, perfect. Yeah, I, you hear the point about the Iron guides and conversation about what Iron Guides provides 'cause to your point of pricing. That's a great question. So if I'm a dealer and I'm trying to understand what I should price a piece of equipment at, I need a comparable, right?

So this is what's really great about our Iron Guides tool too.

Mike Burke: It is so interesting and I feel like these industries specifically are so under leveraged in how they're capturing data. And I know that, the John Deeres of the world are starting to do a lot more of the predictive maintenance kind of pieces on the actual devices themself, but nobody is really assessing the market in that.

The way that you're looking at this data, how are you convincing the [00:11:00] retailers, the insurers and these different large businesses, how are you educating them on this is the right path forward and this is how you're gonna make more informed decisions? I'm sure there's a lot of education that goes into a product like this, right?

Matthew Cox: There is there, and I think it's, the good thing about our products, specifically cab ed and these others, they've been around, they've been in the industry for a long time. They actually have been a standard, right? They are the standard tool to, for them to go leverage. So we have great. We have a great reputation and great products that have been in this industry for some time to bring leverage.

I think the question that I would say is really the next step, right? We start talking about the connected data. We start talking about insights, right? And the pulling things that we've talked about in the past. Predictive, prescriptive actions that would come from analytics. To me, that's the real area of next step growth.

To really to help them understand is okay, it's one thing. To receive this information that we can send you this transactional data that we receive and information we receive from various sources. But really the next step is let's help you understand what those next trigger events should be.

How should you look at the broader, say area or the broader type of [00:12:00] vehicles or assets or whatever that you're trying to monitor? So I think that insight is that insight driven. So it's one thing to be data driven. It's another one to be insight driven. And so they, we've been helping them drive from a data standpoint and how they should act and respond.

Now, I, the next step is really to help them understand the insights we'll provide and how they should be now making decisions on a broader set of information, both from breadth and depth, historical, and across to drive that next series of decisions they should be making. Yeah, it's so interesting.

Chris Detzel: Go ahead.

I was just gonna quickly, I was just gonna say that sale has to be a little bit different than your normal, b2B type of sale. You go into a dealership for example, and these guys are in the middle of nowhere selling tractors and stuff like that. So it's a different clientele.

They're smart, all these things, it's not the same type of sale. You

Matthew Cox: know, like just, and I think that's, their technology is varying. I think that's part of it as well as you have to meet them where they're at, it is, it's not too dissimilar than a lot of any type of B2B organization that you'd actually be working with.

Their need for sales, their need for prospecting, their need for list marketing, [00:13:00] their need for information. Understand, where that, so when you break it down to, again, that's why I always, I tend to break it down into the use cases. The use cases aren't that dissimilar really, because at the end of the day, there's a sales and marketing play.

They've gotta bring in leads, they've gotta understand. Tell me what prospects are out there. Lemme look at a broad area. I may be a dealership wanting to move into, say, Texas, and I don't have any presence there. Let me look at how broad, the activity is within Texas and should I, where should I open up in Dallas?

Should I open up in Austin? Should I up in Houston? So these are the type of things that I think are really valuable to help them. Think of those next series of decisions they need to make as an organization. Really use case. The sales and marketing use case spans all the industries. It's really at the end of the day, just the type of data you're serving up to them.

In our case, we have to really have an amazing set of information that we can drive these particular verticals with that use

Mike Burke: case. So how did you get involved in this? Coming from Google, right? This massive, completely different ecosystem,

Chris Detzel: Mike. It's because he owns his own farm and he does all. Around his farm.

Yeah. And he thought, this is gonna be a [00:14:00] great company for me to work with. I do data and hey, be cdo. This is perfect. No, go ahead.

Matthew Cox: And so I think it, it is different, but it, but what's interesting is the challenge of these industries are very similar at the end of the day. If you're looking at, if you take data.

So let's and I think it's important for me to reiterate that in my, I'm the. I'm in a unique CDO role because it's not just about data. So I'm responsible for product, for data, for engineering, for our cybersecurity, so our InfoSec, all of our enterprise apps, and then also, the areas of professional services that we're getting ready to invoke and deliver as well.

So I really have a, one of the things that was really interesting to me about the, in particular Fusable was the ability for me to see the full breadth of data. So typically in a lot of areas I've been involved in, it's been, it's either been product or it's been corporate data and it feasible.

