[Emerging Tech] Compounding Institutional Knowledge for Growth With Ian Karnell

Ian Karnell is the Co-founder and CEO of VastAdvisor, an AI-powered growth platform that helps RIAs and wealth management firms build scalable, compliant client acquisition systems. He has over 25 years of experience at the intersection of FinTech, SaaS, and digital marketing, leading innovations that combine technology with financial services growth strategies. Before founding VastAdvisor, Ian built one of the largest independent digital agencies in the US and held senior roles driving strategic initiatives at firms like Envestnet. He is focused on replacing traditional lead generation with AI-driven systems that lower acquisition costs and accelerate organic growth for advisory firms.

Here’s a glimpse of what you’ll learn:

• [02:29] Ian Karnell discusses why most advisors don’t have a true growth strategy beyond referrals

• [05:28] Digital-first generations and shifting client acquisition dynamics

• [08:51] How are advisors underutilizing emerging digital channels like TikTok and Reddit?

• [11:59] Governance-first AI design to reduce hallucinations and model drift

• [14:44] Orchestrating multiple AI agents instead of relying on a single tool

• [20:11] Ian explains why learning speed beats traditional marketing in the AI era

• [27:23] Human-in-the-loop controls within AI-driven campaign execution

• [33:08] The strategy behind Vast Assembly as a venture platform for AI innovation
  

In this episode…

Many financial advisors rely heavily on referrals, but that strategy may not be enough in a rapidly changing market. With trillions of dollars shifting to digital-first generations, traditional growth tactics are becoming less predictable and harder to scale. How can advisory firms build a repeatable, data-driven growth engine that compounds over time instead of starting from scratch each year?

Ian Karnell, an entrepreneur and AI strategist specializing in wealth management growth systems, explains that firms must move beyond lead brokers and manual marketing workflows. He emphasizes defining and refining ideal client profiles using data, activating emerging digital channels, and embedding governance into AI systems to ensure compliance and reliability. Ian also highlights the importance of human-in-the-loop controls, measurable learning loops, and compounding institutional knowledge to create a sustainable competitive advantage. By focusing on learning speed rather than just marketing output, firms can reduce acquisition costs and improve long-term growth outcomes.

In this episode of The Customer Wins, Richard Walker interviews Ian Karnell, the Co-founder and CEO of VastAdvisor, about building AI-powered growth systems for financial advisors. Ian discusses governance-first AI design, orchestrating multiple agents instead of single tools, and turning marketing spend into compounding institutional knowledge.

Resources Mentioned in this episode

• Quik!

• Richard Walker on LinkedIn

• Ian Karnell on LinkedIn

• VastAdvisor.ai

• Edoardo Airoldi on LinkedIn

• Jeremi Karnell on LinkedIn

• "[Emerging Tech] Protecting Businesses From Privacy Lawsuits With Richart Ruddie" on The Customer Wins

• "[Emerging Tech] Eliminating Data Complexity for Growth With Mark Ovaska" on The Customer Wins

• "[Emerging Tech] Reimagining CRM With AI for Advisors With Thomas Clawson" on The Customer Wins

• "[AI Series] Transforming Client Services and Advisors’ Workflow Using AI With Mark Gilbert" on The Customer Wins

• Bill Cates on LinkedIn

• Referral Coach International

• Mike Langford on LinkedIn

• PodBox

• The Modern Financial Advisor Podcast

Quotable Moments:

• “I think that most advisors don't actually have a growth strategy…so when referrals slow down, the backup plan typically is they start buying leads.

• “And we built VastAdvisor as a growth operating system, not a marketing tool that turns client acquisition into governed, repeatable systems for the RIAs.”

• “So every dollar spent on our platform to launch a marketing campaign targeting an ICP becomes a learning event.”

• “I think in the next decade, you know, it won't be won by firms with the best marketing.”

• “We're trying to solve a knowledge problem at VastAdvisor. Like that's why I said, you know, we didn't build VastAdvisor as a marketing tool.”

  
  

Action Steps:

1. Define and refine your ideal client profiles using real data: Moving beyond surface-level demographics helps you understand motivations, pain points, and behaviors, allowing you to target prospects more precisely and improve conversion rates over time.

