Latest / The Adults In The AI Room / Taking Control of your Data: How Knowledge Graphs Help to Optimize your Business

Transcript

Taking Control of your Data: How Knowledge Graphs Help to Optimize your Business · The Adults In The AI Room

• 0:01
  SPEAKER_00: This is the Early AI Podcast, hosted by Seth Early and Chris Featherstone.
• 0:06
  SPEAKER_00: Join us as we delve deep into the passions, expertise, and experiences of thought leaders and practitioners to talk about what's possible in artificial intelligence.
• 0:15
  SPEAKER_00: Early AI Podcast is sponsored by Early Information Science, your digital transformation journey with the design and deployment innovative technology solutions.
• 0:28
  SPEAKER_03: Well, let's go ahead and get started.
• 0:30
  SPEAKER_03: Welcome to our podcast this month.
• 0:32
  SPEAKER_03: My name is Seth Early.
• 0:34
  SPEAKER_03: I'm Chris Featherstone.
• 0:35
  SPEAKER_03: Really excited about our guest who is an expert in knowledge graph technology.
• 0:42
  SPEAKER_03: He's a CMO.
• 0:44
  SPEAKER_03: He's uh very passionate about um uh B2B retail, supply chain data.
• 0:52
  SPEAKER_03: He's a tech evangelist who focuses on e-commerce, customer centricity, and getting data and insights from uh getting insights from data to support uh uh business and business management.
• 1:05
  SPEAKER_03: And he's very passionate about enabling digital business.
• 1:08
  SPEAKER_03: Uh Doug Campbell, Chief Marketing Officer of AutoTechs, welcome to the show.
• 1:14
  SPEAKER_02: Thank you, Seth.
• 1:15
  SPEAKER_02: Glad to be here.
• 1:15
  SPEAKER_02: I mean, I've been enjoying our back and forth conversations for almost a year, and to be uh part of this process is pretty exciting.
• 1:22
  SPEAKER_02: And like you said, I have I have fun when I get to talk to the businesses about how technology can be used to make their lives better using data.
• 1:30
  SPEAKER_02: So it's been a fun full circle journey to end up here.
• 1:34
  SPEAKER_03: So, you know, it's interesting because uh knowledge graphs have been around for a while.
• 1:39
  SPEAKER_03: Uh the term ontology has kind of become a little bit more mainstream in the last few years.
• 1:44
  SPEAKER_03: It was really something that caused executives to uh uh I their eyes to glaze over or to uh walk out of meetings.
• 1:52
  SPEAKER_03: But uh now people are starting to realize the importance.
• 1:54
  SPEAKER_03: But when you come when it comes to knowledge graphs, um, maybe you could talk about um the misconceptions.
• 2:02
  SPEAKER_03: Like, what do people not understand about knowledge graphs and ontologies?
• 2:06
  SPEAKER_03: Maybe you could start with that and you can provide kind of working definition of of ontology and knowledge graph, and but talk about what people don't understand.
• 2:16
  SPEAKER_03: Sure.
• 2:17
  SPEAKER_02: I think there's a problem, I think there's a lot that they don't understand about graph technology and knowledge graphs, particularly.
• 2:22
  SPEAKER_02: Um, one of the things that we do encounter a lot really is this a rip and replace?
• 2:27
  SPEAKER_02: In other words, here you come talk to me about another database structure, another architectural process.
• 2:32
  SPEAKER_02: I've already got fill in the blanks, SAP, Hadoop, et cetera, et cetera.
• 2:36
  SPEAKER_02: Do I have to go make all these massive changes?
• 2:39
  SPEAKER_02: I've already spent this money.
• 2:41
  SPEAKER_02: And really, knowledge graph, not graph technology knowledge graphs are not a rip and replace.
• 2:46
  SPEAKER_02: They're an ad to, they're an enhancement of, um, in order to bring all that data together that they've already been using, you know, they have all these disparate systems.
• 2:55
  SPEAKER_02: They've got SAP, they've got it, you know, the infrastructure, they've got Hadoop, they've got databases, they've got unstructured data.
• 3:01
  SPEAKER_02: Um, but nothing has to change in where they are putting things, how they look at their how they store their data.
• 3:09
  SPEAKER_02: It's more inserting this as I like to use the analogy of inserting a better gear, you know, like dropping out a 10-speed gear into your existing engine to make it run better, faster, stronger, all those wonderful little pieces we can say there.
• 3:23
  SPEAKER_02: That's what it's truly about.
• 3:25
  SPEAKER_02: And then when you take an ontology, the ontologies, and you now you're you're taking all this great data that we have and you're layering on top of that knowledge of how to communicate about it, how to use it, how to access it, how to understand, how to make your technology systems use that data, understand that data, applying context to it to make better decisions.
• 3:49
  SPEAKER_02: So I think one of the easiest ones is really it's it's this is not it's not easy.
• 3:53
  SPEAKER_02: I don't listen to this isn't a plug and play and off you go.
• 3:56
  SPEAKER_02: It does take time and it is a serious practice, but it's not a I'm giving up all this great money I spent on something to do something new and shiny.
• 4:06
  SPEAKER_02: I'm doing it better.
• 4:08
  SPEAKER_01: You know, especially to um into an executive or a stakeholder, why they should even care.
• 4:14
  SPEAKER_01: And once they do care, why you know what's that investment gonna look like?
• 4:19
  SPEAKER_01: And you added to that, you know, just in terms of the time and money.
• 4:22
  SPEAKER_01: But part of it too is, you know, like you said, we've got all this expensive stuff.
• 4:26
  SPEAKER_01: Why should they care about a knowledge graph?
• 4:28
  SPEAKER_01: And where does that get them?
• 4:29
  SPEAKER_01: Right.
• 4:29
  SPEAKER_02: You know, it's a strategic tool.
• 4:31
  SPEAKER_02: If I come back to you and say, oh, we need more data, we need you know, more and more.
• 4:35
  SPEAKER_02: We've heard that, you know, the the leaders, whether they're CDOs or CEOs or CIOs, have heard we need more data.
• 4:41
  SPEAKER_02: Now we've got all this data.
• 4:43
  SPEAKER_02: What do we actually do with it to monetize it, to get better personalization, to reduce the use of our resources, to streamline agility, outcome driven.
• 4:54
  SPEAKER_02: And that's where a knowledge graph and the knowledge graph technology comes into play is that you're you're adding this framework, this powerful connected framework on top of all the data structures that you already have, um, to everything that goes back to the business outcomes.
• 5:10
  SPEAKER_02: You want to drive innovation, you want to increase efficiencies.
• 5:13
  SPEAKER_02: That's adding to the capabilities.
• 5:16
  SPEAKER_02: Data by itself is great, and you good it's good that you have it.
• 5:19
  SPEAKER_02: It's nice and clean, organized, disambiguated.
• 5:22
  SPEAKER_02: But what do you do with it?
• 5:23
  SPEAKER_02: And that's where I look at it from a C-level conversation is okay, drive the top down.
• 5:28
  SPEAKER_02: What are the results we're trying to achieve because of having better data?
• 5:32
  SPEAKER_02: Get it, understand the business problems, and then drive that down to the technology, be able to make it make it happen.
• 5:37
  SPEAKER_03: So what's interesting is it it reminds me of the old B A S F commercial.
• 5:42
  SPEAKER_03: We we make uh we don't make a lot of things, we make a lot of things better.
• 5:46
  SPEAKER_03: It's an enabler, right?
• 5:48
  SPEAKER_03: Because they made all these things that improve products.
• 5:50
  SPEAKER_03: But we're saying that the knowledge graph and graph technology uh will will enhance uh the information systems that are in place.
• 5:59
  SPEAKER_03: Um, but sometimes there's a big challenge with that because it's so uh it's it's infrastructure, it's a capability, it's foundational, it's hard to connect that directly to a business outcome.
• 6:11
  SPEAKER_03: Um, but so how do you how do you talk to stakeholders and executives about the business outcome and what they can do with it?