I'm actually responsible for the entire set of data across the organization, which for me is really intriguing. But the other part that I really enjoy is. And I think you alluded to this the technology landscape. 'cause having you obviously running a [00:15:00] ranch and being familiar with that industry there is such an opportunity for growth and leverage of this, of what technology and what data insights and what sort of new things that are being delivered today, how they can take advantage of it.

That was ex it's exciting for me to see and an industry. Really grow and take advantage. I think there was an insight that I had obviously from Google that allows me to really help provide a vision and a guidance into this realm that they can take advantage of. So that for me was what was really exciting.

And really, when you deal with smaller organizations, the ability to bring data, product and engineering altogether is fairly unique in a lot of organizations. Those are separate. Completely separate and they have to work in parallel. In this case, the vision gets to unite all three of those together, which is really, for me, an exciting step because it drives product velocity.

Mike Burke: Amazing. And it's such a, to me, I think this is such an interesting area because up until, I would say a few years ago, I knew nothing about really the agricultural space or manufacturing space. My wife's in manufacturing and. I never realized that when you [00:16:00] invest in a piece of equipment that is the business, like you are putting your entire equity, everything that you've saved into that next piece of equipment.

And if you're a seller of those pieces of equipment, you might only sell one every 10, 15 years. So to that customer, so you really have to know. The market landscape, who's buying, what's happening in that space, and also economic indicators that to do that could be a profession in itself. And so I really think it's interesting with Fusable that it sounds like that's the direction that you're heading is providing more holistic information.

Yes. Along with. The individual data points that somebody would use to make smaller decisions within their ecosystem.

Matthew Cox: Yeah. 'cause you're right. 'cause there's cycles, there's, it's a cyclical nature. It's it, like most industries where the pricing will go up, the pricing will go down to try and understand what those wavelengths look like is an area of value that we can bring.

'cause at the end of the day, our desires to make every one of our customers as successful as possible. So give them. Information about what's happening in their particular area of focus, give them perspective on what that means across a timeline. [00:17:00] And then as much as we can forecast out what we believe, what that's gonna look like in the future, because that, to your point, these are heavy investments, right?

We wanna make sure they're able to get to leverage those as much as possible within their industry.

Chris Detzel: So I'm curious, so we now know a lot about Fusable, really cool company, really cool stuff. And I wanna talk more about your role and what is, you own a lot of different things you said, and what are the things you're working on, the things that you're doing, and what's some passion that you're having right now or some focus. And what are you passionate about is really kinda what I wanna know.

Matthew Cox: Yeah. Me, I'm, I have a lot of passions, right? So actually is multifold. So it's passionate, a number of things.

So I, I think from my standpoint, I have a great appreciation for all these, how all these pieces come together, right? So I've talked in the past about what the need for data quality, right? The need for data governance, and the need for being able to provide a framework for that. We've got, our friends, our common friends like Scott Taylor and others, right?

That, that talk about, it's yes. But the data. And I think. One of the things that I'm particularly excited about is a few parallel paths that we [00:18:00] have, right? So number one is really getting our arms around on the data itself. So I talked about that connected data strategy, right?

But then understanding within that, what is the framework to ensure the data quality, is at the forefront of our concern. And that can be in a number of things, right? Whether that's observability or explainability, they're looking at how we're, common and consistent ingestion that we begin to put into the engine.

So I think that's a big part for me about data is really getting our arms around how do we provide the most robust and most well honed and best standardized data. We actually call it core data, our core data set that we can bring to the table. And so to me that's exciting because.

The first part of it, because trust and value is a big part in my mind of how we bring ourselves to the table with our customers trust that we're doing everything we can to build the best data set possible, but then value that we're actually delivering capabilities that help them actually move their business forward.

So from a data standpoint, it's really getting our arms around that robust and rigorous environment. From a product standpoint, product velocity is a big part for me. Really [00:19:00] giving our customers, our sales teams and our customers, the ability to have multiple ways in which they can engage with our data, right?

So it's not like your typical, I have a data organization. I go, I drop a file off to a company and just say, Hey, go off and have fun and talk to me at six months if you figured out how to use our file, right? We're building common canonical services. We're building new end products, right?

