1 00:00:00,160 --> 00:00:03,680 Essentially, it's a swarm of models, AI models that 2 00:00:03,680 --> 00:00:07,480 emulate human cognition and emotion and become highly 3 00:00:07,480 --> 00:00:11,040 predictive of behavior across populations. So you're 4 00:00:11,040 --> 00:00:14,680 creating synthetic populations of people that are then situated 5 00:00:14,680 --> 00:00:18,360 in context. Forget Personas, Jill Axline is building 6 00:00:18,360 --> 00:00:21,920 synthetic populations that predict real human behavior and that changes 7 00:00:21,920 --> 00:00:23,840 everything. Keep watching to learn how. 8 00:00:32,060 --> 00:00:32,380 Foreign. 9 00:00:39,100 --> 00:00:42,380 Hello, and welcome to Data Driven, the podcast. We explore the 10 00:00:42,380 --> 00:00:46,060 exploding world of artificial intelligence, data science, and of 11 00:00:46,060 --> 00:00:49,540 course, none of this would be possible without the underlying data 12 00:00:49,540 --> 00:00:53,260 engineering. And with me on this road trip down the information 13 00:00:53,420 --> 00:00:57,260 superhighway of the future and Buzzwords 14 00:00:57,920 --> 00:01:01,680 is my most favorite data engineer in the world. How's it 15 00:01:01,680 --> 00:01:05,320 going, Andy? Hey, Frank. It's going pretty good. How are you? I'm 16 00:01:05,320 --> 00:01:08,560 doing all right. I'm still wearing the hipster glasses because they 17 00:01:10,160 --> 00:01:13,680 were recording this about post 3 weeks since my concussion. 18 00:01:13,920 --> 00:01:17,600 And as we were telling our guest in the virtual green room that 19 00:01:18,080 --> 00:01:21,200 we kind of owe the show's name to a concussion. 20 00:01:21,760 --> 00:01:25,560 So true, folks who, longtime listeners, know 21 00:01:25,560 --> 00:01:29,240 the lore, so we won't bore them or waste any of our guests precious 22 00:01:29,240 --> 00:01:33,080 time. With us, we have Jill axlein, who 23 00:01:33,080 --> 00:01:36,760 is Ph.D. and is the co founder and 24 00:01:36,760 --> 00:01:40,520 CEO of Mavera, which is an 25 00:01:40,520 --> 00:01:44,320 interesting company and Maverick Era is what I'm told it's short for. 26 00:01:44,880 --> 00:01:48,080 So welcome to the show, Jill. Hey, thanks. So happy to be here. 27 00:01:48,480 --> 00:01:52,240 Yeah. So you also have three kids and. I 28 00:01:52,240 --> 00:01:55,520 have three kids. Andy has three. Three plus two. 29 00:01:55,920 --> 00:01:59,520 Yes, that's. I think, 30 00:01:59,760 --> 00:02:02,720 I think there's a correlation between number of kids and gray hairs. 31 00:02:04,480 --> 00:02:08,200 I know I have kids and five grandchildren, so there you go. But 32 00:02:08,200 --> 00:02:11,760 I'm old. I'm just saying you have an age today, 33 00:02:12,080 --> 00:02:14,800 you know. So 34 00:02:15,760 --> 00:02:18,480 what does Mavera do and 35 00:02:19,990 --> 00:02:23,670 what is brand and business meaning for? What does that mean in 36 00:02:23,670 --> 00:02:26,310 high growth. Companies, brand and business. 37 00:02:27,670 --> 00:02:30,630 I totally botched that. I'm sorry. I'll blame the concussion because I can do that 38 00:02:30,630 --> 00:02:34,310 for another week or so. So what exactly does Mavera 39 00:02:34,310 --> 00:02:38,070 do? Sure. So essentially it's a swarm of 40 00:02:38,070 --> 00:02:41,590 models, AI models that emulate human 41 00:02:41,590 --> 00:02:45,350 cognition and emotion and become highly predictive of behavior 42 00:02:45,430 --> 00:02:49,190 across populations. So, so you're creating synthetic populations 43 00:02:49,190 --> 00:02:52,190 of people that are then situated in context. 44 00:02:52,830 --> 00:02:56,550 So as opposed to a model that's trained six months ago and 45 00:02:56,550 --> 00:02:59,710 then is rapidly trying to iterate, it actually 46 00:02:59,870 --> 00:03:03,550 pulls its synthetic database will update on a 47 00:03:03,550 --> 00:03:07,390 second to second basis. So you always look at your population in 48 00:03:07,390 --> 00:03:10,510 situ. Additionally, I would say 49 00:03:11,390 --> 00:03:15,070 it provides a really strong pulse of what that population 50 00:03:15,150 --> 00:03:18,600 looks like within the context of your business or your vertical. 51 00:03:18,760 --> 00:03:22,520 Because we support a foundation with deep business context 52 00:03:22,760 --> 00:03:26,400 that takes into account not just your business from the time that it 53 00:03:26,400 --> 00:03:30,000 was instantiated, but it also is updating 54 00:03:30,000 --> 00:03:32,760 temporally and it creates relational, 55 00:03:34,280 --> 00:03:38,040 like relational connections across your business. So for instance, 56 00:03:38,360 --> 00:03:42,000 if there's a marketing spend five years ago or about 57 00:03:42,000 --> 00:03:45,720 the same time that you launch your flagship product or a secondary product, 58 00:03:46,040 --> 00:03:49,780 it's going to show a lot of data on how the context 59 00:03:49,860 --> 00:03:52,660 around that might have influenced your outcomes. 60 00:03:53,380 --> 00:03:56,700 So I guess like long and short of it is you have 61 00:03:56,700 --> 00:04:00,340 populations situated in context and wrapped around your business, 62 00:04:00,740 --> 00:04:04,580 and you can use that pretty expeditiously to make 63 00:04:04,580 --> 00:04:08,340 decisions in a much less expensive way than most market research 64 00:04:08,420 --> 00:04:12,180 or, you know, strategy research, strategy based research. 65 00:04:12,340 --> 00:04:15,400 It's almost like you're taking kind of like the SIMS 66 00:04:15,800 --> 00:04:19,440 approach of having these individual entities, I wouldn't call them 67 00:04:19,440 --> 00:04:23,000 agents because it doesn't sound like they're agents. It sounds like they're simulated entities, like 68 00:04:23,000 --> 00:04:26,360 you said. Right, exactly. That's interesting. Is there like a. 69 00:04:26,680 --> 00:04:29,080 That. That's an interesting approach because that does, 70 00:04:30,360 --> 00:04:34,000 it probably doesn't completely insulate you from model drift, but it 71 00:04:34,000 --> 00:04:37,240 probably does a good job of, well, 72 00:04:38,360 --> 00:04:42,040 we're having a massive windstorm and it's like, you know, negative, whatever. Outside in your 73 00:04:42,040 --> 00:04:45,860 Chicago, it's really cold. It's always sunny and it's always sunny in farmville, as 74 00:04:45,860 --> 00:04:49,380 I like to tell Andy. But, but I mean, you can 75 00:04:49,380 --> 00:04:52,220 insulate against a certain amount of cold, but you can't really stop it. 76 00:04:52,700 --> 00:04:55,820 That's right to think about it. So you can't really stop model drift, but you 77 00:04:55,820 --> 00:04:59,619 probably can prolong how, how, how long your 78 00:04:59,619 --> 00:05:03,420 models are valid for this by this approach. So that's correct. In 79 00:05:03,420 --> 00:05:06,940 addition to that, something that I've pushed on because I've been an 80 00:05:06,940 --> 00:05:10,620 advisor with this team for well over a year. And 81 00:05:10,620 --> 00:05:14,180 since I'm a ph dork and I, you know, I'm always looking at evidence. 82 00:05:14,180 --> 00:05:17,440 Evidence Ev. I was the original skeptic to synthetic 83 00:05:17,440 --> 00:05:21,240 populations. In my last role at Morningstar, I built our market research 84 00:05:21,320 --> 00:05:24,720 team. And when I was first introduced to the idea of 85 00:05:24,720 --> 00:05:28,440 synthetic populations, I was like, you know, tons of skepticism. 86 00:05:28,440 --> 00:05:31,960 I think the big thing here is they've built in a level of AI 87 00:05:31,960 --> 00:05:35,640 governance around things like drift, but also to 88 00:05:35,640 --> 00:05:38,920 model the difference between evidence and inference. And so 89 00:05:39,320 --> 00:05:42,920 they're looking for confidence scores. They'll gather first party data 90 00:05:42,920 --> 00:05:46,720 around your population and then create a synthetic data layer on top 91 00:05:46,720 --> 00:05:49,680 of that. And a good example would say 92 00:05:50,240 --> 00:05:53,920 asset managers like ice cream. Asset managers like cold 93 00:05:53,920 --> 00:05:57,320 things. They like cold, wet things, they like cold, wet, sweet things. And then a 94 00:05:57,320 --> 00:06:01,160 coefficient is that assigned to each of those new synthetic data points. And so 95 00:06:01,160 --> 00:06:04,480 while it makes a more robust data set in the 96 00:06:04,480 --> 00:06:08,320 billions that allows it to draw inference, it's also accounting 97 00:06:08,320 --> 00:06:11,960 for again, what, what's based on evidence and what's based, what is 98 00:06:11,960 --> 00:06:15,680 inference of the machine. And then there's also a governor across 99 00:06:15,760 --> 00:06:19,360 this swarm of models. So it's going to call on the right model 100 00:06:19,920 --> 00:06:23,360 for the right facet of human thinking or 101 00:06:23,360 --> 00:06:26,800 feeling that it's trying to construct. And so 102 00:06:27,360 --> 00:06:31,040 I think in doing that it creates safeguards around confidence. So 103 00:06:31,040 --> 