2. Diversify beyond referral-only growth strategies: Relying solely on referrals limits scalability and predictability, so expanding into digital channels creates a repeatable pipeline that is less vulnerable to market shifts.

3. Embed governance and compliance into AI workflows: Building oversight directly into your systems reduces risk and increases trust in AI-driven decisions, which is especially critical in regulated industries where errors can be costly.

4. Keep humans in the loop when deploying AI systems: AI can handle heavy cognitive lifting, but human judgment ensures strategic alignment and prevents flawed assumptions, improving both performance and accountability.

5. Treat every marketing dollar as a learning event: Measuring outcomes and feeding results back into your system creates compounding institutional knowledge that accelerates growth and builds a sustainable competitive advantage.

   
   

Sponsor for this episode...

This is brought to you by Quik!

At Quik!, we provide forms automation and management solutions for companies seeking to maximize their potential productivity.

Using our FormXtract API, you can submit your completed forms and get clean, context-rich data that is 99.9% accurate.

Our vision is to become the leading forms automation company by making paperwork the easiest part of every transaction.

Meanwhile, our mission is to help the top firms in the financial industry raise their bottom line by streamlining the customer experience with automated, convenient solutions.

Go to www.quickforms.com to learn more, or contact us with questions at support@quikforms.com.

Powered by Rise25 Podcast Production Company

Episode Transcript:

Intro: 00:02

Welcome to The Customer Wins podcast, where business leaders discuss their secrets and techniques for helping their customers succeed and, in turn, grow their business.

Richard Walker: 00:16

Hi, I'm Rich Walker, the host of The Customer Wins, where I talk to business leaders about how they help their customers win and how their focus on customer experience leads to growth. Today is a special episode of my series on new and emerging solutions called the Emerging Tech Series, and today's guest is Ian Karnell, a VastAdvisor. Some of our past guests in this series have included Richart Ruddie of Captain Compliance, Mark Ovaska of Precept, and Thomas Clawson of Slant CRM. And today's episode is brought to you by Quik!, the leader in enterprise forms processing. When your business relies upon processing forms, don't waste your team's valuable time manually reviewing the forms.

Instead, get Quik!. Using Quik!, you'll be able to generate completed forms and get back clean, context-rich data that reduces manual reviews to only one out of 1000 submissions. Visit QuickForms.com to get started. All right, before I introduce today's guest, I want to give a big thank you to Mike Langford, founder of PodBox and host of the Modern Financial Advisor Podcast. In fact, I was just on a show.

Go check out his website at PodBox.com and learn more. All right. I've been looking forward to talking to Ian. Ian Karnell is the founder and CEO of VastAdvisor, an AI-powered growth operating system built for Ria’s and enterprise wealth platforms that want to own, not rent their growth. Yeah, let's talk about that.

He focuses on replacing lead brokers, shallow attribution manual marketing workflows with governed learning based systems that compound over time. With a background spanning wealth management, fintech and growth strategy, Ian works at the intersection of AI compliance and go-to-market execution. All right, you know this, Ian, I always love talking to fellow entrepreneurs. So welcome to The Customer Wins.

Ian Karnell: 02:02

Thank you. Thank you. And again and a shout out a shout out to Mike Langford for the introduction. It's fantastic. Super awesome having me on.

Richard Walker: 02:11

Oh, my pleasure. So looking forward to this. So for my audience, if you haven't heard my podcast before, I just love to talk to business leaders about what they're doing to help their customers win, how they build and deliver a great customer experience. And the challenge is growing their own company. So, Ian, let's understand your business a little bit better.

How does your company help people?

Ian Karnell: 02:29

Awesome question. Thank you. And thank you for having me on. Let's start with a hard truth. First and foremost, I think that most advisors don't actually have a growth strategy.

They have referrals, and when they have vendors, right. And so when referrals slow down, the backup plan typically is they start buying leads from like the smart assets, the zoes they get agencies involved. And that's an expensive shared opaque, and it's impossible to compound. And so we exist to replace that entire model with something fundamentally different. And so when you think about the sector, every serious Aria has systematized investment management, almost none have systematized organic growth.