• 6:18
  SPEAKER_03: Uh, because I think a lot of technical people kind of understand this, you know, from that infrastructure perspective and that data fabric perspective and the whole idea of reference data and so on.
• 6:28
  SPEAKER_03: Tell me more about how you kind of connect that to the tangible business outcomes.
• 6:33
  SPEAKER_02: Sure.
• 6:34
  SPEAKER_02: And I'll use a topic that really is what inspired me to eventually reach out to to you guys.
• 6:39
  SPEAKER_02: Look at product information management.
• 6:40
  SPEAKER_02: I mean, it's a it's a topic that I spoke on before.
• 6:43
  SPEAKER_02: Um, and I'll and I'll I'll up level that to e-commerce and I'll up level that even further to customer acquisition and customer personalization.
• 6:52
  SPEAKER_02: So if I'm a business, whether it's B2B or B2C, those lines blur so much these days, you know, the B2B buyers, you know, they look to have the same kind of experience when they're shopping on Amazon as they do when they're buying 50 rolls of you know steel to make a car.
• 7:06
  SPEAKER_02: How do you provide the most customized, engaging, personalized experience for your shoppers, customers, consumers, whatever it is that you want to call them?
• 7:19
  SPEAKER_02: How do you do that?
• 7:20
  SPEAKER_02: And you need to do that with the right data being served up to the right people at the right time with the right connections.
• 7:29
  SPEAKER_02: And it's that's one thing to do.
• 7:31
  SPEAKER_02: And we were already doing that.
• 7:32
  SPEAKER_02: I mean, we've been doing that for quite a long time in the e-commerce world, but now how do you then make additional connections behind that to be differentiated?
• 7:42
  SPEAKER_02: How do you stand out in the sea of people who have e-commerce or selling relatively the same product in the relatively the same channels?
• 7:50
  SPEAKER_02: You've got to hit people in a way that really appeals to them.
• 7:54
  SPEAKER_02: And that's where, again, the power of knowledge graph technology, with its ability to do what we call inferences or reasoning, to start making connections among the data that aren't implicitly there.
• 8:06
  SPEAKER_02: In other words, we put the data in, it's there, but knowledge graph technology can then say, Oh, Doug likes to shop for these kinds of things in the month of June.
• 8:15
  SPEAKER_02: Maybe I should feed this up to it, but then it takes an abstraction level deeper.
• 8:19
  SPEAKER_02: Well, Doug often buys also these things in June, and it starts to make those connections and those inferences to say, maybe we should serve up something a little bit different to him because his shopping habits aren't necessarily changing.
• 8:33
  SPEAKER_02: Who he is is evolving.
• 8:34
  SPEAKER_02: And so you have to evolve your commerce engine.
• 8:37
  SPEAKER_02: So I'm saying commerce engine is a generalized term there, but serving up material to somebody in order to make a purchase.
• 8:45
  SPEAKER_02: That's I go back to the whole product information management piece.
• 8:48
  SPEAKER_02: Is again, we're managing the information around a product or a service you're trying to sell to somebody in some fashion.
• 8:56
  SPEAKER_02: And the more you appeal to me when I walk into the store, physical, virtual, whatever it is, and you can make connections based on what you know about me already.
• 9:06
  SPEAKER_02: Looks like Doug usually buys these things, serve this up, or even pre-serve it up before I start to make a purchase.
• 9:13
  SPEAKER_02: You bought this kind of stuff before.
• 9:15
  SPEAKER_02: What about this?
• 9:16
  SPEAKER_02: So that's how I take the approach of speaking to it from a C-level perspective is what is it you're trying to drive?
• 9:22
  SPEAKER_02: I mean, volume, profit, and share is what a lot so many things come down to.
• 9:26
  SPEAKER_02: We're trying to try to drive volume, profit, share.
• 9:27
  SPEAKER_02: Okay, cool.
• 9:28
  SPEAKER_02: What are you missing on your volume?
• 9:30
  SPEAKER_02: Why is your profit not where you want it to be?
• 9:32
  SPEAKER_02: Why are you not getting the kinds of share that you want?
• 9:34
  SPEAKER_02: And then continue to ask those questions to drill into that.
• 9:38
  SPEAKER_01: Yeah, I had a CEO.
• 9:40
  SPEAKER_01: We were talking about, you know, data management, governance, and guardrails and adding a knowledge graph to their environment.
• 9:47
  SPEAKER_01: And um, he's like, so what?
• 9:49
  SPEAKER_01: Great, I got all this data.
• 9:50
  SPEAKER_01: So what?
• 9:51
  SPEAKER_01: What's this gonna get me?
• 9:52
  SPEAKER_01: You know, it doesn't it doesn't matter.
• 9:53
  SPEAKER_01: I put all this information now, you give me more insights, who cares?
• 9:58
  SPEAKER_01: And I was like, that's actually I think a good perspective to take because to your point, you know, um, what is the outcomes you want to drive and and you know, and what you're trying to hit.
• 10:10
  SPEAKER_01: And then now you like you said, uh you have all this data now to help supply support and give you the structure in order to now drive one outcome or 50 outcomes, depending on what those are, and not have to go back and say, well, we don't have this information, we can't even really drive that outcome, and or a really good clear recommendation, you know, right on top of it.
• 10:30
  SPEAKER_01: So yeah, it's important.
• 10:32
  SPEAKER_03: Well, what else?
• 10:33
  SPEAKER_03: I guess my question would be you know, the knowledge graph is providing kind of a reference data, it's providing uh relationships between concepts.
• 10:44
  SPEAKER_03: If you want to do customer 360 or personalization or recommendation, what else needs to be in place?
• 10:50
  SPEAKER_03: What else needs to happen?
• 10:51
  SPEAKER_03: Because the knowledge graph is one component to that, but what are the other pieces that need to be brought together?
• 10:57
  SPEAKER_02: The risk of going back to what Chris was saying is yes, you have to have the data.
• 11:00
  SPEAKER_02: I mean, obviously, I guess it sounds obvious, but you have to have good, clean, you know, disambiguated, deduplicated, accurate data that's enriched.
• 11:10
  SPEAKER_02: I mean, just knowing that the Doug Kimball purchases things online is one example, but how do we verify that Doug Kimball A is at this address?
• 11:19
  SPEAKER_02: I mean, I was just telling Chris, I'm living in Oklahoma City where it's been plus hundred-degree weather for the last month and a half.
• 11:25
  SPEAKER_02: I know there's another Doug Kimball out there who lives in California.
• 11:28
  SPEAKER_02: He's probably enjoying much nicer weather than I have.
• 11:30
  SPEAKER_02: So you've got to disambiguate who this person is instead of serving up the Doug Kimball in California, more power fans.
• 11:38
  SPEAKER_02: I'm making this up.
• 11:39
  SPEAKER_02: You know, you've got to have that organized.
• 11:40
  SPEAKER_02: So you know, you talk about customer 360, product 360, asset 360, all those things that go down to having the right data.
• 11:51
  SPEAKER_02: Um, you have the data mastered so that you understand that you don't have multiple records of these things.
• 11:56
  SPEAKER_02: And that's where, again, I see such an inherent value of knowledge graphs in that you know, where you know, in a relational database, you've got you know a series of things in knowledge graphs, you can go different directions and add these different perspectives in to add to the depth, but again, without good, clean access, you know, data governed properly with metadata layered atop of that, so you can make the queries better.
• 12:21
  SPEAKER_02: It's just really good data.
• 12:23
  SPEAKER_02: What do you do with it?
• 12:23
  SPEAKER_02: And that goes back to what Chris you're saying is like, okay, cool, I can do queries faster.
• 12:28
  SPEAKER_02: So what?
• 12:30
  SPEAKER_01: Right.
• 12:31
  SPEAKER_01: So what exactly?
• 12:33
  SPEAKER_03: Well, I think the other thing that's super important is you know, we we built uh an information architecture to do personalization for an organization a number of years back, and when it came down to starting to talk about, okay, what is the difference between this segment and this segment and what information and what content is going to resonate with them more effectively, you really didn't have that answer.