00:06:34,840 we, we produce confidence scores, it will give a spread of opinion across 104 00:06:34,840 --> 00:06:38,560 a population. So unlike a custom GBT or 105 00:06:38,950 --> 00:06:42,510 a Persona and some pre existing platforms that are emulating 106 00:06:42,510 --> 00:06:45,590 language, it's actually taking a look at 107 00:06:46,070 --> 00:06:49,590 where's their entropy across emotional response and cognitive 108 00:06:49,590 --> 00:06:52,990 response in this data set and what does that look like in the spread of 109 00:06:52,990 --> 00:06:56,630 opinion for that audience. So it'll tell you the nature of the spread 110 00:06:56,630 --> 00:07:00,230 and where that spread is happening. So now you can account for almost, 111 00:07:00,310 --> 00:07:04,070 you know, sub segmentation within the population. And that might 112 00:07:04,070 --> 00:07:07,820 look very different at the top of the funnel when we're looking at thought leadership 113 00:07:07,820 --> 00:07:11,620 topics versus the bottom of the funnel in marketing where we're thinking of features, 114 00:07:11,620 --> 00:07:15,180 functions, benefits, et cetera. And so 115 00:07:15,180 --> 00:07:18,300 that allows at least marketers, but I think others, 116 00:07:18,540 --> 00:07:22,300 anyone go to market to really think about what is their message for the right 117 00:07:22,300 --> 00:07:25,740 audience at the right time based on, you know, where they are in their 118 00:07:25,740 --> 00:07:29,500 buyer's journey. And so that to me is a little bit 119 00:07:29,500 --> 00:07:33,180 different because I would say the last facet of this is 120 00:07:33,180 --> 00:07:36,440 the response stability. We're also providing a level of 121 00:07:37,240 --> 00:07:41,080 test retest reliability. If you go into ChatGPT 122 00:07:41,240 --> 00:07:44,120 recently, someone was flaming me because I've never made 123 00:07:44,760 --> 00:07:48,120 caramelized onions. And so, you know, as a joke, he kind of went in and 124 00:07:48,120 --> 00:07:51,320 said how many people who are 40 something, you know, like know how to make 125 00:07:51,320 --> 00:07:54,840 caramelized onions? And these percentages swung 126 00:07:55,800 --> 00:07:59,520 quite significantly from the first time he queried to the second time to the 127 00:07:59,520 --> 00:08:03,030 third time. Whereas we're looking at population response stability 128 00:08:03,110 --> 00:08:06,790 and modeling that, projecting it into the future and looking at the trend 129 00:08:06,790 --> 00:08:10,350 line from the past on how this population would continuously 130 00:08:10,350 --> 00:08:14,190 answer the question. So I kind of guess like when we think about model 131 00:08:14,190 --> 00:08:17,630 drift, I think that's likely inevitable. But if you're 132 00:08:17,630 --> 00:08:21,270 situating and updating with minute to minute context and then you're surfacing 133 00:08:21,270 --> 00:08:24,630 some of these governance factors around what the Outputs are, 134 00:08:26,220 --> 00:08:29,940 we're getting to a closer place where we can actually be collaborate collaborators 135 00:08:29,940 --> 00:08:33,340 with the AI and govern it and then build, 136 00:08:33,580 --> 00:08:36,220 you know, a greater level of trust is the hope. 137 00:08:37,020 --> 00:08:40,540 That's interesting. I'm glad you addressed the skepticism because that was going to be my 138 00:08:40,540 --> 00:08:43,500 next question. Like, how do you know this is real? How do you know that 139 00:08:43,740 --> 00:08:47,340 it's accurate? The other question I had, and sorry, Andy, 140 00:08:48,300 --> 00:08:50,540 I had a bunch of monster energy drinks today. 141 00:08:52,940 --> 00:08:56,700 You could probably run different simulations, like in 142 00:08:56,700 --> 00:09:00,500 parallel, right. Assuming you had the compute. So 143 00:09:00,500 --> 00:09:04,260 you can see if this happens, if that happens, right. If there's 144 00:09:04,980 --> 00:09:08,260 a recession, people are going to do this, go this way. If there's a boom, 145 00:09:08,740 --> 00:09:12,500 if it kind of meanders somewhere in the middle, you could probably run 146 00:09:14,660 --> 00:09:18,100 only limited to what compute you have, right? I mean, 147 00:09:18,340 --> 00:09:21,900 yeah, I mean, it's a credit based system. So, you know, you buy 148 00:09:21,900 --> 00:09:25,600 credits like a tank of gas and it's going to, you 149 00:09:25,600 --> 00:09:29,400 know, give you enough gas to, to build whatever it is you 150 00:09:29,400 --> 00:09:32,640 want within limits. But I would say, 151 00:09:33,840 --> 00:09:37,520 yeah, I don't think you're really, yeah, I don't think you're really 152 00:09:37,520 --> 00:09:40,880 restricted in terms of what outputs look like on, on a scenario 153 00:09:40,880 --> 00:09:44,560 analysis. I think obviously if the more 154 00:09:44,560 --> 00:09:48,040 data we have, let's call it for a specific company, when I was working at 155 00:09:48,040 --> 00:09:51,680 Morningstar, that's 40 plus years of data on the back end in 156 00:09:51,680 --> 00:09:55,330 that deep business context, that makes that prediction that much easier. 157 00:09:56,690 --> 00:10:00,530 And so I think it also depends on what's coming into the model and 158 00:10:01,570 --> 00:10:05,410 what its power is and its ability to be predictive. I 159 00:10:05,410 --> 00:10:08,530 guess I should say that's cool. Because I think this is an interesting, it seems 160 00:10:08,530 --> 00:10:11,570 like it's an interesting mix of kind of predictive modeling and 161 00:10:12,050 --> 00:10:15,890 LLMs. Right. Because predictive models, I mean, they're not 162 00:10:15,890 --> 00:10:19,390 new. Right, but they're not. But they do. I think 163 00:10:19,390 --> 00:10:23,030 they're, they're traditionally, they're 164 00:10:23,030 --> 00:10:26,710 very susceptible to drift. Right. But 165 00:10:26,710 --> 00:10:30,350 I think also by simulating the individual actors, because a society 166 00:10:30,350 --> 00:10:34,150 and economy, a customer base is, consists of, you know, 167 00:10:34,150 --> 00:10:37,990 X number of, you know, not sovereign 168 00:10:37,990 --> 00:10:41,190 but unique individuals that are going to have certain 169 00:10:41,830 --> 00:10:45,450 personality traits. And some of those you kind of can 170 00:10:45,450 --> 00:10:49,010 guess from. Like you said, you know, asset managers. Asset 171 00:10:49,010 --> 00:10:52,730 managers, everybody likes ice cream, but asset managers probably really 172 00:10:52,810 --> 00:10:55,850 like luxury cars. I'm going to go out on a limb. Right, 173 00:10:56,410 --> 00:11:00,249 right. And probably how much the, how many luxury cars they have and which model 174 00:11:00,249 --> 00:11:03,890 of luxury car they have is probably going to determine, is probably not, not 175 00:11:03,890 --> 00:11:07,130 determine how successful they are. But it's probably a correlation between 176 00:11:07,530 --> 00:11:11,360 how successful they are versus like how not. You know, I don't 177 00:11:11,360 --> 00:11:14,640 know. I. If you're an asset manager and you're driving around the Hyundai, 178 00:11:15,920 --> 00:11:18,880 there's gotta be a good story behind that. That's right. 179 00:11:19,840 --> 00:11:23,560 I agree with you. And I think again, when 180 00:11:23,560 --> 00:11:27,040 you can ask the synthetic audience and pull them, you can start to get into 181 00:11:27,040 --> 00:11:29,680 further nuance whether those are B2B 182 00:11:30,400 --> 00:11:33,360 dimensions of, you know, like firm type, role type, 183 00:11:33,760 --> 00:11:37,400 etc. AUM or it can get into that more 184 00:11:37,400 --> 00:11:40,830 psychographic or it can get into start, start to break down 185 00:11:40,830 --> 00:11:44,390 archetypal differences and you know, all of those 186 00:11:44,390 --> 00:11:48,150 then can be mapped into attributes that are built into the channels where we 187 00:11:48,150 --> 00:11:49,150 communicate with them. 188 00:11:52,430 --> 00:11:55,989 Go ahead, Andy. I don't want to hog the mic. No, no, it's all good. 189 00:11:55,989 --> 00:11:59,790 I'm fascinated and 190 00:12:00,030 --> 00:12:03,870 kind of playing off your, your idea of model drift, Frank, 191 00:12:04,190 --> 00:12:08,040 and your questions along those lines. I 192 00:12:08,040 --> 00:12:11,640 mean, in one sense I would say, you know, 193 00:12:11,640 --> 00:12:15,400 a synthetic audience or you know, a synthetic sample 194 00:12:15,400 --> 00:12:19,040 or cohort, however you want to classify that. Is 195 00:12:19,520 --> 00:12:23,000 model drift happening in that 196 00:12:23,000 --> 00:12:26,400 context is probably not unheard of because 197 00:12:26,880 --> 00:12:30,400 there's cultural drift. And if you're looking for 198 00:12:30,960 --> 00:12:34,200 ways to effectively simulate that 199 00:12:34,760 --> 00:12:38,480 and run marketing campaigns against, you know, the 200 00:12:38,480 --> 00:12:42,240 synthetic cohort, it doesn't strike me 201 00:12:42,240 --> 00:12:46,040 as out of the realm of possibilities that you may want 202 00:12:47,000 --> 00:12:50,600 some of that you may want to even tune for, especially 203 00:12:50,600 --> 00:12:53,000 if you're looking at a younger audience. 