And we built VastAdvisor as a growth operating system, not a marketing tool that turns client acquisition into governed, repeatable systems for the Aria. So that's what we're doing.

Richard Walker: 03:34

That's complex. Let's unwrap that a little bit more. You know, I just had Bill Cates, the the growth referral coach, on on the show with his books, talking about the referral engines and it being such a strong source of where you get growth from. So when you talk about a growth operating system, how does that fit into the advisors world exactly? What are you taking from them and leveraging and making better.

Ian Karnell: 03:56

Sure. Well, first and foremost, I want to I want to address the whole referral thing. Yes, referrals have driven and have been, well, have been a predominant channel for driving organic growth for all, you know, nearly every advisor for over 100, I mean, since, since this sector. You know, was born. So and obviously referrals are a big part of every market sector.

That being said, we have about $87 trillion beginning to, you know, change hands from baby boomers to digital first generations that are nothing like their parents, that are nothing like their grandparents. So assuming past, you know, kind of antiquated channels will continue to perform like they have, I think is a fundamental flaw in anybody's supposition about how growth is going to, you know, kind of move forward with these new digital first generations. They act nothing like their parents, and nor do they value referrals the same way their parents did. And so that's a fundamental shift. Yep.

Richard Walker: 04:54

Can I validate that for a second? Because when I became an advisor, I joined my mentor in his business. Okay. All of his clients were his generation, right? They had kids, maybe kids already in college here.

I was just out of college. I didn't have kids. I didn't relate to all the things they're going through. And he thought, well, Rich, you're going to take their kids as clients. I didn't like their kids as clients.

So it didn't work. Right.

Ian Karnell: 05:25

That's right, that's right.

Richard Walker: 05:26

Anyway. Keep going.

Ian Karnell: 05:28

Yeah. No, no. So, I think fundamentally accessing growth, organic growth or participating in that $87 trillion wealth transfer needs, you know, requires an advisor to think differently about tactics, tactics that they've used in the past. I mean, wealth is infamous for just being a laggard with any sort of digital transformation investment whatsoever. Kitces talks about this a lot, and he writes a lot of reports about it.

And I think when you look at the next generation of advisor, they they're they're thinking differently than their predecessors, which is refreshing. And it's great. They don't think about this business as a lifestyle business. They're looking at it completely in a different way, and they're thinking digitally about their practice and thinking about digital channels. And they're looking at, how do I embrace this next generation of wealth, and where do I find them?

And so that's fun. So that's part one: we help advisors activate audience discovery. Like what are their ideal client profiles using large data sets using wealth signals. And so and you know, and then we help them activate those audiences across all of the different digital channels that exist today. There are dozens of them.

You know, we all know the core ones, the LinkedIn, the Metas, the Googles of the world. But there are some exciting new ones, like Reddit, and there are a vast number of self-directed, high-net-worth individuals that are in those Reddit channels, in Fortran channels. And these channels are releasing or these platforms are releasing targeting and media capabilities for the first time. I think that's going to that's going to be, you know, so there are new emerging digital channels all the time. And I think they're going to activate new opportunities for Rias to access audiences that have been otherwise difficult to reach through digital channels.

And so that's one of the things that we do to help them.

Richard Walker: 07:32

Do, you know, do you know where I see Dave Ramsey the most?

Ian Karnell: 07:35

Where?

Richard Walker: 07:36

TikTok.

Ian Karnell: 07:37

TikTok. Yeah. Well.

Richard Walker: 07:38

It's the only place I see Dave Ramsey. Yeah.

Ian Karnell: 07:41

Yeah. I mean, so TikTok, it's interesting like my own and some of my own investments that I've made because I'm one of those self-directed individuals. I've, I have I follow a number of analysts like these are serious analysts, but I follow them on TikTok, and I and I, and I, and I, they're constantly in my feed. And I validate these insights that they provide through a multitude of other ways as well, through a multitude of different LMS that I utilize to validate facts. But that's where I am.

That's how I generate my own investment thesis and where I've made my own investments. And I've done quite well utilizing those two methods. And I'm so I'm not, you know, so I think I'm, I'm, I'm in good company with other self-directed investors utilizing these channels. And I think it's just untapped by advisors. Yeah.

So we want to one of the things of many things that we're doing, we want to help them target those individuals.