• 12:55
  SPEAKER_03: So you can build a great infrastructure, you can have great data, but if you don't have maturity and you're supporting processes where you really understand that customer and their needs in detail, and then have infrastructure that allows you to start to do some experimentation uh to try to uh optimize certain messaging combinations.
• 13:14
  SPEAKER_03: I mean, again, that's where a lot of the um uh the multivariant testing comes in, but you kind of have to start off with uh, you know, what is this custom, who is this customer, what's their journey look like, and then what's going to, you know, uh what's going to help them move from one stage to another.
• 13:32
  SPEAKER_03: And that's really knowledge, that's really expertise about the customer.
• 13:36
  SPEAKER_03: But then that can be captured in the knowledge graph in order to be applied uh to that runtime uh environment where you're actually serving up things that make sense to that customer and then testing the details of that.
• 13:49
  SPEAKER_03: So it seems to me that that would work in conjunction with things like a customer data platform, product information management system, content management system, any other tools or technologies that you usually see in this kind of ensemble approach is usually some kind of orchestration engine.
• 14:04
  SPEAKER_03: So the knowledge graph fits in, but this ecosystem is bigger uh and uh um has multiple uh components.
• 14:13
  SPEAKER_03: Completely agree.
• 14:14
  SPEAKER_02: Yeah, you can't a knowledge graph by itself, we're we are just a really, really cool database, really, really powerful database.
• 14:20
  SPEAKER_02: And I don't mean to simplify it, but that's you know, all we are is a database.
• 14:24
  SPEAKER_02: And data, all data is is just data until you do something with it with the different engines that that go back to that.
• 14:29
  SPEAKER_02: I want to touch on something you just said about serving the right uh, you know, I think I've got you phrased, serve the right content at the right time.
• 14:34
  SPEAKER_02: Here's an example of thought I've used in the past.
• 14:37
  SPEAKER_02: Remember the 2006, the big hurricane that hit, like there's Katrina that hit New Orleans, and so many people moved out of New Orleans, moved out of Louisiana, they start going out to other states.
• 14:49
  SPEAKER_02: Retailers were struggling for a while because they had a whole new mix of people, demographics, coming into states like Texas and other places like that, who are now well, where's the spices I use to cook this?
• 15:00
  SPEAKER_02: You know, where's the shrimps I used to get, you know, because they were used to buying certain things designed for the demographics of Louisiana, but they spread out because they just said I've had enough of the floods.
• 15:09
  SPEAKER_02: Now imagine in that back then you'd had knowledge graph powered solutions to that to know that Seth moved from Louisiana, he likes to have these things and start to serve that content up.
• 15:20
  SPEAKER_02: Lots of opportunity going back to why the sea cares, the sea levels care, is it's an opportunity for profit or volume or share.
• 15:28
  SPEAKER_03: Chris, I know you were uh thinking about uh how this connects to kind of some of the emerging things.
• 15:33
  SPEAKER_03: Do you want to go down?
• 15:35
  SPEAKER_01: Oh, yeah, given given you know, the the the really the basis for this podcast and stuff too is you know, all the data and stuff is important, but how often are you um having folks and people you influence, you know, asking about AI, right?
• 15:49
  SPEAKER_01: And then I would love to you know get that perspective just in general, and then I've got a follow-on with how generative AI is influencing you and or you know what it's what's what's been the outcomes so far.
• 16:02
  SPEAKER_02: So love to get a lot is the easy short answer.
• 16:05
  SPEAKER_02: I'll just stop there and the podcast is boring.
• 16:07
  SPEAKER_02: Sorry, no.
• 16:07
  SPEAKER_02: Um when I first joined uh Ontatex back in early November, machine learning was you know kind of the hot thing, and AI as a more generalized term was the big thing we were talking about because you need ideally, you want to have a knowledge graph power you know to power some of the algorithms behind machine learning to feed that properly so you can intelligently you know expand that.
• 16:27
  SPEAKER_02: Um, so you know, AI as a part of what we do with graph technology is extremely important.
• 16:33
  SPEAKER_02: To to fast forward to the next question, then generative AI, of course, everybody has been exploding with this, and we could have some very interesting dialogue about how how long the hype is going to last and how how real it is.
• 16:45
  SPEAKER_02: But uh, the best indicator of how much interest we've seen in this is we did a webinar, I think just over a month ago, uh, talk about generative AI and knowledge graphs and large language models.
• 16:56
  SPEAKER_02: Had our largest attendee or largest sign-up today we've ever had, over 800 and some odd people signed up for that.
• 17:04
  SPEAKER_02: And you know, we're you know, we're not a huge company.
• 17:06
  SPEAKER_02: We're I think the marketing engine's great, of course.
• 17:08
  SPEAKER_02: I'm biased, but the fact that that many people sign up and the actual uh registration, the actual attendee numbers were significantly higher than I expected.
• 17:17
  SPEAKER_02: So that to me indicates that not only, of course, generative AI, hot topic, large language models, hot topic, but how does that plug into, or how did knowledge gap, excuse me, plug into supporting and training large language models of that corpus of information that's out there?
• 17:33
  SPEAKER_02: People were very engaged.
• 17:34
  SPEAKER_02: And we I continue to get follow-ups coming in, both myself and our salespeople saying, Hey, can I get the link to that webinar?
• 17:40
  SPEAKER_02: I missed it, I want to hear more about it.
• 17:42
  SPEAKER_02: Uh, it's created some good conversations for us to have uh on our side.
• 17:45
  SPEAKER_02: So I got the long answer, but I think that definitely important.
• 17:50
  SPEAKER_02: If I go back to putting that hat on the so what C-level people are spending money on an innovation, we've got to drive innovation, you know, chat GPT, AI is showing up in the boardroom conversations.
• 18:01
  SPEAKER_02: You see that happening, you know.
• 18:03
  SPEAKER_02: They're telling you they're being told go do something with AI.
• 18:06
  SPEAKER_02: That's uh but um what are those opportunities that you can do?
• 18:10
  SPEAKER_02: And that's where knowledge graphs help to uncover those say hidden opportunities in the data to help feed some of those things around that.
• 18:19
  SPEAKER_03: We'll have to have we'll have to have the link in our show notes to that webinar.
• 18:23
  SPEAKER_03: I think I might have missed it, so I need to dig it up as well.
• 18:25
  SPEAKER_03: But I know I saw it coming.
• 18:28
  SPEAKER_01: Do you see too?
• 18:29
  SPEAKER_01: Because I mean you sit square in in you know, marketing the CMO and stuff too, because there's you know, I can't help but think that there's a lot of shadow related IT now that goes on, right?
• 18:40
  SPEAKER_01: How people get content created quicker, faster, cheaper, you know, those types of things.
• 18:46
  SPEAKER_01: What do you see in there?
• 18:47
  SPEAKER_01: And just in terms of the broad aspects of marketing, because part of this, I mean, you you, you know, we go into it, of course, we see all this in AI, but as you guys apply this to what you're seeing, not only internal to a company, but just in marketing in general, you know, what do you what are your you know, thoughts there?
• 19:03
  SPEAKER_01: Because I'm sure you have an opinion.
• 19:05
  SPEAKER_02: And when I first started digging into chat GPT, there was I think that kind of common marketing especial reaction of oh boy, there goes my job.
• 19:12
  SPEAKER_02: Yeah, part of my job, and I've been especially the last few years, has been positioning and messaging and you know, making things sound good.
• 19:18
  SPEAKER_02: And chat GPT can make things sound really, really good.
• 19:22
  SPEAKER_02: Um, although I'm getting to the point you can start to read output and go, that sounds like it might be chat GPT related.
• 19:31
  SPEAKER_02: Yeah, and there's nothing wrong with that.
• 19:32
  SPEAKER_02: I mean, you know, I've used the analogy before of a dishwasher.
• 19:35
  SPEAKER_02: I mean, I'm assuming both of you have dishwashers in your house, so you have loosely some AI that's running part of your household activities, it's washing dishes for you, so you can gain an hour back of your time.