204 00:12:54,200 --> 00:12:57,960 There are emerging trends that come out of 205 00:12:57,960 --> 00:13:01,370 those demographics. It's just part of the nature of those 206 00:13:01,370 --> 00:13:04,930 demographics. I mean, I'd love to hear your thoughts on. On that. 207 00:13:05,970 --> 00:13:09,650 Yeah, I mean, I don't know that it's a function of. 208 00:13:09,890 --> 00:13:13,410 I don't want to make it like a generational distinction, but I do think 209 00:13:13,410 --> 00:13:16,770 that anything that's current to context is going to 210 00:13:17,170 --> 00:13:20,890 impact on a minute to minute basis in some cases how 211 00:13:20,890 --> 00:13:24,450 the population is going to make decisions and what level of like 212 00:13:24,450 --> 00:13:28,200 arousal they have. And I don't mean that in the, you know, cheeky 213 00:13:28,200 --> 00:13:31,280 sort of way, but I would say like we're working with 214 00:13:32,160 --> 00:13:35,880 an index team in financial services and they asked me on the spot, 215 00:13:35,880 --> 00:13:39,200 can you please model a high net worth investor in Denmark? 216 00:13:40,000 --> 00:13:43,840 You know, and this was last week just to, just to say, are you thinking 217 00:13:43,840 --> 00:13:47,400 about, you know, rebalancing out of blah, blah, 218 00:13:47,400 --> 00:13:51,160 blah, US broad index? And you know, the 219 00:13:51,160 --> 00:13:54,720 answer was not immediately, but here's my thinking on that 220 00:13:55,650 --> 00:13:59,010 and here's what I would be investing in instead. So now they're trying to think 221 00:13:59,010 --> 00:14:02,770 through what's their messaging around outflows in that 222 00:14:02,930 --> 00:14:06,690 predominant US broad index? And then how are we 223 00:14:06,690 --> 00:14:10,210 surfacing the rest of our family of indexes in its 224 00:14:10,210 --> 00:14:13,890 stead? And then he asked, how does this, does 225 00:14:14,290 --> 00:14:17,810 the audience, is there a large spread here? And if so, 226 00:14:17,890 --> 00:14:21,370 you know, what is the nature of that? So now we can think about 227 00:14:21,370 --> 00:14:25,130 discrete campaigns across this population, which 228 00:14:25,130 --> 00:14:28,490 is pretty narrow of, you know, ultra high net worth investors in 229 00:14:28,490 --> 00:14:31,850 Denmark. Right. So I think it's 230 00:14:31,850 --> 00:14:35,690 applicable depending on what, what is that trigger, you know, that what 231 00:14:35,690 --> 00:14:39,450 is that zero moment of truth for any given population that is going to be 232 00:14:39,450 --> 00:14:42,290 influenced by their immediate context. And 233 00:14:43,090 --> 00:14:46,930 you know, with that responsibility score, we can then tell them this is something 234 00:14:46,930 --> 00:14:50,730 we think will persist over time versus this is ephemeral. And based on what's 235 00:14:50,730 --> 00:14:54,370 happening in the news around tariffs today. So here's something to push out in 236 00:14:54,370 --> 00:14:57,710 your channels today versus here's something to build into, 237 00:14:58,510 --> 00:15:02,350 you know, a long tail campaign and how to think about product strategy in 238 00:15:02,350 --> 00:15:06,110 a different sort of way. That, that's pretty fascinating. 239 00:15:06,590 --> 00:15:09,790 So pivoting just a little bit, you, 240 00:15:10,350 --> 00:15:13,310 you mentioned quite a few instances of 241 00:15:13,470 --> 00:15:16,750 incorporating evidence into this. And I would 242 00:15:16,750 --> 00:15:19,710 imagine that I'm an engineer. Okay, that's a warning. 243 00:15:21,150 --> 00:15:24,190 So, so is our cto. I'm getting used to it. 244 00:15:25,230 --> 00:15:28,930 I think about open and close loops all the time. It's just, you know, I 245 00:15:28,930 --> 00:15:32,410 don't even have to think about thinking about it. It just happens. But 246 00:15:33,850 --> 00:15:37,130 being able to, to become predictive 247 00:15:37,930 --> 00:15:41,730 and have that feedback where you, you 248 00:15:41,730 --> 00:15:45,290 made some, you know, you made some prediction, some predictive 249 00:15:45,290 --> 00:15:49,050 analytic, and then you collect evidence on 250 00:15:49,050 --> 00:15:52,770 how accurate you were and not just, you 251 00:15:52,770 --> 00:15:56,570 know, percentage wise, it doesn't really apply that much, especially in 252 00:15:56,730 --> 00:15:57,690 marketing type 253 00:16:00,630 --> 00:16:04,150 and especially in the age of AI where you can collect information and feed it 254 00:16:04,150 --> 00:16:06,710 back into the system as training data, 255 00:16:08,070 --> 00:16:11,870 effectively as responses to prompts. So the 256 00:16:11,870 --> 00:16:15,270 prompts themselves become part of the data 257 00:16:15,430 --> 00:16:18,870 that goes in and then the outcome that was 258 00:16:18,870 --> 00:16:22,390 predicted, that's very easy to see. That 259 00:16:24,230 --> 00:16:27,180 part happens. But then supplying the evidence 260 00:16:27,900 --> 00:16:31,340 you predicted this, the delta between the 261 00:16:31,900 --> 00:16:35,700 predicted and the actual, that's evidence. And 262 00:16:35,700 --> 00:16:39,540 so being able to quantify that, being able to 263 00:16:39,540 --> 00:16:43,340 feed that back into the engine, I think in early 264 00:16:43,340 --> 00:16:47,020 2026, as we're talking about this, we've not 265 00:16:47,100 --> 00:16:50,220 had the ability to, 266 00:16:51,350 --> 00:16:55,190 I'd say in, you know, in, in natural language, to provide that 267 00:16:55,190 --> 00:16:58,630 sort of information with any sort of confidence that 268 00:16:58,870 --> 00:17:02,590 the algorithm that we're supplying that information to, that feedback, 269 00:17:02,590 --> 00:17:06,110 closing the loop on the evidence, supplying the 270 00:17:06,110 --> 00:17:09,630 evidence, we just hadn't had the confidence that the 271 00:17:09,630 --> 00:17:13,390 machine was going to understand what we meant. And one of the 272 00:17:13,390 --> 00:17:17,199 things that sort of slipped into invisibility over the 273 00:17:17,199 --> 00:17:20,839 past, gosh, what's it been, three years and a few 274 00:17:20,839 --> 00:17:24,399 months since Chat GPT was released? 275 00:17:25,439 --> 00:17:29,079 Is that the model mostly understands what you're 276 00:17:29,079 --> 00:17:32,559 saying now. And I mean by, by mostly some number well above 277 00:17:32,559 --> 00:17:36,319 90%, you know, it's going to get what you mean 278 00:17:37,199 --> 00:17:40,679 and when it hallucinates, you know, it's going to be because it 279 00:17:40,679 --> 00:17:44,360 misunderstands what you said, not because it went off, you 280 00:17:44,360 --> 00:17:47,920 know, and started interpolating what you said and 281 00:17:47,920 --> 00:17:51,600 made something completely different out of it. It's the way it was 282 00:17:51,600 --> 00:17:55,440 stated, wasn't quite clear. And nowadays 283 00:17:55,440 --> 00:17:59,120 I hang out mostly in Claude and Claude code. 284 00:17:59,760 --> 00:18:03,440 So when I'm going back and forth with, you know, with the engine, 285 00:18:04,560 --> 00:18:08,240 it's, especially in Claude code, it very often 286 00:18:08,240 --> 00:18:12,070 will pause the conversation and stop and say, hey, I have this question, 287 00:18:12,070 --> 00:18:15,430 you know, and here's the options. I think you're, you know, based on what you 288 00:18:15,430 --> 00:18:18,990 said, I give you 1, 2, 3. And then number four is you just type 289 00:18:18,990 --> 00:18:22,790 and tell me if I completely missed it. And I rarely find myself 290 00:18:22,790 --> 00:18:26,590 on that bottom option. Most of the time I'm picking the, the 291 00:18:26,590 --> 00:18:30,390 top option, which the one it thinks is most likely. And 292 00:18:31,430 --> 00:18:34,630 so having having that sort of evidence based 293 00:18:35,430 --> 00:18:38,630 feedback, number one, be so much easier 294 00:18:39,180 --> 00:18:42,820 than it is before. And so I can see that limiting model 295 00:18:42,820 --> 00:18:46,660 drift. I can also see it kind of making 296 00:18:46,660 --> 00:18:49,900 your predictions align with 297 00:18:51,180 --> 00:18:55,020 the timescale that you mentioned. So not just the population 298 00:18:55,180 --> 00:18:58,220 being so, so small, which is 299 00:18:59,260 --> 00:19:02,980 infinitely harder than dealing with big data, right? Dealing with a 300 00:19:02,980 --> 00:19:06,710 small set of data. How do you predict in all of that? And before I 301 00:19:06,710 --> 00:19:10,070 ramble anymore, I'll just stop and let you respond. How about that? 302 00:19:11,350 --> 00:19:15,070 Well, it's interesting and I don't want to get over my skis 303 00:19:15,070 --> 00:19:17,270 because this is really where our CTO shines. 