Richard Walker: 08:40

So okay, so when you talk about ideal client profile, are you taking what is their current profile and figuring that out for them, or are you helping them assess that and realize, oh, I have a better profile I could go after?

Ian Karnell: 08:51

Yeah. So great question. So what? So initially when a client onboards with us, one of the very first experiences that they have is they register with our site and they're asked to enter the URL of their Ria into our website. So once they hit submit, it activates a number of agents and subagents and workflows that have been built to power that advisor.

Those agents crawl the advisors websites. They look at their LinkedIn profile. It does a Google search on the advisor as well as it looks at the form ADV, and I like to view it as it is the foundational institutional knowledge that we're capturing, which is all of the publicly available information about that Ria, you know, through large language models and through internet search, and we're ingesting it into our database, and we're utilizing it in a number of different ways. We utilize it to pre-configure our application. So there are compliance settings get set automatically there.

Firm information gets set automatically there, firm fee rate gets set automatically. But another thing that happens is that we preload in three what we infer as three ideal client profiles. So we have a number of agents just dedicated to ingesting this data, looking at various wealth signals, enriching with various other third party research about that client segment, be it Pew Research or Boston Consulting Group. And then we look at the firm's geolocation. We look at the firms, how the firm positions its offering.

There are some firms that are very specific about which niches it wants to target. We evaluate how competitive those niches are. When I say we these agents, do it in a matter of minutes and it presents to the firm. Three. Ideal client profiles now with robust psychographic, demographic and geographic insights about that particular profile.

So it's not just kind of broad segmentation. We get into what are the key pain points with respect to this ICP. What are their key values and motivators. We surface all of this data by the way. We also source where we get this data from.

And we also allow the firm to click add ICP, ICP meaning ideal client profile. And and we give the firm the ability to describe their own ideal client profile or and or we give them the ability to click into the three that we load in and refine it. So we have some firms that may want to refine the AUM. We have some firms that may want to refine the or the geographic profile of that audience. Maybe they only want to target Massachusetts and not New England, you know, that sort of thing.

And that changes the data. It changes the profile significantly sometimes. So that's that's the first thing that we help them do is really understand. Who do you who are you targeting and what represents the ideal client profile for your firm? It goes way beyond AUM and household income, way beyond it.

Richard Walker: 11:54

So does your system ever get it so wrong? It's obvious. Like like laughingly wrong?

Ian Karnell: 11:59

Yeah. So no, no it hasn't now. And there's a reason for that. By the way. We've taken steps.

Governance is a is a core component of our value proposition. Go to our website. You'll see governance everywhere. And there are steps that we've taken to to build observability into every single AI generated insight or outcome and measure that and audit that. And those are all immutable, by the way.

So that's really, really important, not just from a compliance perspective. It's important from a AI perspective, we didn't want to just build a wrapper around a publicly available LLM. We're building learning systems. And so it's really, really important that we mitigate hallucinations, we mitigate drift, we mitigate latency and costs related to all of that. So we've taken the steps to really make the smart investments, to do all of those things so that the results are reliable.

Not to say that they're perfect to the extent that they're not perfect. We have systems to be able to catch that and learn from that before it becomes something that is an issue, like something so grossly, so grossly amiss on behalf of the Ria. Now the Ria might say, well, I think about my segmentation differently, and we give them the ability to modify it and then retrain it or just build build something from scratch. But we haven't seen major misses at all.

Richard Walker: 13:37

So okay, I got to bring this out to the open because a lot of people in this industry are super focused on I need an agent to do my job or this job or that job, right? The workforce is real. People want it. And that's why we see all sorts of hype and super fast interest. And like I've had Parker from jump, I've had Mark from Zocks on here, etc. those guys have taken off faster than I've ever seen in this industry.

You are actually talking about something that a lot of people I don't think, see or understand, and that is the orchestration of a lot of agents together to do different things. Now, I'm personally doing that from a software development standpoint. And my own example is if if I have an agent write code, how do I know it wrote the code I needed? Right, right. You can test it and say, oh, did it work?

But how do I know it really followed the specification? You can introduce other agents who come in and verify check. Right. You can use different language models.

Ian Karnell: 14:35

Absolutely.