• 19:48
  SPEAKER_02: Okay, so you know, AI is doing similar things for us within the marketing, whether it's the sales, whether it's code, you know, code review.
• 19:56
  SPEAKER_02: Um, from our standpoint, we are we looked at it.
• 19:59
  SPEAKER_02: One of the things we are already doing is connecting, we've got an enhancement in our in our recent releases, connecting Chat GPT to our knowledge graph.
• 20:08
  SPEAKER_02: So there's a lot of power behind that.
• 20:09
  SPEAKER_02: We've also got what's called the on-to-text knowledge graph, where we're using that to tie into our existing corpus of knowledge to feed that back, connecting it with a large large language model concepts to say, hey, here's how you can serve our own content up so we can understand it even better.
• 20:26
  SPEAKER_02: Um, to me, the ability of knowledge graphs to uh you know to make those relationships, to have that understanding of things, to find that context within the data gives me, I'll say, reassurance sometimes that the what comes back from a large language model isn't just a bunch of hallucinations.
• 20:47
  SPEAKER_02: And that's where I think you know, we keep hearing hallucinations in chat GPT and like by having that context applied and verified, you have the potential.
• 20:58
  SPEAKER_02: I'm not saying 100%, potential to get better accuracy and more efficacy of the predictive nature of that large language model as opposed to well, it's just mining everything.
• 21:09
  SPEAKER_02: So that's a lot of directions you can go that one.
• 21:12
  SPEAKER_02: Sorry, that's uh that's a long-winded answer, but there's a lot to it.
• 21:14
  SPEAKER_03: Not at all, not at all.
• 21:16
  SPEAKER_03: And um, I don't know if you can speak more uh specifically to the connections between um uh the knowledge graph and the large language model.
• 21:26
  SPEAKER_03: Do you have any more detail you can provide about exactly how that's being done?
• 21:30
  SPEAKER_03: Are you ingesting your your knowledge into a vector space?
• 21:34
  SPEAKER_03: And then what's how is the knowledge graph kind of being referenced in that context?
• 21:39
  SPEAKER_03: Or is that a little too much in the weeds?
• 21:41
  SPEAKER_02: Yeah, it's probably outside my depth.
• 21:42
  SPEAKER_02: Sorry.
• 21:42
  SPEAKER_02: I I know we have talked about both doing it using a vector and a non vectorized approach, but as far as that's I would want to go any further than that, I'll be I'll be confusing myself more than more than anybody else.
• 21:52
  SPEAKER_03: No worries, no worries, no worries.
• 21:54
  SPEAKER_03: Um, but uh it so have you seen that being deployed in production or Is it mostly POCs these days that you're seeing?
• 22:02
  SPEAKER_02: Right now it's mostly POC with us, it's mostly mostly POCs.
• 22:06
  SPEAKER_02: I've seen some people putting this in place in other organizations, other types of business applications.
• 22:13
  SPEAKER_02: But what's been most interesting just in the last I'd say month, starting to see a little more buzz on you know on the internet and Slinkedin about is this thing going to really take off, or we all jumped into this full speed and are running really fast and not really sure where it's going.
• 22:30
  SPEAKER_02: And I'm I always take a bit of a uh weather eye to the hot, shiny, you know, new object.
• 22:37
  SPEAKER_02: Um, so I'm seeing where it goes, but I've seen you know, mostly proof of concepts, you'll asking questions about it.
• 22:42
  SPEAKER_02: How do you do that?
• 22:43
  SPEAKER_02: But it's what we do, much like when you say I've got you know, what do I start?
• 22:49
  SPEAKER_02: Is asking the question, what is it you're really trying to accomplish?
• 22:54
  SPEAKER_02: And is chat GPT or or fill in the blank of those?
• 22:56
  SPEAKER_02: Is you know, is generative models the right way to go, or could you solve it in a different fashion?
• 23:02
  SPEAKER_02: Is it just a search and browse function that you're trying to get?
• 23:05
  SPEAKER_02: Is it just knowledge discovery you're trying to get?
• 23:08
  SPEAKER_02: Do you need chat GPT?
• 23:09
  SPEAKER_02: So not steering people away from it, but asking the questions to help them understand.
• 23:14
  SPEAKER_02: So they we'll take as much money as they want to give us or for our partners, but our goal is to do the right thing for them.
• 23:23
  SPEAKER_01: Is there anything that concerns you about it?
• 23:25
  SPEAKER_01: Like things that uh you know, you're coming into it from the perspective of, hey, listen, this is all great, you know, see like uh CXO.
• 23:33
  SPEAKER_01: However, one thing I want you to keep in mind is X.
• 23:37
  SPEAKER_01: What is that X for you?
• 23:38
  SPEAKER_01: Like, what are the things that you're advising not only to do, but what not to do?
• 23:43
  SPEAKER_02: Data governance is an easy one.
• 23:45
  SPEAKER_02: Um, so you've got all this great information out there, and depending on which how you listen to it, what the up-to-date timing is, but data governance and compliance jumps out to me.
• 23:56
  SPEAKER_02: How do you, if you're using the existing base of knowledge and it gives you back an answer, do you have the ability to get compliance and verification?
• 24:04
  SPEAKER_02: Do we know this this is the right answer?
• 24:06
  SPEAKER_02: Now, if you're just using a marketing language and it says something pretty and fun, it's probably okay.
• 24:10
  SPEAKER_02: But if you're looking for some very specific answers to a very business and business critical question, you can't accept poor data.
• 24:18
  SPEAKER_02: You can't accept sorry, poor answers driven by non-governed or non-good data for lack of a better term.
• 24:25
  SPEAKER_02: Um, if you don't have a framework, uh you don't have a framework in place that ensures when I ask a question, I'm getting answered from a good, accurate, trusted, verified source or sources of data, then you run the risk of you know, being like overdone, but garbage in, garbage out.
• 24:46
  SPEAKER_02: Um, so that's what we talk a lot about.
• 24:48
  SPEAKER_02: You know, what you know, what's your you know, what if you're doing this kind of an AI initiative, what are the regulatory requirements that you're looking for?
• 24:54
  SPEAKER_02: Um, what kind of risk factor are you willing to accept in doing this process?
• 25:00
  SPEAKER_03: What do you think are the biggest mistakes that organizations are making or the C-suite is making when it comes to AI, large language models, generative um capabilities, and and you know, again, looking at your technology as well.
• 25:16
  SPEAKER_03: But what would you say they're what kind of mistakes are they making?
• 25:18
  SPEAKER_03: How are they not understanding this?
• 25:20
  SPEAKER_03: Or what what should they be doing differently?
• 25:23
  SPEAKER_02: Kind of hopscotching back what I said.
• 25:25
  SPEAKER_02: I think it again it goes back to asking the questions.
• 25:27
  SPEAKER_02: Are they are they reacting to market pressures?
• 25:32
  SPEAKER_02: My competitor is doing something with fill in the blank technology, AI, you know, GPT.
• 25:36
  SPEAKER_02: We better do something too.
• 25:38
  SPEAKER_02: Well, there's there's rationale behind that because again, market share, as I said earlier, is important, but just jumping into it without a plan and just hurrying up to do it is is never going to be a good result.
• 25:50
  SPEAKER_02: I mean, I've been very impressed with seeing some of the apps have come out using Chat GPT, etc.
• 25:54
  SPEAKER_02: But you know, it's not an inconsequential amount of money or time or effort to do these things.
• 26:01
  SPEAKER_02: I think just jumping in before first asking internally the question what are we trying to get out of this?
• 26:07
  SPEAKER_02: What is the you know, what is what you know, document and agree on what it is we're trying to accomplish, going back to to Chris's point about what are the outcomes, put that in place so we have a business-driven need being solved and not just a data-driven need.
• 26:24
  SPEAKER_02: Both are important, but you know, data drip data supports the business outcomes.
• 26:30
  SPEAKER_03: There any place that you're seeing specific success or traction uh in the marketplace, either again with knowledge graphs or with uh knowledge graphs and generative AI, generative AI, what are you seeing that's kind of taking hold?