304 00:19:18,790 --> 00:19:21,190 He has the ability to create 305 00:19:22,470 --> 00:19:26,150 some audiences out of what he would say he would call dark 306 00:19:26,150 --> 00:19:29,950 matter. The best way for me to think that through is when I look at 307 00:19:29,950 --> 00:19:33,550 a tree and I see its various branches. I'm looking at the 308 00:19:33,550 --> 00:19:37,310 tree to define the tree, but there's so much more sky 309 00:19:37,710 --> 00:19:41,150 and negative space around that tree that also defines it. 310 00:19:41,630 --> 00:19:45,150 And so he's starting to look at data and how it affects other 311 00:19:45,150 --> 00:19:48,790 data and then putting that in context and using that 312 00:19:48,790 --> 00:19:52,550 kind of negative space to then define the audience that's 313 00:19:52,550 --> 00:19:56,270 so small. So that is, you know, in the case 314 00:19:56,270 --> 00:20:00,010 of when I was at Morningstar, Acid owners, really small group of 315 00:20:00,010 --> 00:20:03,770 people, but one that Morningstar really wanted to understand a 316 00:20:03,770 --> 00:20:07,050 lot better. And so that institutional audience, they're 317 00:20:07,050 --> 00:20:10,890 regulated. It's hard to, you know, get permissions because they're so small. 318 00:20:10,890 --> 00:20:14,490 Their time is worth a lot. So it's an expensive panel to construct. 319 00:20:14,970 --> 00:20:18,690 And here he was able to build from again, like that negative 320 00:20:18,690 --> 00:20:22,210 space to then recreate the audience. And, and he is 321 00:20:22,210 --> 00:20:25,770 surfacing that confidence variable. And if there is a hallucination, 322 00:20:25,850 --> 00:20:29,610 hallucination risk, it's tagged and it will prompt you for what sort of 323 00:20:29,610 --> 00:20:33,290 data it then needs. Or it's going to say, actually have to refractor the 324 00:20:33,290 --> 00:20:37,050 audience a little bit differently. There's too much entropy for me to continue and 325 00:20:37,050 --> 00:20:40,410 it will go and run it again. So. And again, I don't want to get 326 00:20:40,410 --> 00:20:43,770 over my skis because I'm the social scientist in the mix, but that's how it's 327 00:20:43,770 --> 00:20:47,610 been described to me that I can, I can best understand it. That makes 328 00:20:47,610 --> 00:20:50,570 a lot of sense actually. And like you can kind of, I think there's a 329 00:20:50,570 --> 00:20:53,690 lot of inference here in terms of what you can infer. Right. So 330 00:20:54,810 --> 00:20:58,330 my, my kid, my 331 00:20:58,330 --> 00:21:02,130 middle kids, my two younger kids are really into and really the three 332 00:21:02,130 --> 00:21:05,730 year old just likes hanging out with his big brother. They watch Dragon Ball Z, 333 00:21:05,730 --> 00:21:09,530 they watch the Jujutsu Kaizen, like all the crazy anime that's 334 00:21:09,530 --> 00:21:13,130 very popular now. I bet one of the things you could do, I, 335 00:21:13,210 --> 00:21:16,210 I've actually gotten into it. I was never much of an anime fan, but like, 336 00:21:16,210 --> 00:21:19,210 you'd say, like say Dragon's Ball Z. Right. Dragon Ball Z has been around 337 00:21:20,020 --> 00:21:23,460 that I'm aware of, maybe 20, 30 years. Right. But. So you can probably, 338 00:21:23,620 --> 00:21:26,580 you could probably make a solid assumption that there might be some Gen X folks 339 00:21:26,740 --> 00:21:30,180 that are Dragon Ball Z fans, probably a lot of millennials, a lot of Gen 340 00:21:30,180 --> 00:21:33,940 Z, Gen Alpha, whatever they're calling them now. But there's probably not a 341 00:21:33,940 --> 00:21:37,620 lot of people in retirement homes, boomers and 342 00:21:37,620 --> 00:21:41,300 up there are big fans of it. Is it because they wouldn't like it? 343 00:21:41,700 --> 00:21:45,300 I don't know. Maybe. But it's just, it tends that since that demographic 344 00:21:45,300 --> 00:21:49,030 skew is kind of small, you're probably not going to find 345 00:21:49,030 --> 00:21:52,430 a lot of them that are going to be into that in the retirement. I 346 00:21:52,430 --> 00:21:55,870 don't know that that's just me just firing an analogy. 347 00:21:56,110 --> 00:21:59,670 I mean, my parents liked K Pop Demon Hunter when my kids made them watch 348 00:21:59,670 --> 00:22:01,750 it, but I have girls, so I don't know, 349 00:22:01,750 --> 00:22:05,510 they're just really cute though. That's really 350 00:22:05,510 --> 00:22:09,270 cute. It's a very well done kind of cross genres, but yeah, yeah. 351 00:22:09,270 --> 00:22:12,720 And K pop is very, very, very 352 00:22:12,720 --> 00:22:16,480 addictive. Yeah. You know, so like it just 353 00:22:16,480 --> 00:22:18,400 sticks in your head. I don't know how they did it, but 354 00:22:20,640 --> 00:22:24,040 who, who are the industries? What are the industries that are really interested in this? 355 00:22:24,040 --> 00:22:27,600 You obvious, you mentioned Morningstar, obviously, I would imagine financial 356 00:22:27,600 --> 00:22:30,800 services. And 357 00:22:32,080 --> 00:22:35,800 Morningstar is asset management. Right. Is that what it is? Or a hedge 358 00:22:35,800 --> 00:22:39,520 fund or it's, I'm. Not exactly sure, data and research. So I mean, I think 359 00:22:39,520 --> 00:22:42,880 primarily they're known for their research and data and how they've rated 360 00:22:43,110 --> 00:22:46,870 funds over the years and they've expanded from there by way of acquisition. 361 00:22:46,950 --> 00:22:50,750 So PitchBook is a part of it. DVRS is an index business. So 362 00:22:50,750 --> 00:22:54,470 they, they have seven different pianos that really like traverse 363 00:22:54,550 --> 00:22:58,270 financial services. At this point I 364 00:22:58,270 --> 00:23:01,990 think financial services has been interested partially because I'm in financial 365 00:23:02,070 --> 00:23:05,670 services and I'm literate and being able to discuss it and showcase its 366 00:23:05,670 --> 00:23:09,350 benefits. Right, right. I would say this is more like 367 00:23:09,350 --> 00:23:13,170 functionally, like accurate for any 368 00:23:13,170 --> 00:23:16,690 place that needs human intelligence. Right. So I've worked with 369 00:23:17,890 --> 00:23:20,690 private equity teams that are helping to arm their 370 00:23:21,250 --> 00:23:24,970 portfolio companies with a marketing tool that doesn't 371 00:23:24,970 --> 00:23:28,650 have them, then looking to boutique agencies to do this level of market 372 00:23:28,650 --> 00:23:32,290 research and understand their ICP and find product market fit or message 373 00:23:32,290 --> 00:23:36,050 market fit. So there for them, it's very easy to kind of get in 374 00:23:36,050 --> 00:23:39,690 there, even the technical founders, and try to augment maybe a gap in 375 00:23:39,690 --> 00:23:43,460 their marketing acumen. I would say marketing 376 00:23:43,460 --> 00:23:47,100 agencies, creative performance, et cetera, they have taken 377 00:23:47,180 --> 00:23:50,780 to it really easily because they're already wizards who 378 00:23:50,780 --> 00:23:54,540 wield, you know, traditional wands on doing this kind 379 00:23:54,540 --> 00:23:58,100 of work to understand a market, to understand the message that's going to 380 00:23:58,100 --> 00:24:01,660 fit with that market and then to make sense of what the real results were 381 00:24:01,660 --> 00:24:05,460 when the market either engaged or didn't. Right. So and 382 00:24:05,460 --> 00:24:09,300 building the creative around that. So the ability to pre test all of that with 383 00:24:09,300 --> 00:24:13,040 the audience gets them to the starting line before they put money behind it 384 00:24:13,040 --> 00:24:16,800 or have their client put money behind it with the best possible set of 385 00:24:16,800 --> 00:24:20,560 options. So I think agency has been pretty prolific there too. And then 386 00:24:20,560 --> 00:24:24,120 the last. And again, I'm kind of biased because I came out of enterprise. 387 00:24:24,600 --> 00:24:28,359 Enterprise marketers who are finding gaps 388 00:24:28,360 --> 00:24:31,880 in the kind of the traditional products that are, have easy distribution 389 00:24:31,880 --> 00:24:35,560 within the enterprise are looking to a tool like 390 00:24:35,560 --> 00:24:38,040 Movera to try to get more 391 00:24:39,370 --> 00:24:43,210 what decision intelligence that's human based in what they're doing 392 00:24:44,490 --> 00:24:48,250 and so that's, that's where we're seeing a good amount of traction would be 393 00:24:48,250 --> 00:24:51,930 like in that mid market and enterprise level marketing team, 394 00:24:51,930 --> 00:24:54,930 whether that be product marketing or demand gen or market 395 00:24:54,930 --> 00:24:58,770 intelligence. And I came out of brand strategy so I found great 396 00:24:58,770 --> 00:25:02,490 utility for it there in corporate comms. So again I think 397 00:25:02,490 --> 00:25:06,330 it's really that go to market team where human intelligence becomes so 398 00:25:06,330 --> 00:25:10,160 important to decisions and current like traditional research methods 399 00:25:10,310 --> 00:25:13,670 are really slow and they're quite expensive and 400 00:25:13,910 --> 00:25:17,590 not everyone can do them, you know, or they think to just grab 401 00:25:17,590 --> 00:25:20,870 the information from within the four walls of the firm and 402 00:25:20,870 --> 00:25:24,630 anecdotes of talking to customers. Right. So this is a good 403 00:25:24,630 --> 00:25:28,470 way to augment an expensive way to augment some of that decision 404 00:25:28,550 --> 00:25:31,670 support. So you can like throw together like a, 405 00:25:32,390 --> 00:25:36,150 what's the, like a test market simulations and you can probably, 406 00:25:36,230 --> 00:25:39,740 there's probably knobs and dials you could do. So you can kind of like get 407 00:25:39,740 --> 00:25:43,500 multiple answers and I, I get it. So you can kind of, you can hit 408 00:25:43,500 --> 00:25:47,140 your, whatever your campaign is going to be with the running start as opposed 409 00:25:47,140 --> 00:25:50,900 to it's a little bit more guided than just throwing 410 00:25:50,900 --> 00:25:54,700 stuff at the. Wall and seeing what sticks. That's right. You know what to throw. 