So there's really, and there's really two types of end products. There's one, which is a, say a Fusable environment. That's our standard platform. That's unique and specific to make available. But then also looking at platform native systems, right? So if we want introduce an applet into Salesforce or Microsoft Dynamics or another area where, again, the idea is it with product is to really focus on that moment of truth, right?

Where is the point? Where our customer exists, where they need our data to be in order to make a decision, right? I wanna be as close to that moment as possible so that we're delivering the highest value and that they're not having to, leave their environment, then come back and figure out how to work with feasible and some really infusing ourselves into that arena.

And I think that's particularly important from a [00:20:00] product standpoint. But then from an engineering standpoint, one of the things we've talked about is a microservices architecture, right? And really building reusable components and creating an environment where we can very quickly build. Variety of capabilities.

If you think about Lego blocks, right? You can, in the same Lego set, you can build a tie fighter, an X-Wing fighter, you could build all sorts of different things, right? Just depending on how you assemble them. And so the idea from an engineering standpoint is we're building a robust stack that allows for these small components to be assembled very quickly in a new products, which to me, builds that pro, builds on that product velocity view and went on top of that.

I talked about being responsible for our info, our information security, we have to wrap everything with a security first priority, right? When you're dealing with device to cloud, right? You really have to build a robust environment that your customers can have faith in, and then trust that once they give you their data, I.

That whole transaction is secure and safe for them. And so that's a piece that I felt was very important to wrap. So if you think about those three pillars of data, product engineering, wrap with security, I [00:21:00] think is a great message to bring out to our customers and say, listen we're as concerned about your data as you are and making sure that in our negotiation, our relationship, it's safe.

And so those things really, for me are important. And then if you think about then taking that robust environment and adding our AI spin of how do I now enable. Machine learning and reinforced learning. How do I look at potentially agent tech ai To start to actually have run transactions, actually begin to do specific roles for me, builds a value proposition.

That's very. Very strong over the next upcoming years. So to me, those were all very important pieces that we need to establish and things that we will feel in our roadmap are very important for us as a company to move forward with.

Chris Detzel: I heard the term, just this recent, that I've started hearing the term ATech ai.

Can you talk a little bit about that? Because, it feels like another buzzword an important one that we need to know about. It sounds like

Matthew Cox: another buzzword. So they just had to have another AI thing, yeah, lemme talk about that for a second. 'cause I, I'm actually really energized about agent ai.

Yeah. Lot of people are

Chris Detzel: [00:22:00] obviously.

Matthew Cox: It's because it, so what's important for me is a couple things. Number one, we've talked about LLMs for a long time. We've talked about our, a GI, artificial generative intelligence, right? So we've talked about all that work, right? That has the process of you developing human type capabilities.

I, you can ask it a question like chat, GBT, it gives you an answer back. All that's great, and I think there's some things we can build upon, but what is, I think, actionable now are, is our ability to put. This what we call this agentic ai or AI for usefulness in very specific roles, right? So think about if I have someone who's, that's trying to perform a task that I now need to automate.

So Agentic AI automates that role. And what's really cool about agentic AI is that I can leverage ml, right? So I can have reinforced learning that helps me make a decision, but more importantly, I can place that against even my existing. Capabilities. So say my existing APIs, my existing capabilities and pros, agent AI doesn't have to have ML or a GI or anything else to operate, right?

It allows you to actually say, I [00:23:00] have a role or a function I need to play, say customer service. And that agent AI role can actually now begin to perform all those functions and decisions that someone in customer service would be performing. And so I think that's where we can talk a lot about the future of ML and what it does in the future of a GI, et cetera.

But what's. Actionable today, I believe is agent ai. And I'm really looking forward to seeing how we can put that to work as products, but also internal, right? You could, think about processes that you'd run internally, right? There's things you'd even do internally as a company that, that those, you can really suit up that role.

'cause its purpose is within a bounded area to make decisions right on, on behalf of what? Of the input that it's receiving. Very exciting steps, I think for agentic ai.

Mike Burke: I think this is one area where when we started talking about generative or general intelligence, I was like, okay, that's great, but it's gonna be a long time before we get there.

Unless we see major strategic technical advances in the offerings that we have today. But with agentic ai. We're essentially arming or enabling AI to [00:24:00] take action without the need for human intervention to some degree in a specialized task with controls. How do you think about feasible and ag agentic ai?