• 26:44
  SPEAKER_02: What I saw was what I think most interesting and relevant, mainly because I'm planning a trip.
• 26:48
  SPEAKER_02: Uh I'm going to Greece in just over a month now.
• 26:50
  SPEAKER_02: Oh, that's scary.
• 26:52
  SPEAKER_02: Um, is travel agents, travel sites, excuse me, using GPU do using AI to help you to plan a trip where they'll basically you know you take your parameters, you put a couple of things in, bang, it spits out a nice little itinerary, um, which is simplistic by its nature, but I've I played with it a bit and started customizing.
• 27:10
  SPEAKER_02: You know, we're we we like these sorts of things, we don't like long walking tours, and it it made some pretty interesting recommendations.
• 27:17
  SPEAKER_02: And then the fun part was which is why there's a so what it made commercial recommendations for vendors you could use.
• 27:24
  SPEAKER_02: Here's what you use for your taxi.
• 27:25
  SPEAKER_02: We suggest these, and then just say one, but you get like three, these three vendors for XYZ.
• 27:31
  SPEAKER_02: Here's three great places to eat when you're in Crete, but you know, and it's so now gave commercial connections.
• 27:36
  SPEAKER_02: And so I'm assuming someplace in the back end, there's like, hey, you give us some money, maybe what's your name up there.
• 27:42
  SPEAKER_02: To me, that was just a neat example.
• 27:43
  SPEAKER_02: Now, is that a yeah, it's a business-driven example because you're helping to drive commerce of some sort.
• 27:51
  SPEAKER_03: And are you seeing organizations um uh with any success or focus on say conversational commerce asking questions or their product catalog?
• 28:02
  SPEAKER_03: Um, what are your thoughts there?
• 28:04
  SPEAKER_02: Again, going back to the the PIM, the PIM discussion, just wading into that is what I'm seeing.
• 28:09
  SPEAKER_02: Um, we've developed, for example, we've developed what we call a demonstrator for our product information management plugin.
• 28:16
  SPEAKER_02: I'll call it a plugin, which is my my development team will smack my hand.
• 28:18
  SPEAKER_02: It's not really a plug-in, but it's an enhancement to that to work with that product information in order to serve that more quickly.
• 28:26
  SPEAKER_02: It's not chat GPT connected, but it has an AI, the AI uh mindset behind that to feed that.
• 28:33
  SPEAKER_02: How do you going back to that?
• 28:35
  SPEAKER_02: How do you ask the right multi-level convert multi-level questions of your data to get the right answer?
• 28:43
  SPEAKER_02: This goes back to something you were saying a couple threads back, is where again I see the power of graph technology connected with AI is not just asking a one-level question about something, but adding in the you know, using natural language processing to really ask the question to get to where I want to get.
• 29:04
  SPEAKER_02: That's where I'm starting to see some activities, especially with the product information management types, the um digital asset management, to cat to have information categorized, organized, and be able to be queried using AI tools, you know.
• 29:19
  SPEAKER_02: Because again, that's a space I look at a lot.
• 29:20
  SPEAKER_02: That's why I'm most interested in it.
• 29:22
  SPEAKER_02: Right.
• 29:23
  SPEAKER_02: I've seen it pop up in other places.
• 29:24
  SPEAKER_02: I haven't spent time digging into those because not as interesting.
• 29:28
  SPEAKER_03: Yeah.
• 29:28
  SPEAKER_03: And uh are there particular tools or environments that you work uh particularly well with in terms of um e-commerce and product information management?
• 29:38
  SPEAKER_02: Not yet.
• 29:39
  SPEAKER_02: Not we only can't still we're still we are still getting our fingers into this space.
• 29:43
  SPEAKER_02: Yeah, and I came on board, it's one of the things I started talking about early on because of my background with master data management.
• 29:48
  SPEAKER_02: I'm just getting a PIM experience with retail, uh, which is why again I reached out to you guys.
• 29:52
  SPEAKER_02: I saw this this case study that you had done.
• 29:54
  SPEAKER_02: And you know, I've talked with a Gardner analyst about this.
• 29:57
  SPEAKER_02: We've actually we're doing a briefing with them next week, and they're you know, there's they're acknowledging and validating that what it is I'm proposing as a it as being part of a technology stack makes a lot of sense to support a PIM vendor.
• 30:11
  SPEAKER_02: I talked to a damn digital asset management vendor uh last week.
• 30:15
  SPEAKER_02: So yeah, we should bring this into our technology.
• 30:17
  SPEAKER_02: I'm like, yes, you should.
• 30:18
  SPEAKER_02: So there seems to be a lot of people who get it, and now it's just moving it from here to there, right?
• 30:25
  SPEAKER_01: You could definitely, I mean, Seth is a digital apparition anyway, right?
• 30:29
  SPEAKER_01: A digital persona.
• 30:30
  SPEAKER_01: So you can say that I use Seth as a tool, that's fine, you know, that works.
• 30:35
  SPEAKER_02: I've read most of not all, I've read most of the books.
• 30:38
  SPEAKER_02: I I it's the stuff fascinates me.
• 30:40
  SPEAKER_02: It's just we it's I've been in and around data until I look at where I am now.
• 30:44
  SPEAKER_02: I didn't realize how much I've been using, not really working with data in you know in my jobs.
• 30:49
  SPEAKER_02: When it all starts to come together, that's what's been so cool about being at Autitex is seeing how this can all be brought together for pardon me, the greater good of what we do from a business standpoint.
• 31:02
  SPEAKER_02: Um, you know when I read books like what you know, what did you you know, the AI for everybody?
• 31:05
  SPEAKER_02: Like, yeah, that just makes sense.
• 31:08
  SPEAKER_02: And I'm not trying to butter you up, trust me.
• 31:09
  SPEAKER_02: I just oh go ahead by all means butter me up.
• 31:12
  SPEAKER_02: What did you find?
• 31:14
  SPEAKER_03: Go ahead.
• 31:14
  SPEAKER_03: Sorry, go ahead.
• 31:15
  SPEAKER_01: I was just gonna say you can't butter up a digital assistant because they just, you know, they you know, yeah, they just continue, continue in the binary path of of positivity.
• 31:23
  SPEAKER_03: So and and what were the big takeaways from your perspective?
• 31:27
  SPEAKER_03: You're referring to the AI powered enterprise, the book I wrote a few years ago.
• 31:31
  SPEAKER_03: Um, what struck me, you know, when when I first saw Chat GPT is I thought, oh, my work is no longer necessary.
• 31:38
  SPEAKER_03: And then the more I investigated, the more I realized everything in that book that I wrote three years ago is still valid and applicable today.
• 31:44
  SPEAKER_03: What were your biggest takeaways when you uh uh took a look at that book?
• 31:48
  SPEAKER_02: And I wear my bias tab because that's my background, but like when you talked, you talked about, I mean, I was I mentioned PIM, product information management, supply chain.
• 31:56
  SPEAKER_02: You know, you touched a bit about how this all this stuff that we do could be applied against the supply chain.
• 32:01
  SPEAKER_02: I spent nine years in the supply chain vendor side of things.
• 32:04
  SPEAKER_02: Uh another area I've been thinking about.
• 32:05
  SPEAKER_02: I mean, if you look at it from a control tower perspective, just you know, the overall arching view of looking at your supply chain across a whole part of the value chain, connecting the dots.
• 32:15
  SPEAKER_02: Um, you know, we use the term connecting the dots a lot of onto techs, you help to connect those.
• 32:19
  SPEAKER_02: To me, it's like you you what you said in there about you know bringing that together, and I'm summarizing the you know my chapter, that stood out because the regardless of the supply chain provider, they might say they're end-to-end.
• 32:34
  SPEAKER_02: You know, it might be warehouse to to DC, but being able to make sure that information is being shared in an agreed-upon, timely fashion, etc., um, is not an easy process.
• 32:47
  SPEAKER_02: And when you look at the potential to stick a knowledge graph underneath that, that's pretty powerful.