411 00:25:55,340 --> 00:25:58,740 You have better idea what to throw and where to throw it. That's right. And 412 00:25:58,740 --> 00:26:02,100 I mean we had, even when I was still at Morningstar, pre 413 00:26:02,100 --> 00:26:05,900 tested like the first time ever they built commercials. You know, they 414 00:26:05,900 --> 00:26:09,340 didn't, they don't really do brand level, you know, television commercial. 415 00:26:10,040 --> 00:26:13,760 They were deploying in Chicago, New York and London. And it was 416 00:26:13,760 --> 00:26:17,600 shown that in London it wasn't, whatever it was, the voiceover, the 417 00:26:17,600 --> 00:26:21,080 creative itself wasn't going to resonate with that audience as well. 418 00:26:21,320 --> 00:26:24,960 And so that gave us the foresight to take a look at what the voiceover 419 00:26:24,960 --> 00:26:28,640 is, what channels we might use, how much money we would put behind it before 420 00:26:28,640 --> 00:26:32,480 we deployed in that market. And so that, that kind of helped with channel 421 00:26:32,480 --> 00:26:35,920 strategy, it helped with content strategy. It certainly helped to 422 00:26:35,920 --> 00:26:39,230 evaluate that creative before any money 423 00:26:39,550 --> 00:26:43,070 changed hands. And so I think that was a super helpful thing. And now it's 424 00:26:43,070 --> 00:26:46,310 an award winning campaign. I'd love to feel like Movera had something to do with 425 00:26:46,310 --> 00:26:49,150 it along with all the brilliant minds that worked on it. 426 00:26:49,870 --> 00:26:53,629 That's cool. So you can get down to the macro, not macro, micro level of 427 00:26:53,629 --> 00:26:57,470 like the voiceover may not work in this market and things like that. That's 428 00:26:57,470 --> 00:27:01,150 cool. Yeah. In fact there's a. So we're in multiple modalities. 429 00:27:01,230 --> 00:27:04,990 We had used, I helped to co author 430 00:27:04,990 --> 00:27:08,750 the CEO's speeches for multiple years. And so we made him 431 00:27:08,750 --> 00:27:12,410 pract again and again and again, and we would. We would record 432 00:27:12,490 --> 00:27:16,290 them. And so the video analysis tool would look at the 433 00:27:16,290 --> 00:27:19,810 substance of what he was saying, the creative that was behind him on the 434 00:27:19,810 --> 00:27:23,610 deck, and then also his performance. So as it evaluated him, 435 00:27:23,610 --> 00:27:27,010 it said, you know, you're not taking time to pause for emotional 436 00:27:27,010 --> 00:27:30,850 resonance. And it gave all the timestamps across his speech where he 437 00:27:30,850 --> 00:27:33,850 should pause and why, and potentially even for how long. 438 00:27:35,680 --> 00:27:39,360 So it was looking at audience engagement and emotional connection. Then it started 439 00:27:39,360 --> 00:27:43,080 to take a look at, well, your message isn't that highly differentiated. And because we 440 00:27:43,080 --> 00:27:46,920 have this deep business context, we know that X, Y and 441 00:27:46,920 --> 00:27:50,680 Z are also talking about the convergence of public and private markets. This 442 00:27:50,680 --> 00:27:53,920 is what they're saying, here's what you should say so that it sounds uniquely 443 00:27:53,920 --> 00:27:57,680 Morningstar. So it now is helping to differentiate the message. 444 00:27:57,920 --> 00:28:01,520 And then when we got down to the creative, it's saying, you should do things 445 00:28:01,520 --> 00:28:05,250 that are a little bit more dynamic. You should back up what you're saying here 446 00:28:05,250 --> 00:28:08,810 with, you know, more data, graphs, charts, et 447 00:28:08,810 --> 00:28:12,370 cetera, less imagery. And so it was giving us guidance on three 448 00:28:12,370 --> 00:28:16,050 dimensions of that speech. And as we did it over time and recorded 449 00:28:16,050 --> 00:28:19,690 him, we saw his scores go up and up and up. And then 450 00:28:20,250 --> 00:28:23,450 it ended up being a really successful speech at 451 00:28:23,850 --> 00:28:27,610 the flagship conference that spring. So, you know, I 452 00:28:27,610 --> 00:28:31,090 had even said to him, like, maybe we should use this before earnings calls. You 453 00:28:31,090 --> 00:28:32,970 know, you never know. 454 00:28:34,900 --> 00:28:38,420 I could see the. I could see the applications and, you know, in fintech, 455 00:28:38,660 --> 00:28:42,100 I could also see applications of this in political campaigns. 456 00:28:42,500 --> 00:28:46,340 Yes. I was just thinking that. I'm like, you know, yeah, they 457 00:28:46,340 --> 00:28:50,179 would eat this up. Yeah. So we have been in 458 00:28:50,179 --> 00:28:54,020 some conversations, and I obviously can't talk about it with someone in the House 459 00:28:54,020 --> 00:28:57,740 of Representatives because we also have a news digest that 460 00:28:57,740 --> 00:29:01,380 will metabolize the news and give you the perspective of specific 461 00:29:01,380 --> 00:29:05,200 audiences. So he wanted to look at the two counties, you 462 00:29:05,200 --> 00:29:09,040 know, that he. That are part of his constituency. But then he was 463 00:29:09,040 --> 00:29:12,840 also looking at the committees, you know, so he's on two 464 00:29:12,840 --> 00:29:16,520 different committees and how are they responding to the news and what is it that 465 00:29:16,520 --> 00:29:20,239 they're doing? So it was doing this kind of social listening and moderate, you know, 466 00:29:20,239 --> 00:29:24,040 modeling of the audience. And then he could say, well, this is what my response 467 00:29:24,040 --> 00:29:27,720 would be to it and get them to vet it before he, you know, would 468 00:29:27,720 --> 00:29:31,380 push send on a communication. So, yeah, that was. That was 469 00:29:31,380 --> 00:29:34,380 something that. It's so timely Particularly with that news 470 00:29:34,860 --> 00:29:38,380 digest. Yeah, sure. And you know, 471 00:29:38,380 --> 00:29:42,180 particularly in it's, you know, the sentiment 472 00:29:42,180 --> 00:29:46,020 analysis angle on that's huge. And 473 00:29:46,020 --> 00:29:48,940 being able to do that in near real time, 474 00:29:49,820 --> 00:29:53,580 I think has, you know, applications across not just those two markets, 475 00:29:53,580 --> 00:29:57,410 but a bunch of different verticals as well. Because you 476 00:29:57,410 --> 00:30:00,210 almost. The perception is if you don't 477 00:30:00,770 --> 00:30:04,050 respond or react, that's a response or 478 00:30:04,050 --> 00:30:07,330 reaction, you know, so. 479 00:30:07,810 --> 00:30:11,490 Yeah, that's right. So I, I'd say between access 480 00:30:11,490 --> 00:30:15,290 to news content and then also connection with APIs. So 481 00:30:15,290 --> 00:30:18,970 we have Bloomberg flowing through the platform Pitchbook. We've got it 482 00:30:18,970 --> 00:30:22,730 for marketers, Ahrefs and Semrush data. If you're looking at SEO and you have 483 00:30:22,730 --> 00:30:26,500 thoughts towards what does it mean to show up in answer engines, all of 484 00:30:26,500 --> 00:30:30,300 this data flows and could be called through the platform so that you're 485 00:30:30,300 --> 00:30:34,140 looking at real data again, we leave a receipt of like this is where we 486 00:30:34,140 --> 00:30:37,940 drew this data from. You can see it. And here's where we 487 00:30:37,940 --> 00:30:41,540 inferred. So now you can use your own best thought 488 00:30:41,540 --> 00:30:45,220 and strategic thinking on. Okay, do I need to get that 489 00:30:45,220 --> 00:30:48,980 inference score down or do I feel good about this 490 00:30:48,980 --> 00:30:52,420 and I can build it into my argument in a really defensible way? 491 00:30:53,970 --> 00:30:57,650 So just curious. That's cool. Yeah, I'm, I'm down 492 00:30:57,650 --> 00:31:01,330 with it. I'm just curious how, 493 00:31:01,730 --> 00:31:05,530 in your experience, how have the, how's 494 00:31:05,530 --> 00:31:09,010 the opportunities presented themselves for someone to kind of step 495 00:31:09,010 --> 00:31:12,730 out and be creative is probably a nice way to 496 00:31:12,730 --> 00:31:16,410 say it. Or, and, or controversial. You know, 497 00:31:16,410 --> 00:31:19,930 there's, there's value in that some of the time. I mean, from a. If you're 498 00:31:20,160 --> 00:31:23,880 talking about marketing a product or service, you 499 00:31:23,880 --> 00:31:27,120 definitely want the differentiation. You mentioned that earlier. 