Are there gonna be focuses on certain areas like risk modeling or deal intelligence? Where do you think that's headed in the next couple years?

Matthew Cox: I think all of those are opportunities and I would say and one quick thing about agent tech ai, one of the things I think is really important 'cause we've talked about data quality.

Over and over again. And I think that there's, I think everyone's taking it seriously to a certain point, but then they pause because, investments in data quality has always been a challenge. Even with even things like even MDM, you're always having to make this broad case for why you have to go and deliver these capabilities to improve data.

But what I think is gotta make people a little more unnerving or unnerved in this environment is the idea that I'm having agent take AI work autonomously, right? So before there was this idea that, hey, I go and build a report. Or I build a transaction, I've got a human standing there that can make it a gut check decision.

It's oh, call on the sleeve. I'm like I don't know that company. I'm not really [00:25:00] sure. I'm gonna wait on that. So data didn't have to be as the bar for the level of quality data didn't have to be as hard, high before. 'cause you always had someone who could intervene at the moment before a decision was made.

We're at the point now with that, specifically as we looked at agent ai, there's not, and we've removed the intervention moment, so data quality and making sure that what you're. Passing into these, whether it's agent AI or even ML as we move forward, is really important because there's not someone stopping or intervening at the moment before a decision is made.

So I think data quality and the work that we're doing with our connect to data strategy really, and all the frameworks that running is really preparing us for that next mission one way or the other. So I think it's really important people understand you. You can't. Go to agent, take AI and go, oh, this is a great wizard.

I'm gonna throw it out there and just start running things against you. You've really gotta take very seriously the fee, the what, you're the feeding into that and have the level of quality and assurance that you're able to deliver the right information for it to act on. So first of all is that, so secondly, to your point on the risk, so to me, where does a agent take AI actually begin to play in our future is [00:26:00] I look at those use cases.

To me, it all starts at the use case, right? So I look at sales and marketing. We'll just take two for now. I'll take. Sales and marketing and risk management. So within those use cases, there are personas, right? So you have people that are, actors that are performing a certain role. Any of those roles that can be mapped within those use cases, to me, is an opportunity for agent AI to play, right?

So whether it's potential for being brokering between, for being inside sales, potentially for being lead acquisition, any of those personas or actors that have to take in data, make a decision, and then act, they all become potentials for agent. Take ai. As we move forward, then of course, the question, Mike, would be the value prop.

The trust piece, how much is an organization willing to hand off automation of those roles to an agent versus a, an individual or human being making the decision? That's some of those series of steps we need to make. But at the end of the day, if you can't. When you're looking at delivering out product, you, you can actually scale into that discussion because the first thing is if a customer can't trust you to deliver insight [00:27:00] you'll never get to the next step, right?

But if you can deliver insight, then maybe you can move into, say, predictive. Analytics, right? So if, and then if you can move that next step, they're like, okay, that's great. I can now trust how you predict things. Then the next step is, okay, maybe you can prescribe some behaviors that I should do, or some prep some action.

So if you move to that next step of prescribing what should be done, there's another tier of trust, and then that next trust could then move you to agent ai. If you wanna go from today, which is like a transaction piece of data to agent ai. You've missed a series of steps of trust that you need to move through to move that point.

So I do think it's a journey, Mike, that will be on, but I do think you can make steps over time to reach that point where both the value of what you're delivering and the trust that you can actually deliver can come together. I. And meet that need.

Mike Burke: I love it. And there's so much there that we're seeing this with a lot of customers that I work with as well.

Like prescription is really the first major focal point of a agentic ai. And what does that mean? It means we're gonna tell you that we think you should do something, but we're not just gonna recommend a action. We're [00:28:00] gonna explain to you why this action was important and how we came to the conclusion that you should take this action.

And so I think of it like peer programming or having a teacher review your assignment. There's somebody who's gonna be sitting there saying, we saw 10 other deals or 10 other claims be filed this way, and so we're gonna recommend this approach. You don't have to take that information and act on it as an actuary or as whatever use case you're playing within that role.

But you now have a really informed decision by someone who was able to analyze a massive amount of data that wouldn't be worth the time for you to go and investigate unless you'd worked at the company for 20 or 30 years. And that's the kind of experience that you're getting. With this first step of kind of informative instruction with some reasoning behind it.