• 32:53
  SPEAKER_02: There is one vendor that I have talked to that does actually have a knowledge or sorry, a graph technology built into their actual offering.
• 33:00
  SPEAKER_02: Um I was reading about the yeah, I've known that the space for them for a while, then I realized they actually have a graph technology, like somebody gets it.
• 33:08
  SPEAKER_02: And then there needs to be more supply chain providers who on who bring in graph technology, right?
• 33:15
  SPEAKER_02: Um, and I think you know, resource subscription framework versus LPG has more expandability power.
• 33:20
  SPEAKER_02: They need to see that.
• 33:21
  SPEAKER_02: So there's a lot of things that get stuck out.
• 33:23
  SPEAKER_02: That was one, obviously, the product information management side of things.
• 33:26
  SPEAKER_02: Um, you know, I think you use the term, at least maybe I'm making that I put it in my head, layered ontology.
• 33:33
  SPEAKER_03: Yes, right.
• 33:34
  SPEAKER_03: Yeah, yeah.
• 33:34
  SPEAKER_02: So it just you know looking at how you and I use Doug's term, so I'm I'm a smaller brain thinker, you know, putting just you've got all this stuff now, putting knowledge on top of that in order to get to intelligence to get to results.
• 33:47
  SPEAKER_03: Right.
• 33:47
  SPEAKER_03: And you know, it's interesting.
• 33:49
  SPEAKER_03: Uh, when you talk about supply chains, I've always said that a physical supply chain is also a digital supply chain, right?
• 33:56
  SPEAKER_03: It's an information supply chain.
• 33:58
  SPEAKER_03: You're moving physical goods, but with those goods, you have to move data and data not only about the journey, the logistics, but the products themselves, the end usage, all of that.
• 34:08
  SPEAKER_03: So it's a really complex ecosystem.
• 34:10
  SPEAKER_03: So that's great.
• 34:11
  SPEAKER_03: Yeah.
• 34:12
  SPEAKER_02: When I was at when I was at JDA, which is now Blue Yonder, they're the largest supply chain company in the world, I introduced a concept called the supply chain grid.
• 34:19
  SPEAKER_02: Um it was a concept.
• 34:21
  SPEAKER_02: We didn't sell it, it was a concept, but it was a way to explain a lot of what you just said there, Seth, is that moving an item from point A to wherever to the finished product, it doesn't always take a straight line.
• 34:32
  SPEAKER_02: In fact, a lot of times it can't because something happens or disruption or whatever it's going to be.
• 34:36
  SPEAKER_02: So if you visualize your grid, your supply chain as a grid with multiple connection points and opportunities, and then you manage your business with that approach, uh, it gave you a lot more risk mitigation, a lot more failover, failover opportunities, more ways to get to the customer, more ways to communicate.
• 34:52
  SPEAKER_02: But it's really been kind of neat to look back at the diagram that I might create a person do then.
• 34:57
  SPEAKER_02: And I look at knowledge graph visualizations now, I'm like, cool.
• 35:02
  SPEAKER_03: So, what are what are the things that might concern you um moving forward, either from your perspective as the CMO or things in the marketplace?
• 35:11
  SPEAKER_03: Uh, are you do you have things that you're concerned with that you're worried about, you know, in terms of how this is evolving, or the messaging, or the uptake, or um adoption, whatever it might be.
• 35:23
  SPEAKER_03: What kinds of therm things uh uh concern you?
• 35:26
  SPEAKER_02: The biggest struggle we continue to have is awareness and understanding.
• 35:31
  SPEAKER_02: Um, I've used the term here a couple of times about not wanting to talk in an echo chamber.
• 35:36
  SPEAKER_02: And by that I mean I don't want us to just sell to people who get it.
• 35:43
  SPEAKER_02: It does help.
• 35:44
  SPEAKER_02: It shortens the sales cycle, it makes it more powerful.
• 35:47
  SPEAKER_02: But if we only sell to people who get it, we're not gonna be talking to the larger business perspective, we're not gonna be talking to the enterprise mindset, we're not gonna be helping to educate how data can be used and reused across an organization using graph technology structure.
• 36:06
  SPEAKER_02: Um, I you know, selfishly, we're we're gonna be missing out on sales opportunities and client names, etc.
• 36:13
  SPEAKER_02: So I think part of it is education to understand why this is so important.
• 36:18
  SPEAKER_02: Going back to this is not a rip and replace, this is an enhancement perspective.
• 36:23
  SPEAKER_02: Those, I think to me, are two of the biggest challenges.
• 36:26
  SPEAKER_02: Generating leads as a you know, as a CMO, you know, leads and market awareness are my are my two bugaboos.
• 36:32
  SPEAKER_02: Generating leads is functionally easier.
• 36:36
  SPEAKER_02: You know, and I don't just mean I'm scanning you because you walked up and took a you know a copy of one of our books or something.
• 36:41
  SPEAKER_02: You know, we've had a conversation.
• 36:42
  SPEAKER_02: Getting a lead is fun, but getting somebody to go, oh, I get it, and I understand at a high level, depending on who you are, how this might help either my business unit or my business or challenges, etc.
• 36:57
  SPEAKER_02: Because it is so potentially powerful, I think that's harder to educate on.
• 37:03
  SPEAKER_01: We find a lot of a lot of folks to that point, Doug, where you you know, like for instance, we're all we're humans still, so there's still a lot of psychology that goes into that answering of why and what's going to be valuable.
• 37:17
  SPEAKER_01: And then also timing is in is perfect, right?
• 37:20
  SPEAKER_01: Because you know as well as I do that you, you know, like it makes me it makes no sense for me to market to you while it's a hundred goddamn degrees in your in your neck of the woods to you know send you know like parkas and stuff for the winter time, right?
• 37:34
  SPEAKER_01: Because winter's coming, yeah, it does does me no good for those kind of things.
• 37:38
  SPEAKER_01: So timing is everything.
• 37:38
  SPEAKER_01: You almost you know can do yourself a disservice like some of those activities.
• 37:42
  SPEAKER_01: But part of it, like you said, is you know, hey, trying to make sure that we understand the who, the when, the what, and the how, and then taking a good step back and saying, is is uh are these technologies going to enable us to do more, see more, get deeper insights, and therefore take more, you know, put things into automation or not.
• 38:03
  SPEAKER_01: And then in fact, a lot of questions I ask organizations too if you look across your customer journey and your 360, are you causing problems or solving problems?
• 38:12
  SPEAKER_01: And can you answer that in each and every one of those, you know, those uh touch points, right?
• 38:17
  SPEAKER_01: Because a lot, yeah.
• 38:19
  SPEAKER_01: It's it's and and you know as well as I do that many of them.
• 38:22
  SPEAKER_01: I love the fact when they come back and they're like, you know what?
• 38:25
  SPEAKER_01: In some aspects, we're hurting, in some aspects, we're actually helping, but many of them have no idea, no idea at all.
• 38:35
  SPEAKER_02: So I mean playing that timing things.
• 38:37
  SPEAKER_02: I again I yeah, please don't try to sell me a parker right now because I would just laugh.
• 38:41
  SPEAKER_02: But let's play that out going back to me, Seth, you mentioned customer data platforms, and we talked about the power behind AI and prediction models and reasoning.
• 38:51
  SPEAKER_02: Going back to my example about traveling.
• 38:53
  SPEAKER_02: So maybe I'm not, but maybe I'm going to Iceland as a part of my journey and I'm gonna spend two or three days over here.
• 38:59
  SPEAKER_02: And because of my search history or because of things I've bought through Amazon, it makes a connection that hey, Doug doesn't normally want to buy a parka in July when he lives in Oklahoma, or I guess it's August now.
• 39:10
  SPEAKER_02: Um, but I can make inferences based on the fact that he's bought a ticket, he's been searching history books or and make recommendations.
• 39:18
  SPEAKER_02: So yeah, you're completely accurate.
• 39:20
  SPEAKER_02: But that is, I think, is one of the powers of that kind of thing is to sell it to the right person at the right time using that re that reasoning.