500 00:31:28,320 --> 00:31:32,160 If you're talking about a campaign, whether it's a marketing 501 00:31:32,160 --> 00:31:35,839 campaign or a political issues type 502 00:31:35,839 --> 00:31:38,720 campaign, the opportunity to 503 00:31:40,320 --> 00:31:43,880 either be portrayed as a maverick or see what I did 504 00:31:43,880 --> 00:31:47,560 there or to, or to be, you 505 00:31:47,560 --> 00:31:51,200 know, just portrayed as somebody kind of breaking the mold, stepping outside 506 00:31:51,200 --> 00:31:54,880 the talking points. You know, 507 00:31:54,960 --> 00:31:58,240 how's, you know, how's your, how's your product and service 508 00:31:58,240 --> 00:32:02,079 addressing that. But also too, there might be some. I'm sorry, I 509 00:32:02,079 --> 00:32:04,760 didn't mean to cut you off. No, that's trying to cut off Andy. And then 510 00:32:04,760 --> 00:32:08,520 I cut you off by mistake. But also to the 511 00:32:08,520 --> 00:32:11,880 inverse of that. Like maybe there's some things you people, you don't want 512 00:32:11,880 --> 00:32:15,700 Mavericks, you don't. We want stability. Financial services kind of comes to mind. 513 00:32:15,930 --> 00:32:19,450 So sorry, I'll shut up. Yeah. So I mean, you can 514 00:32:19,450 --> 00:32:23,130 construct your own Brand identity that's going to say, you 515 00:32:23,130 --> 00:32:26,850 know, typically, here's our brand standards and here's our 516 00:32:26,850 --> 00:32:30,650 brand expression, which can come across creatively or tone or 517 00:32:30,650 --> 00:32:33,530 what have you. So that can be constructed and put on the back end so 518 00:32:33,530 --> 00:32:37,170 that everything is then scored against that and can tell you how far away from 519 00:32:37,170 --> 00:32:40,250 that you're drifting. Then you can put it in front of the audience. 520 00:32:40,810 --> 00:32:44,050 Typically, anyone who's working with is going to have their own framework for 521 00:32:44,050 --> 00:32:47,730 understanding. You know, how do I evaluate whether this message, message can go to market 522 00:32:47,730 --> 00:32:51,250 under my brand and how much risk am I willing to take? You can ask 523 00:32:51,250 --> 00:32:54,570 it even to assess the risk given the audience response. 524 00:32:54,890 --> 00:32:58,290 And as it splits that audience where people are having a difference of 525 00:32:58,290 --> 00:33:01,850 opinion, you can isolate that and say, is this my most 526 00:33:01,850 --> 00:33:05,450 likely buyer or is this the part of the audience that maybe there's a huge 527 00:33:05,450 --> 00:33:09,130 population that would like this more provocative 528 00:33:09,130 --> 00:33:12,810 message, but it's a, it's an audience, as it's described, that would churn. 529 00:33:13,380 --> 00:33:16,740 So, like, it allows you to make a little bit like, more strategic business 530 00:33:16,740 --> 00:33:20,260 decisions based on like, what. What are the attributes of that 531 00:33:20,260 --> 00:33:23,780 audience that are going to resonate with that more provocative message. 532 00:33:24,900 --> 00:33:28,460 The other thing I would say is just, oh, no, it's okay. This is built 533 00:33:28,460 --> 00:33:32,260 on a gan. So it's an adversarial network. And I 534 00:33:32,260 --> 00:33:36,060 would say, as opposed to being sycophantic, like so many models that 535 00:33:36,060 --> 00:33:38,260 are like, oh, yeah, I agree with you. And then you're like, no, don't agree 536 00:33:38,260 --> 00:33:41,930 with me. Be like adversarial. You know, push back. It's built 537 00:33:41,930 --> 00:33:45,570 to push back. In fact, we have a Persona specifically meant to 538 00:33:45,570 --> 00:33:49,410 poke holes and ask you questions and get you to question your assumptions. And 539 00:33:49,410 --> 00:33:53,170 I always start there. It's called Osprey. And I, like, that's my number 540 00:33:53,170 --> 00:33:56,810 one first stop on the bus is here's how I'm thinking about 541 00:33:56,810 --> 00:34:00,490 this competitive analysis. Let's like sort through what. 542 00:34:00,810 --> 00:34:04,330 What is wrong with that or how I can improve it. Same thing with a 543 00:34:04,330 --> 00:34:08,110 market sizing exercise. It feels like that should be wrote, but as you lend 544 00:34:08,110 --> 00:34:11,950 more specificity to it, I might be market sizing against not just 545 00:34:12,270 --> 00:34:15,750 a product, but a specific use case that I want to build up, campaign around. 546 00:34:15,750 --> 00:34:19,510 And now it becomes like a much more nuanced way of modeling 547 00:34:19,510 --> 00:34:22,710 an audience. So I always, again, start with that 548 00:34:22,710 --> 00:34:26,270 adversarial model to get me to think better, you know, like, really improve 549 00:34:26,430 --> 00:34:30,110 my strategic critical thinking. Kind of like the 550 00:34:31,310 --> 00:34:34,950 10th man in world War Zone. Okay, I don't know what that is, 551 00:34:34,950 --> 00:34:38,630 but should I watch it? I'm sorry, Andy. Andy, I cut you off. Yes, 552 00:34:38,870 --> 00:34:42,510 it's an interesting concept. I don't want to spoil it for you, but, like. And 553 00:34:42,510 --> 00:34:46,310 it's based on a real, real army unit where 554 00:34:46,470 --> 00:34:50,270 they basically become their contrarian. If nine people agree 555 00:34:50,270 --> 00:34:53,910 on something, then it's. They randomly will. 556 00:34:53,990 --> 00:34:57,470 If 9 out of 10 people agree on something or something like that, or 10 557 00:34:57,470 --> 00:35:00,960 out of 10, they will randomly pick one to. You have to 558 00:35:01,040 --> 00:35:03,040 poke holes in it. Oh, 559 00:35:05,440 --> 00:35:09,160 sorry. Encountered. That's okay. I first encountered that in World War 560 00:35:09,160 --> 00:35:12,960 Z. So. Yeah, that. That was where I saw 561 00:35:12,960 --> 00:35:16,720 that. The. It sounds what I 562 00:35:16,720 --> 00:35:20,160 was thinking as you were describing that. I guess the phrase that popped into my. 563 00:35:20,160 --> 00:35:23,520 My mind was, you know, there's no such thing as bad publicity. 564 00:35:24,410 --> 00:35:27,690 And if you are peaking interest, whether it's 565 00:35:28,090 --> 00:35:31,890 positive or negative interest, if you're provoking some sort 566 00:35:31,890 --> 00:35:35,690 of reaction in that, and I think a lot of the social media 567 00:35:35,690 --> 00:35:39,530 algorithms are tuned around being able to do that very thing, 568 00:35:40,330 --> 00:35:43,490 you know, to. To get a reaction, either an agreement or a 569 00:35:43,490 --> 00:35:46,170 disagreement, then that can lead to 570 00:35:46,570 --> 00:35:50,410 engagement. And if that's the goal, that makes perfect sense. 571 00:35:50,880 --> 00:35:54,400 That's right. I. In fact, I have a book right here called Filter World. 572 00:35:54,560 --> 00:35:58,240 I think that's what it's called. Yeah, Filter World. And it's really all about 573 00:35:58,320 --> 00:36:01,760 how algorithms can. Can do that, feed you back things that are more 574 00:36:01,760 --> 00:36:05,560 sensationalized, kind of like yellow journalism going back to Hunter S. Thompson. 575 00:36:05,560 --> 00:36:09,320 Right. That are meant to create some sort of response, whether good, bad, 576 00:36:09,320 --> 00:36:12,800 or ugly. So, yeah, I think that's right. But at least 577 00:36:12,960 --> 00:36:16,320 you could test. Yeah, at least you can test some assumptions first 578 00:36:18,650 --> 00:36:22,170 prior to taking it to market and getting slammed for it and 579 00:36:22,490 --> 00:36:26,130 having unintended consequence, potentially. Yeah, Well, 580 00:36:26,130 --> 00:36:29,930 I mean, if you think about it, I'm just basing this on my 581 00:36:29,930 --> 00:36:33,130 experience, because I have the most experience with my experience. 582 00:36:35,130 --> 00:36:38,490 I love a comeback. Right. I just. I love it. And 583 00:36:38,730 --> 00:36:42,410 often the way that that comeback begins, the. The arc 584 00:36:42,410 --> 00:36:45,380 starts with me first 585 00:36:46,340 --> 00:36:49,620 noticing something and having a negative reaction to it. 586 00:36:50,260 --> 00:36:53,900 And then as I get more information, I go, well, yeah, I could kind of 587 00:36:53,900 --> 00:36:57,580 see where they're coming from and, you know, begin to identify with it and 588 00:36:57,580 --> 00:37:01,340 empathize and. And then every now 589 00:37:01,340 --> 00:37:05,180 and then it's rare, but when it happens, it happens huge. And I 590 00:37:05,180 --> 00:37:08,820 think part of the reason is because I started so negative with it, my support 591 00:37:08,980 --> 00:37:12,370 skyrockets, you know, a little. It's not a line, it's an 592 00:37:12,370 --> 00:37:16,130 exponent, you know, very exponential curve of 593 00:37:16,530 --> 00:37:19,970 support that Grows out of that. And like I said, I think it's powered by 594 00:37:19,970 --> 00:37:23,170 stretching that rubber band in the opposite direction to start with. 595 00:37:23,650 --> 00:37:27,010 Yep. Although I would say some people are built that way because my 596 00:37:27,010 --> 00:37:30,610 dissertation looked at processes of empathy and processes of 597 00:37:30,610 --> 00:37:34,370 perspective taking and how counter. Counterargumentation happens. 