Matthew Cox: And this is where I think the value prop that Fusable brings to the table is we have 20, 30 years of data. We have 20, 30 years of being in the business in a lot of these areas. So we do have that IP that we can bring to the table that says, it's not just about here's something that's happened, it's here's something that happened and let me give you.

[00:29:00] Context, here's the context around that you should be perceiving. 'cause that should then help drive, just like we talked about the pricing, right? I've got a tractor. What pricing level should I put on it? When should I try and sell it? What's happening in the market that I should cite either to sell now or sell maybe in another quarter?

Depending upon when the market looks right. That context and insight, I think is where we really bring huge value to our customers, but that moves us then toward that point of. Predictive prescriptive and then saying, Hey, if you've trusted me this far, maybe you can trust me the next step of allowing an agent to actually start to have even a bigger role in that piece for you.

Mike Burke: Yeah. I love it. The example I always like to use for folks that aren't technical on the call is like coin evaluation. When you think of like a, an old penny. And one that's worth millions of dollars for an ai. It's what makes it valuable? Is it the year? Is it the condition? Is it what was done to it?

There's so many pieces of context that, some might be in, in structured databases like sales of previous pennies out there, but others, you need to analyze it to a specific degree. There are a lot of technologies out there, but there also a lot [00:30:00] of opportunity for areas where data hasn't really been captured on those areas.

And agent AI is going to get there very quickly, I think a lot faster than everybody realizes. And if you have a platform like what Fusable has, and there's many other businesses out there doing similar things in different industries. That has the dual flywheel where you have really good inputs and you have really good abilities to output.

It's just a recipe for success and I think that's where any air period in AI is gonna blow up is where you have that multi-use asset flywheel of proprietary data. So much value there.

Matthew Cox: Yeah, I agree. But I think to, to the point, I think the question is why would they trust you? So why would organizations move to that?

I think it's, it goes back to again the being able to speak to what is the framework. That I'm putting around the quality of my data. What is the observability and explainability that I have around the decisions that I'm now offering up. So I'm, as part of what we're doing as a thrust internally is to begin to expose more and more of the work that's happening behind the scenes.

That we're putting in place prior to that product that our product each to [00:31:00] having data ready for our customer consume, right? It's easy to say, here's my ui. Go use the ui. Go pull information and be happy. But that doesn't build that next level of trust to say. But let me explain to you what's happening behind the scenes.

Let me show you the volume of data we're bringing in. Let me explain to you the series of steps we go through in order to make that data prepared for you to consume it. Let me show you from a. Observability standpoint, how this is all flowing through and where we're at. At any one point, lemme explain to you why you're seeing what you're seeing.

All those series of steps and that information then begins to build that trust in, okay. It's not just the UI I'm dealing with. It's not just that piece of information that you gave me, the data I. But there's a whole wealth of activity that happened prior to, which is really the, ultimately the great value you provide.

Yeah the product looks great and they interface the product, but it's all that work that goes into what's being delivered that I think is gonna help customers and others really begin to trust and respect the decisions we're trying to enable for them and move to that predictive, prescriptive, agen AI series of steps.

They've gotta understand what's [00:32:00] behind and how that's serving them, I think to be able to trust to that next. That next step forward. So you've gotta be willing to open up the curtain a little bit and let them see what's going on behind the scenes.

Mike Burke: Absolutely. How do you see that transparency evolving as the business grows and also AI grows in the industry?

How vital is that to success? You guys have established a long tenure of credibility and transparency with these organizations. How do you see that taking flight in a realm of AI where explainability might become a little bit more difficult? You might need to be an expert to understand some of these things.

Matthew Cox: Yeah, I think so. And I think so. I think there's a couple faults. Number one, you know we are, I. We are, creating the ability for us to have our own professional services and individuals who have that SME to walk our customers through a lot of those steps. 'cause your point is correct that the step between what I'm operating today and where I need to go as an organization is opaque.

In a lot of cases, right? And so you need almost a tour guide to walk you through those steps to say, listen, here's how you get from A to B, so you can actually start to take advantage of that. So I think part of that is incumbent on the organization like us to provide [00:33:00] some of that expertise and enable our customers to do, to be more effective and to understand more quickly where things are going.