• 39:31
  SPEAKER_02: If I go to salesman who's just bought Snowflake or just bought you know uh you know Databricks, my timing might not be good, but maybe I need to understand to ask the right questions again to your point.
• 39:43
  SPEAKER_02: Is Databricks helping you, or is there anything that's not helping you with?
• 39:47
  SPEAKER_01: I'm not picking any vendor.
• 39:48
  SPEAKER_01: No, I think I think, but your po your point though is is well valid because oftentimes you're like, listen, great decision.
• 39:54
  SPEAKER_01: Because you know what that tells me is you you're getting ahead of the governance, and um, and you know, just in terms of just putting together a good compliance system in place.
• 40:03
  SPEAKER_01: Now, one thing that you will need to think about going forward, once you get this in place is X, Y, and Z.
• 40:08
  SPEAKER_01: That's where we want to make sure that we help you with those kind of things because you will get there.
• 40:12
  SPEAKER_01: It's not it's not now, but you will, and then that's chat, right?
• 40:16
  SPEAKER_01: Because once you get that in place, because that's generally, you know, those lit, like you said, it's all about asking the right question at the right time.
• 40:22
  SPEAKER_01: I would love to know your perspective too, on especially when we we talk about graphs and product caught catalogs and all this kind of information, you know.
• 40:31
  SPEAKER_01: Um, where you see people missing the boat on always fine-tuning it, not to take the time fine-tuning perspective from general generative models, but in terms of just always care and feeding your graph environment so it's always relevant, right?
• 40:48
  SPEAKER_01: What do you guys think?
• 40:49
  SPEAKER_01: How do you think about that?
• 40:51
  SPEAKER_02: So I go back to if you go back to the product example, the fine-tuning, let's use, and if I'm not in the same mindset path as you, please please correct me, but let's use uh text analytics as a part of that.
• 41:04
  SPEAKER_02: So you've got your catalog of information, your catalog of products that you're publishing out there, and you've got a set of images or hopefully even videos that you're using as part of that.
• 41:14
  SPEAKER_02: Um being able to run text analytics as an example on all the different product categories, all the descriptions you've got for your products, and then you and use the mining.
• 41:25
  SPEAKER_02: So you look at you know how often this this topic shows up, comparing that to search and browse or query activity.
• 41:32
  SPEAKER_02: If people are searching on these terms on a regular basis or these requests on a regular basis, or these two things not in the same line, but the same if they're part of it, you compare that to your text analytics.
• 41:44
  SPEAKER_02: Maybe I should tweak that.
• 41:46
  SPEAKER_02: But do you have to do that every day, every week, every month?
• 41:49
  SPEAKER_02: Probably not, because it's not gonna change that much.
• 41:51
  SPEAKER_02: But that's where we do we do a lot with text analytics, and that's basically we're we've started off, you know, on ontologies and text, you know, text that's what we focus on.
• 42:00
  SPEAKER_02: Um, I think constant tweak, you know, not regular tweaking, not constant, regular tweaking and reanalyzing what it is that you are, how you are organizing and how you're presenting the information based on the data that you have is important, but you know, analysis, paralysis, uh not so good.
• 42:19
  SPEAKER_02: You know, I was in the I was in the Coast Guard for 10 years, and our our um motto was Sperat Paratis, always ready, which we often translate it to semper paralysis.
• 42:31
  SPEAKER_02: Because we're always getting ready to get ready to get ready to get ready.
• 42:34
  SPEAKER_02: Um, I see a lot of organizations.
• 42:35
  SPEAKER_02: Oh, we can't do this yet because we have to progress over perfection in everything that we do.
• 42:42
  SPEAKER_02: Uh most things, you know, some things not, but that's awesome.
• 42:46
  SPEAKER_03: So let's uh turn a little bit toward who you are and uh where you're from and how you got into this space.
• 42:53
  SPEAKER_03: Do you want to give us kind of a world according to uh Doug?
• 42:56
  SPEAKER_02: Sure.
• 42:56
  SPEAKER_02: Uh my journey has been one of evolution, as I always say.
• 43:01
  SPEAKER_02: Um, actually, my master's degree is in counseling education.
• 43:04
  SPEAKER_02: So I spent seven years as a professional counselor doing first level career and and uh that kind of counseling.
• 43:10
  SPEAKER_02: So how I ended up in this role some days, there's days I shake my head.
• 43:13
  SPEAKER_03: But um, I've had a journey of work serves you though.
• 43:16
  SPEAKER_03: I'm sure the counseling serves you in your business.
• 43:18
  SPEAKER_02: So it if you break it down into the kinds of conversations we're having, the conversations I have when we do sales and marketing and when I at a trade show, who are you?
• 43:27
  SPEAKER_02: What what's what's your problem?
• 43:30
  SPEAKER_02: What are you challenged by?
• 43:31
  SPEAKER_02: How do we help you to move past that?
• 43:32
  SPEAKER_02: And let's put together a solution plan.
• 43:34
  SPEAKER_02: It's very similar.
• 43:35
  SPEAKER_02: So I mean I worked for the Nielsen company for almost uh, I guess 10 years in total, went for coming to UC Cena to become trans, you know, worked in Transor to become One World Sync.
• 43:44
  SPEAKER_02: I've just migrated through jobs in different fashions product management, product development, sales, uh, to product marketing.
• 43:51
  SPEAKER_02: Uh, had the opportunity to come here, and it was like I said, it's been neat just to watch it all come together.
• 43:59
  SPEAKER_02: That's the best thing I can.
• 44:00
  SPEAKER_02: And say.
• 44:00
  SPEAKER_02: Um, one of the coolest things is I'm blessed with a wife who has listened to me blather on about data for a long time, uh, and she gets she gets it sometimes better than I do, to the point there's one day we're driving, I was trying to explain knowledge graphs to her, and she'd heard me talk about master data management.
• 44:17
  SPEAKER_02: She's so knowledge graphs are like sentences, and MDM is like the letters of the alphabet.
• 44:24
  SPEAKER_02: And so I got really I was I'm driving, I got really quiet.
• 44:26
  SPEAKER_02: She's like, was that dumb?
• 44:27
  SPEAKER_02: I'm like, no, that was really good.
• 44:30
  SPEAKER_02: And I expanded on it, but I'm like, that's what I look at, and that's what to me is fun about what I've gotten to do in my career is do and work for so many different organizations, talk to so many different clients, listen and learn to what they need, and try to articulate it in a fashion that they go, yeah, that takes away my pain point.
• 44:52
  SPEAKER_01: You mentioned you mentioned too when we when we were talking before that your um your parents were anthropologists, is that correct?
• 44:59
  SPEAKER_01: Yeah, yeah.
• 45:00
  SPEAKER_01: So I'd love to get a perspective on what it's like to be the child of anthropologists.
• 45:05
  SPEAKER_03: Um you learn a lot by just being around.
• 45:11
  SPEAKER_02: Well, so my my parents divorced when I was about eight.
• 45:14
  SPEAKER_02: And I remember my mom might like I I heard the story later.
• 45:17
  SPEAKER_02: I know then my mom told me to you know just see a counselor, make sure Doug was okay.
• 45:21
  SPEAKER_02: And they my the counselor pulled my mom aside afterwards.
• 45:24
  SPEAKER_02: They did the Rorschark test, you know, Rorschark test.
• 45:26
  SPEAKER_02: Long, this is long, obviously.
• 45:27
  SPEAKER_02: I'm old long time ago.
• 45:28
  SPEAKER_02: I said, I'm a little concerned.
• 45:29
  SPEAKER_02: Doug seems to have an unhealthy obsession with death.
• 45:32
  SPEAKER_02: It's like, what do you mean?
• 45:34
  SPEAKER_02: They would show pictures.
• 45:35
  SPEAKER_02: I would say, Oh, that looks like a femur, or that looks like a because bones were around our house.
• 45:40
  SPEAKER_02: You know, I go I go to my dad's lab and you know, there's bones, right?