598 00:37:34,450 --> 00:37:38,010 Right. What are the various factors, either in an environment or in a 599 00:37:38,010 --> 00:37:41,590 message that are going to create that? But there are also some things just in 600 00:37:41,590 --> 00:37:45,350 you that might have that approach to say. I would say 601 00:37:45,350 --> 00:37:49,070 my 7 year old, my little guy has like, he comes from a space of 602 00:37:49,070 --> 00:37:52,910 no. We always start with no. He's also like 603 00:37:52,910 --> 00:37:56,510 in the 99th percentile for math. I think he has like an engineering mind. Like, 604 00:37:56,510 --> 00:38:00,310 I just, I was gonna say. He sounds like an engineer before you even 605 00:38:00,310 --> 00:38:04,150 mention math. Yeah, yeah, yeah. Likes to take things apart 606 00:38:04,150 --> 00:38:07,840 and put it back together. So that's it. No is a good spot. Yeah. Yes. 607 00:38:10,480 --> 00:38:14,120 That's funny. It reminds me 608 00:38:14,120 --> 00:38:17,960 of. Here's a story from way back when. Once upon a 609 00:38:17,960 --> 00:38:21,440 time, I worked for a fintech startup. We'd call it. It wasn't called 610 00:38:21,440 --> 00:38:24,960 fintech then, but it was basically in early 611 00:38:24,960 --> 00:38:28,600 2000s. And it was a banking portal, but it was meant to be kind 612 00:38:28,600 --> 00:38:30,800 of banking for people who 613 00:38:32,320 --> 00:38:35,910 weren't comfortable with finance. Right. But the, 614 00:38:36,230 --> 00:38:39,350 the rationale was they wanted to make the site really friendly. And one of the 615 00:38:39,350 --> 00:38:42,950 things they did was they put little cute cartoon characters 616 00:38:44,070 --> 00:38:47,710 on every page, which people. 617 00:38:47,710 --> 00:38:51,390 And this was in Germany. So like it was a, you know, banking 618 00:38:51,390 --> 00:38:54,550 culture in the US is very conservative. Even 619 00:38:55,030 --> 00:38:58,830 Germany is even more so. And 620 00:38:58,830 --> 00:39:02,670 that's being kind of. Turns out 621 00:39:02,670 --> 00:39:06,030 people didn't want to put their money into a website. 622 00:39:06,350 --> 00:39:09,870 Which again, early 2000s. Right. Still, that was already a stretch 623 00:39:09,950 --> 00:39:13,670 with cute little cartoon characters. They wanted serious, they wanted stable, 624 00:39:13,670 --> 00:39:17,070 they wanted boring, they wanted, they wanted the suits, they wanted that. 625 00:39:17,310 --> 00:39:21,030 And it was kind of like when I saw the website, the design rolled 626 00:39:21,030 --> 00:39:24,870 out, I was like, I don't think this is gonna work. I better have my 627 00:39:24,870 --> 00:39:28,190 plane ticket home just in case. And 628 00:39:28,680 --> 00:39:32,480 you know, it turns out I was right. You know, trust me, 629 00:39:32,480 --> 00:39:35,760 I, you know, I didn't want to be right because I would have, you know, 630 00:39:35,760 --> 00:39:38,760 had dot com dreams and, you know, all that. But. 631 00:39:41,000 --> 00:39:44,800 But I mean, you're right. Like sometimes it would have been helpful 632 00:39:44,800 --> 00:39:48,520 if they were to test out, if they had the capacity to test out. 633 00:39:49,000 --> 00:39:52,520 Hey, what if we went for a cutesy K pop kind of demon hunter thing 634 00:39:52,840 --> 00:39:56,430 for a banking portal. It might fly today maybe, 635 00:39:56,510 --> 00:39:59,390 but probably not. 636 00:40:00,750 --> 00:40:04,590 Just depends on the audience. Again, yes, Exactly. Know your audience. Right. 637 00:40:05,950 --> 00:40:09,710 That seems like a tough sell. It, you know, in Germany in the late 638 00:40:09,710 --> 00:40:13,230 90s, early 2000s. I don't know. Right. It definitely was. 639 00:40:13,470 --> 00:40:17,270 I think after half a billion euros 640 00:40:17,270 --> 00:40:20,990 were spent, I think they acquired 120 new customers. 641 00:40:21,070 --> 00:40:24,790 So, yeah, it was br. It was bad 642 00:40:25,110 --> 00:40:28,950 right there. It was bad. And I might be rounding 643 00:40:28,950 --> 00:40:31,110 up ratio right there. I can do that. 644 00:40:35,190 --> 00:40:38,230 Yeah. So, I mean, again, I think 645 00:40:39,830 --> 00:40:43,670 audience, you can't really replace, like human response to something. You have to 646 00:40:43,670 --> 00:40:47,390 get something out into market and see if trust is established and people engage 647 00:40:47,390 --> 00:40:51,070 and ultimately make a decision to purchase. But I think getting 648 00:40:51,070 --> 00:40:54,420 to the starting line with the best set of options, with 649 00:40:54,660 --> 00:40:58,460 defensible reasons behind why he went with those options, 650 00:40:58,460 --> 00:41:02,180 is kind of a better spot than we were a year ago or two years 651 00:41:02,180 --> 00:41:05,060 ago. Right. And so I think, 652 00:41:06,020 --> 00:41:09,420 I mean, we can only go up from here, but I think, you know, I'm, 653 00:41:09,420 --> 00:41:13,180 I'm optimistic that if people were to start integrating this, it doesn't have 654 00:41:13,180 --> 00:41:16,700 to take them out of the job force. It just can help them do their 655 00:41:16,700 --> 00:41:19,820 job a lot better, you know. No, absolutely. 656 00:41:21,900 --> 00:41:22,460 Yeah. 657 00:41:26,460 --> 00:41:29,620 How did you get into this? How did you get into this? Because your background 658 00:41:29,620 --> 00:41:32,220 is in. Your PhD is in communications. 659 00:41:34,780 --> 00:41:38,620 You're getting used to dealing with engineers. Yes. 660 00:41:38,700 --> 00:41:42,380 How did you. How did you end up at a company that is largely driven 661 00:41:42,380 --> 00:41:46,220 by engineers? That seems. Yeah, this is a great question. 662 00:41:46,460 --> 00:41:50,140 So again, I was kind of that skeptic who was running a market research 663 00:41:50,140 --> 00:41:53,820 team and always being pressed on my budget. So the budget was, 664 00:41:54,060 --> 00:41:57,860 you know, in the high six figures. And it's like that's the 665 00:41:57,860 --> 00:42:01,620 first place everyone wants to cut when everyone's looking at margins. But 666 00:42:01,620 --> 00:42:04,900 it's also such an important place to make sure that product 667 00:42:04,900 --> 00:42:08,540 strategy, message strategy, all these things are kind of coming together in the right sort 668 00:42:08,540 --> 00:42:12,100 of way instead of wasting money downstream. And 669 00:42:12,100 --> 00:42:15,870 so I was trying to, you know, A, look for a way to 670 00:42:15,870 --> 00:42:19,630 cut cost, but B, I also really wanted to understand 671 00:42:19,630 --> 00:42:23,390 what was coming with this whole, like, generative AI thing, you 672 00:42:23,390 --> 00:42:26,590 know. So when I heard about let's scan LinkedIn, 673 00:42:26,670 --> 00:42:30,270 LinkedIn profiles and create synthetic Personas, I 674 00:42:30,270 --> 00:42:34,030 really started to pound the pavement to try to understand who's approaching this in 675 00:42:34,030 --> 00:42:37,670 the right sort of way aligned to how I think about modeling human 676 00:42:37,670 --> 00:42:41,450 populations, which is what I was studying. So when 677 00:42:41,450 --> 00:42:44,850 the strategist I was working with kind of heard me thinking out loud about it, 678 00:42:44,850 --> 00:42:48,530 he introduced me to the co founders at Marvera and, 679 00:42:49,010 --> 00:42:52,690 you know, I think I asked some hard questions. They could see that I was 680 00:42:52,850 --> 00:42:56,690 nerdy and skeptical and willing to try. And 681 00:42:56,690 --> 00:42:59,970 so they gave me access to it for almost a full year. 682 00:43:00,290 --> 00:43:03,890 I took it through the compliance process, which was helpful for them, and it was 683 00:43:03,970 --> 00:43:07,490 good to see how Morningstar was thinking about this progressively 684 00:43:07,890 --> 00:43:11,710 and then just hammered it and, you know, brought it into the C suite and 685 00:43:11,710 --> 00:43:15,510 brought it across the firm in my presentations. And I 686 00:43:15,510 --> 00:43:19,350 think through that, it really helped me to understand what the true value 687 00:43:19,350 --> 00:43:23,030 of it was. And after seven years at an enterprise, I, you 688 00:43:23,030 --> 00:43:26,790 know, I was definitely someone that liked to make decisions quickly, thoughtfully, 689 00:43:26,790 --> 00:43:30,470 but quickly. And I was kind of looking for, you know, maybe 690 00:43:30,470 --> 00:43:34,070 there's another opportunity to take my expertise and apply it in a different 691 00:43:34,070 --> 00:43:37,910 way. So I had a sabbatical. It was like 692 00:43:37,910 --> 00:43:41,630 a, you know, six weeks every four years. Thank you, Morningstar. 