I think number two, you have simplicity, right? Be sophisticated, but simple. And so I think if you're, if what you're delivering out is too sophisticated and it takes three months to learn it, you've probably got a problem. You probably over overestimated your need to deliver the experience that's happening to that customer.

So I, I suck a lot about being sophisticated, but simple. And so that's another approach that I think we're gonna be pushing through. And then I think thirdly, giving them that as much insight as you can into the work that's going on, right? So we delivering insights about how many. You like from an ingestion standpoint, how many of this type of file, this type of source information did we process?

How long did it take to go from here to the point you consumed it? Looking at the overall volume, I'm, I've always been a fan of, we used to call it customer intelligence, right? It was really just how you're doing with your customer data and how you're trying to deliver information. I think it's really important as we move forward, we're asking our customers to trust us.

Even more to give them even more [00:34:00] of that insight into what's happening behind the scenes. And I'm happy to give that more of that intelligence out. There's still ip, you wanna keep your secret sauce, but helping them understand the results of the secret sauce is really, I think, imperative to have that trust.

'cause we're in several areas, even at prescriptive, we're literally asking our customers to trust. What we're telling them for them to go act upon. So they need to be able to trust that we're as serious about the consequences of that decision they make as they are. And so we wanna make sure we're with them and have skin in the game for that knowledge that they're trying to bring to bear.

Chris Detzel: I'm very passionate about, the simplicity as you mentioned, but that's hard to do, right? Yes. Like you think about bringing data stuff to people in general, MDM, there's nothing easy about it. There's even some of the stuff that you're doing and trying to understand that. So I'm always a big fan of, I.

Either building, communities or programs that will help your customers, get the most out of your products. Because you know what, there's always more complic complications than you think there is for anybody. It's just the reality. The deeper you can go into to [00:35:00] showing people how to use it, what it's about, use cases.

And things like that the better because you show one person's, oh, I didn't know you could do that. They just 'cause they're one. 'cause peop we in general just think, we want this data for this, but you can actually use it for the these things too, and so I'm just a big proponent of building engagement type programs that go deep into the products and show people how to use it and everything.

I didn't do that necessarily with you, but I did that with your team at Google. It was. Or if they wanted to.

Matthew Cox: I think part of it's, I think there's a couple ways of having that simple mode. I think number one it is looking at, I talked about the moment of the truth, meeting them in their natural place of work, right?

So I think if you're, you've asked, if you ask an individual to leave the work that they're in, say they have a CRM tool and they're used to working that CRM tool all day long. And now they've gotta leave to go find prospects. They've gotta leave to go get data and then come back. You're taking them out of that natural environment.

So the intent would be in a CRM, why [00:36:00] can't I bring sales and marketing data directly into the world that they live in? So really need to matches them with their point of need so that, that makes it simple, right? If you look at, if you define things less on, here's the breadth of what I'm doing for a data set or an industry, and you get down to.

Hey, let me talk to you, company A about your use case, say sales and marketing. Let me talk about that persona, which is what you happen to be, which is a sales manager. Let me explain to you how we're helping you in that specific role. So I think if you take the message to the realm they're comfortable in makes it lot, the path for them to engage is simplified versus coming in saying, oh, I have a CRM solution.

Why don't you just go, engage and have fun. We try to take it to the person and make it much clear on how can they engage, but at the end of the day, it's about the moment of truth. Can I be there at the can our data, can our solutions, can our insight be there? At the point you need us to make a decision 'cause you have to leave that environment.

I've become sophisticated with a sophisticated solution and that's not where I want to be.

Mike Burke: This is super helpful. I am really excited to see where you go with this and [00:37:00] following feasible as they continue to grow and evolve. Thank you so much for joining us on the show today. That's all. Anything else that you'd like to ask his last questions?

Chris Detzel: No, I think that's it. Matthew certainly appreciate you coming on as usual a ton of stuff about this space so we could even go deeper. I did have questions about data quality that nobody's ever solved, it seems, but we could go deeper into that at some point. 'cause I'm very passionate about how to solve that problem of data quality and how.

People or companies are trying to solve that problem. We won't talk about that today, but thanks everyone for tuning into another data hurdles. I'm Chris Tetzel. Don't forget to rate and review us. Until next time, Matthew, thanks again.

Matthew Cox: Thank you. Appreciate it gentlemen. We'll talk again soon.

Chris Detzel: That was excellent, man.

As usual,

I.