Richard Walker: 14:36

Yeah we do.

Ian Karnell: 14:37

That. Yes.

Richard Walker: 14:38

So yeah talk about that. How are you doing that? How did you even come to this conclusion that this is needed? Because I don't think the majority are there yet.

Ian Karnell: 14:44

So great question. By the way, if you look at our team, first and foremost, I think it's grounded in the team that we've pulled together. We are a team of, you know, very experienced entrepreneurs in the Wealthtech space. We are a team of very experienced engineers in the Wealthtech space, and we are a team of data scientists that are data scientists who are heavyweights in academia, Harvard, MIT. These are individuals who, when I say heavyweights, like some of the most published academics in the field of causal analysis, in the field of causality and big data sets and machine learning, that's our team.

So we're fusing some of, you know, I would argue, the best and brightest in, you know, from a startup perspective in Wealthtech with the best and brightest in academia to solve for a pretty complex problem. Now when you hear the word agentic. I've gotten to a point where I kind of like pause when somebody just throws it out there because like, do you really understand what it means? AI that has agency, AI that you can trust to have that agency. And so I don't my, my thesis is this.

And by the way, we have an agentic framework within our application. And the framework is based on this thesis that I'm, you know, which is I don't think there's a company, I don't think there's an individual, well, at least in wealth, just because of the legal and regulatory bars that are set there that is ready to hand the keys over to an agentic agent and just let it run autonomously and can trust with that, that it should have that agency. Now, will there be a day that that happens? Absolutely. I have no question about that.

How soon we get there, I think is probably sooner rather than later. How quickly we're, you know, we're seeing, you know, we're seeing innovation in this space. But like, our genetic framework enables the user to be able to toggle on or off specific agents or to set sensitivity threshold for an agent or activation triggers for an agent. And then it gives you that ability to through our advisor IQ, our agent IQ dashboard, excuse me, is that we surface these signals for each agent. How well are they performing their job?

How how good or how well have they predicted an outcome. And we measure all of that through VastAdvisor IQ. This is the difference, I think, between I think platforms that are attacking AI in a different way, not just using AI to do what most people have been doing with AI, which is can I make something more efficient? Can I can I save time in doing whatever x, y, z job I'm doing? And that's where we had a category.

We see a bulk of AI innovation solving that problem. It's an efficiency problem that it's solving. We're trying to solve a knowledge problem at VastAdvisor. Like that's why I said, you know, we didn't build it advisor as a marketing tool. We built it as a growth operating system.

So what we want, what we built is a system that is ingesting just an infinite number of data signals that we're putting into learning loops, so that we're training these agents to perform in a way where we're mitigating hallucinations, drift latency, we're improving predictability. All of these things have an impact on outcome. We're talking about institutional knowledge. So every dollar spent on our platform to launch a marketing campaign targeting an ICP becomes a learning event. And that's very, very different than, you know, working with an agency or launching a campaign on meta.

You know.

Richard Walker: 18:28

You know, I think as AI entered our Our World a few years ago, it was more like strapping rocket boosters onto your vehicle, boat, car, roller skates, whatever. You're just like, I can go faster, I can go faster. What you're talking about is a whole new vehicle. Yep. Designed from the ground up.

And look, I want to actually give a framework kind of thesis of my own around this because what I realized is there's two kind of perspectives people have towards AI itself. And before I even say this, I want to say what my litmus test is. Can you ask AI to do the same thing twice and get identical results? If you can't, you don't have control over the AI.

Richard Walker: 19:06

That's a problem. That's a problem. Yeah, especially.

Ian Karnell: 19:08

In a regulated industry like wealth, it becomes a problem, right?

Richard Walker: 19:12

Yeah. So so here's my mindset how it's evolved. I think there's two camps on this. And I still feel like I'm in the minority. But you're probably there.

Tell me what you think. One is you treat AI like a toolset. And anytime you have a tool, it has a purpose. It has a job, it has this known thing you're going to make it do. Hence the rocket being strapped to your roller skates.

It just makes you go faster. I think of it as a human.

Ian Karnell: 19:34

Yes.

Richard Walker: 19:35

Okay. But here's why. A group.

Ian Karnell: 19:37

Of humans.