• 45:43
  SPEAKER_02: So, you know, that's that's growing her up around two PhDs.
• 45:47
  SPEAKER_02: Um yeah, it was me.
• 45:50
  SPEAKER_02: I've got two extremely smart people that are my parents.
• 45:53
  SPEAKER_02: I've been very fortunate in that, different, different backgrounds and interests, but it made me really, I think, take a good look at how people and societies communicate and evolve and how sociology plays into all that, and how it truly is about being a community, which is like bringing data together.
• 46:16
  SPEAKER_02: But it is interesting.
• 46:18
  SPEAKER_03: That's great.
• 46:19
  SPEAKER_03: Um, so what do you do uh outside of work?
• 46:24
  SPEAKER_03: What's fun for you?
• 46:25
  SPEAKER_03: You have stuff you do, the family, you have stuff uh your hobbies, what kind of kind of tell us about the world, uh your your life outside of work.
• 46:34
  SPEAKER_02: I've got 24 year old 24-year-old twins, and they are about eight hours apart from the opposite direction.
• 46:39
  SPEAKER_02: So um don't you know I don't see them as much as I would like to.
• 46:42
  SPEAKER_02: I'd like to see them every day, but they're awesome kids, they've grown up extremely well.
• 46:45
  SPEAKER_02: Um, so you know they're you know, they are a very girl, two boys, two girls, one of each, boy and girl.
• 46:50
  SPEAKER_02: Nice, yeah.
• 46:51
  SPEAKER_02: So hit hit hit the jackpot, and they're both you know uh employed, they're doing well out of college, and they're you know, taking good care of themselves.
• 46:58
  SPEAKER_02: So we did something right in that process.
• 47:00
  SPEAKER_02: Uh my hobby is I do Brazilian jujitsu.
• 47:03
  SPEAKER_02: I've done that for coming up on 16 years, got my black belt uh almost two years ago now, I guess, right?
• 47:12
  SPEAKER_02: Um, so I teach two to three times a week.
• 47:15
  SPEAKER_02: Uh I train when my body says it's okay to do so.
• 47:18
  SPEAKER_02: I mean, I turned I turned 59 a couple of weeks ago, so you know, get a little older, but um it's 4019 in my book.
• 47:26
  SPEAKER_03: That's your very late 40s, 2019.
• 47:28
  SPEAKER_02: Yeah, yeah, it's it's round you round you round down that way, but it's it's I love it because it allows me to I can have a bad day at work and go and get my just butt kicked for an hour, and I walk out like I'm just a the best mood.
• 47:40
  SPEAKER_02: And I think it also keeps the mind sharp because once you get to the elevated porn, actually jujitsu, it's you're not just making move one or move two or move three, you're having to go you're you're you're doing three-dimensional active chess.
• 47:56
  SPEAKER_03: Yeah, it's wonderful for learning and keeping yourself sharp.
• 48:00
  SPEAKER_03: As you know, I do some martial arts as well, and uh definitely uh have done a little bit of Brazilian jujitsu, not not that much.
• 48:08
  SPEAKER_03: I I currently do the most the least uh uh practical uh martial art in the world, which is Aikido.
• 48:14
  SPEAKER_03: It takes a long time, and then uh some Shotokan.
• 48:18
  SPEAKER_03: Uh you did another, you did something else, right?
• 48:20
  SPEAKER_03: Didn't you do that?
• 48:21
  SPEAKER_02: Yeah, I did uh Yoshukai karate for about nine years, I did Taekwondo for just under nine years, and a couple of smattering of other ones.
• 48:28
  SPEAKER_02: Yeah, um, yeah.
• 48:29
  SPEAKER_02: I was I used to be pretty good in the stand-up area, that's why I got into Brazilian Jiu-Jitsu, is that I knew if you took me down the ground, right?
• 48:37
  SPEAKER_03: Very different, yes, very different.
• 48:39
  SPEAKER_03: Everything turns into grappling, yeah.
• 48:41
  SPEAKER_03: Yeah, that's fun.
• 48:43
  SPEAKER_03: That's great.
• 48:44
  SPEAKER_03: Well, we are just about the end of our time, and I have one other uh question for you, kind of an introspective question.
• 48:52
  SPEAKER_03: If you could go back in time to the time you were graduating from college, what might you say to yourself with the perspective that you have now?
• 49:02
  SPEAKER_02: Be more focused.
• 49:04
  SPEAKER_02: Yeah, be more focused.
• 49:05
  SPEAKER_02: I went straight into the military after after high school, and I knew that I was mature enough to know that if I'd gone straight to college, I would not have done very well because I was not mature enough, if that makes any sense.
• 49:17
  SPEAKER_02: Right, sure.
• 49:18
  SPEAKER_02: And when I got to college, the first year, two years, I was not as focused as I as I should have been.
• 49:24
  SPEAKER_02: Um I mean, I don't I'm not a natural learner for lack of a better term.
• 49:29
  SPEAKER_02: I I you know I have to pick things up and really hammer them in my head, but I think being more focused would have made a difference.
• 49:35
  SPEAKER_03: And and do you apply your trainings or your learnings or your philosophy from martial arts into business on a day-to-day basis, would you say?
• 49:43
  SPEAKER_02: Uh that's where I think actually a keto is even more powerful in some ways, because if I come at you with an attack or a I mean, you can you deflect.
• 49:51
  SPEAKER_02: And with Brazilian jujitsu, I do.
• 49:54
  SPEAKER_02: I was actually talking to my class about this last week.
• 49:57
  SPEAKER_02: Um, jujitsu makes you be comfortable in uncomfortable situations.
• 50:03
  SPEAKER_02: So I can walk into any business environment, any trade show, any place, and like I'm maybe a little nervous, and there'd be people I don't know, topics I don't know, conversations, but nobody's gonna try to break my arm or choke me up.
• 50:15
  unknown: Right.
• 50:15
  SPEAKER_02: So there's not a 300-pound person laying on me trying to squash me.
• 50:18
  SPEAKER_03: Right, right, right, right.
• 50:20
  SPEAKER_02: I had worse.
• 50:21
  SPEAKER_03: Yeah, yeah.
• 50:22
  SPEAKER_03: We've had we I used to compare it to something we call special training, which was like marine boot camp in a weekend, but uh I always thought if you can do that, there's everything else is gravy, right?
• 50:33
  SPEAKER_02: Yeah, well that's great, guys.
• 50:35
  SPEAKER_02: Always adapting and improvising is the best way to look at it.
• 50:38
  SPEAKER_03: Yep, yep, and knowing you can get through it, whatever it is.
• 50:41
  SPEAKER_03: Yeah, that's great.
• 50:43
  SPEAKER_03: Well, Doug, it's really been a pleasure having you.
• 50:45
  SPEAKER_03: I've really enjoyed our conversation.
• 50:47
  SPEAKER_03: Thank you so much for uh spending your time with us today.
• 50:50
  SPEAKER_02: Stuff and Chris, thank you.
• 50:51
  SPEAKER_02: This has been a lot of fun.
• 50:52
  SPEAKER_02: I really appreciate it.
• 50:53
  SPEAKER_03: And we'll have your contact information in the show notes.
• 50:56
  SPEAKER_03: You can be found on LinkedIn.
• 50:57
  SPEAKER_03: It looks like Doug Kimball is your LinkedIn.
• 51:00
  SPEAKER_03: Uh, we'll have your Twitter and company address all in the uh uh show notes.
• 51:04
  SPEAKER_03: Uh and again, thank you so much for your time today.
• 51:07
  SPEAKER_03: It's been great to chat.
• 51:11
  SPEAKER_00: Thank you for joining us on another deep dive into AI innovation.
• 51:15
  SPEAKER_00: Join us next time when we introduce another expert and discuss how to maximize AI in your world.
• 51:21
  SPEAKER_00: Early AI Podcast is sponsored by Early Information Science, CMS Wire, and Marketing AI Institute.
• 51:27
  SPEAKER_00: To learn more about us, visit early.com at a D A R L A Y.com.
• 51:34
  SPEAKER_00: See you next time.