693 00:43:41,630 --> 00:43:45,370 And during that time, I just spent some time with them to really understand 694 00:43:45,450 --> 00:43:48,730 the technology, really understand the go to market motion 695 00:43:49,130 --> 00:43:52,810 and look at their capital raise and try to get involved in that 696 00:43:52,810 --> 00:43:56,450 process. And then six months later, they asked me to join 697 00:43:56,450 --> 00:43:59,530 them. Oh, that's cool. Yeah, that's cool. 698 00:44:01,050 --> 00:44:04,410 It was cool. I have to say, I'm drinking from the fire hose because 699 00:44:05,930 --> 00:44:09,050 working with the AI engineer, Full Stack 700 00:44:09,610 --> 00:44:13,220 developer and. And looking at operations and looking 701 00:44:13,220 --> 00:44:17,020 corporate taxes and all these things. No, that was not really. I carried 702 00:44:17,020 --> 00:44:19,500 my. You didn't wake up and you were like, I didn't want to do that. 703 00:44:19,500 --> 00:44:22,820 Like, that's interesting. 704 00:44:23,460 --> 00:44:26,300 The first thing that comes to mind, and I totally lost my train of thought. 705 00:44:26,300 --> 00:44:30,140 So if, Andy, this is an opening for you while I kind of reboot my 706 00:44:30,140 --> 00:44:32,820 blue brain blue screen. So give me a second. 707 00:44:34,900 --> 00:44:38,670 Oh, now I remember. You're welcome 708 00:44:39,550 --> 00:44:43,070 anytime, man. Having 709 00:44:43,390 --> 00:44:47,070 you mentioned regulations, this is what kind of. True. I was very 710 00:44:47,070 --> 00:44:50,750 skeptical of synthetic data, just in general, just 711 00:44:50,750 --> 00:44:54,590 because, you know, you're basically feeding machines into machines. And I'm old enough 712 00:44:54,590 --> 00:44:57,710 to remember when you took it like a tape cassette and you copied it and 713 00:44:57,710 --> 00:45:01,550 you did that enough generations, whether it was VCR or audio cassette, 714 00:45:02,030 --> 00:45:05,870 you had an issue. Right? You would get some kind of degradation. However, in 715 00:45:05,870 --> 00:45:09,450 reality, I've seen synthetic data do amazing things in the AI 716 00:45:09,450 --> 00:45:12,810 space, in the data space, more than it has any right to, 717 00:45:12,890 --> 00:45:16,450 basically. So that's why I was not skeptical when you mentioned synthetic 718 00:45:16,450 --> 00:45:20,250 crowds, because it's one of those things where it's worked better. 719 00:45:20,330 --> 00:45:23,610 But one of the upshots of synthetic data is that 720 00:45:24,170 --> 00:45:27,850 the reg, particularly around generating synthetic 721 00:45:29,050 --> 00:45:32,450 health data and things like that, you don't quite have the same 722 00:45:32,450 --> 00:45:35,900 regulatory constraints. Right? There is no PII 723 00:45:36,140 --> 00:45:39,820 to speak of. And you mentioned that there were regulatory hurdles for, for this. 724 00:45:39,820 --> 00:45:43,660 Like what, what were the regulatory hurdles in 725 00:45:43,660 --> 00:45:47,220 this case? I'm curious. Well, how, how could 726 00:45:47,220 --> 00:45:51,020 you use the outputs? Where would they be applied? If you're reconstructing 727 00:45:51,020 --> 00:45:54,820 the brand voice, what are you basing that off of? Is that, you 728 00:45:54,820 --> 00:45:58,500 know, is that considered for them proprietary information that would 729 00:45:58,500 --> 00:46:01,740 then feed the system for other, you know, competitors or 730 00:46:01,980 --> 00:46:05,610 just writ large? I think that was something that they were looking 731 00:46:05,610 --> 00:46:09,290 at. They were of course looking at data privacy. So 732 00:46:09,450 --> 00:46:13,290 you know, I was uploading not just our creative, 733 00:46:13,290 --> 00:46:17,130 but I was looking at our business strategy across the P Ls and trying to 734 00:46:17,130 --> 00:46:20,890 get it to incorporate when things are launching and where is their convergence. 735 00:46:20,890 --> 00:46:24,530 So that if I'm trying to create an umbrella level message at the brand level, 736 00:46:24,530 --> 00:46:28,370 it can render really strategically down to the different business units 737 00:46:28,370 --> 00:46:31,370 and create continuity and coherence in the message. 738 00:46:31,930 --> 00:46:34,980 So but that's, you know, that's their strategy. 739 00:46:35,620 --> 00:46:39,220 So they're really worried about like, you know, at what point 740 00:46:39,620 --> 00:46:43,340 can we feel like this is safe? And so, you know, in earnest, 741 00:46:43,340 --> 00:46:47,060 the team approached the ISO 742 00:46:47,940 --> 00:46:51,660 42001. They had a SoC2, the 743 00:46:51,660 --> 00:46:55,220 ISO37000 something because I'm never great at 744 00:46:55,220 --> 00:46:58,780 remembering all the numbers but like they really did like get after it in terms 745 00:46:58,780 --> 00:47:02,290 of, of ensuring that enterprises specifically would feel, 746 00:47:03,570 --> 00:47:07,250 you know, really like safe in this environment and 747 00:47:07,250 --> 00:47:10,930 everything. It was abundance of caution. What's that? It was, I'm sorry. 748 00:47:11,410 --> 00:47:15,249 Okay, that's fair. Yeah. Because one of the big selling 749 00:47:15,249 --> 00:47:18,490 points I've seen is it's not real. Right. So I'm sorry to cut you off 750 00:47:18,490 --> 00:47:22,330 again, but. Yeah, no, because it's not. But it's a synthetic data layer 751 00:47:22,330 --> 00:47:25,960 that sits on top of proprietary data and data gathered 752 00:47:26,270 --> 00:47:30,030 from like first party sources externally. So I think 753 00:47:30,030 --> 00:47:33,510 once you have the mix of multiple things, they just have to ensure that 754 00:47:33,510 --> 00:47:36,670 whatever's put in there is proprietary, is protected. 755 00:47:38,030 --> 00:47:40,910 That makes a lot of sense. Cool. Yeah. 756 00:47:41,630 --> 00:47:45,070 This is the world we live in, you know, that's cool. 757 00:47:45,390 --> 00:47:49,150 So any other questions Eddie or. I didn't mean to hog them up. 758 00:47:49,230 --> 00:47:52,440 I'm just fascinated by, 759 00:47:53,080 --> 00:47:56,600 is fascinated by the discussion. It's, it's one of those other. 760 00:47:58,120 --> 00:48:01,640 Well there's other discussions and topics where we see 761 00:48:02,520 --> 00:48:05,760 the kind of the real world interacting with the 762 00:48:05,760 --> 00:48:09,280 artificial and I don't say artificial in any kind of 763 00:48:09,280 --> 00:48:12,760 negative way, you know, in the sense of 764 00:48:12,760 --> 00:48:16,520 synthetic and, and to me it feels a lot more like art 765 00:48:16,520 --> 00:48:19,680 imitating life, you know, and 766 00:48:20,080 --> 00:48:23,200 as we, we find more and, and 767 00:48:23,760 --> 00:48:27,040 better ways to have technology enrich 768 00:48:27,600 --> 00:48:31,280 our ability to do our jobs well. I just. I find it fascinating. 769 00:48:31,440 --> 00:48:35,040 So. And it's. It's cool. I can tell 770 00:48:35,680 --> 00:48:38,800 that you found a real. A real fit 771 00:48:39,520 --> 00:48:42,800 for your education and your skills and it sounds like your 772 00:48:42,800 --> 00:48:46,080 personality and, you know, and kind of likes 773 00:48:47,220 --> 00:48:50,940 and that. That's always good, you know, when. When you can do what 774 00:48:50,940 --> 00:48:54,700 you are. Yeah, it's so true. I love nerding out 775 00:48:54,700 --> 00:48:58,300 every day on this stuff. Plus, like, I'm. I don't. I'm 776 00:48:58,300 --> 00:49:01,820 just. I can't naturally sell anything. I have no selling 777 00:49:01,820 --> 00:49:05,580 ability. But I can talk about it from the perspective of, like, 778 00:49:05,580 --> 00:49:09,300 a practitioner, you know, and a skeptic one at that. So 779 00:49:09,300 --> 00:49:13,110 that's really where I'm coming from in any conversation is like. Like, tell me 780 00:49:13,110 --> 00:49:16,870 why you don't buy it. Because I, like, I'm gonna get in your bandwagon 781 00:49:16,870 --> 00:49:20,270 and not buy it with you until we can figure out how it actually, like, 782 00:49:20,270 --> 00:49:23,230 works and fits into this process, you know? So. 783 00:49:25,070 --> 00:49:28,190 That'S a good way to look at it. That's. That. That's really not just selling 784 00:49:28,190 --> 00:49:31,870 with empathy, but selling with, like, sympathy, I guess. Right? Like, yeah, 785 00:49:31,870 --> 00:49:35,670 yeah, that's cool. That's cool. Where can folks find out more about 786 00:49:35,670 --> 00:49:39,510 you and Mavera? Yeah. I love talking to everyone, so please 787 00:49:39,510 --> 00:49:43,070 connect with me on LinkedIn. My name is Jill Axlide. 788 00:49:43,070 --> 00:49:46,890 Again, Mavera IO is where you can go and check it out. 789 00:49:48,010 --> 00:49:51,490 We liked people to kick the tires, so there's a free trial for everyone for 790 00:49:51,490 --> 00:49:55,170 14 days and you can connect with anyone on our team to 791 00:49:55,170 --> 00:49:58,970 walk through how to use it. Cool. Awesome. And we'll let our 792 00:49:58,970 --> 00:50:01,210 AI finish the show. Awesome.