Richard Walker: 19:38

A group of humans. Yes, yes. Because everything you're describing and the checks and balances and the learnings and the, you know, analyzing, did they do it right? And the sensitivity, dialing up and down. Those feel like human decisions and they interact like human decisions.

And in fact, if you think about how AI operates with you when you're talking to it, it feels the most human of any tech we've ever seen in our lives.

Ian Karnell: 20:01

Yep.

Richard Walker: 20:02

And it starts to mirror you and reflect you. Oh, by the way, Ian, you're absolutely right. You're brilliant idea. Yes. You're so right to ask me that, right?

Ian Karnell: 20:11

I don't want the toxic affirmations. I want. I want you to. So. So I mean, you hit, you hit the nail on the head.

And I think what humans exhibit more than anything is the the propensity to learn, the propensity to, to, to over time build knowledge and that knowledge affect outcome or affect our behavior. Right? And so from childhood to adulthood, we see that that's probably one of the most defining elements of being human. And as we grow in scale, is just how our knowledge grows in scales and how that changes our capabilities and our ability to have empathy and knowledge and and to be able to have a job and all of these things. And so most and I think that would that's probably one of the most important attributes that we sought to build within our platform is the ability to be able to capture knowledge and learn from that knowledge and improve the outcome over time.

So when you think about all of the knowledge within an Ria, all of it, in all of the different component parts, from the CEO to the advisors to the marketing teams, to, you know, to, to to the front desk person. Right. And all of the different kind of systems and workflows that they utilize and how Disaggregated, that is. And where is there a central source of truth within any area? I would argue no.

And and I would argue that what we're doing is to or is to present, at least in the front office, a unified way to do that, a unified way to be able to capture and compound institutional knowledge for the area. Because I think in the next decade, you know, it won't be won by firms with the best marketing. It's going to be won by firms that have the fastest learning and and the ability to compound the knowledge that they have. I think, you know, in a in a world where everyone has access to AI, everyone has access to AI.

Richard Walker: 22:07

Right?

Ian Karnell: 22:08

Those that harness AI to compound learning and institutional knowledge will win. It's the new competitive mode, I would argue.

Richard Walker: 22:17

So that's a really good point. And over the time since I've been interviewing people on my show, I've always talked about AI. I've learned, you know, like you said, we all have access to the same ChatGPT, Gemini, etc.. What makes us different then? And my answer has been the data.

If you look at what data you uniquely have access to versus me, right? The LMS not being trained on the department, Department of Defense classified materials, they're not being trained on Kellogg's cereal eaters, etc.. They're they're being trained only to the extent you have the data and you train them yourself. So I looked at it and said, what data does Quik! have? Right?

We've built over 100,000 forms. We have all that data of building forms. Nobody in the world has that data. So what can I do with it? That's right.

And now you've introduced kind of an elevated concept of it, which is it's not just data, it's compound learning. That's right. I think that's really elegant, Ian. I mean, I'm going to take this as a whole new another way of looking at this because yeah, look, in my own company, we're in a process right now using AI to document all of our code repositories. Why smart?

Ian Karnell: 23:18

Smart.

Richard Walker: 23:19

Why why would we do that? Right. We've never had to do that before. So why are we doing it now? It's not like it's going to go to end consumers.

I don't want to see my source code. No. If I onboard another person to my team, they need to be able to figure out what our code does faster and easier.

Ian Karnell: 23:32

And that person might be an agent that's writing code.

Richard Walker: 23:35

Might be an agent. It's going to be an agent. Come on.

Ian Karnell: 23:38

It's likely going to be an agent. You're like I mean like like versus like hiring. I mean, hiring more human, you know, human.

Richard Walker: 23:46

Oh, no, I'm hiring more humans, too. Don't get me wrong. I love the humans, but I.

Ian Karnell: 23:50

Do.

Richard Walker: 23:51

I know it does. It does depend on the role. This is true. I love what Jeff Woods, who's an author of the AI Driven Leader, he was on my show as well. Sorry, I've dropped a bunch of people from my show, but it's all been relevant.

Jeff has said a job is a set of skills applied to tasks and work, right? So the job can change because the tasks can change, the skills can change. So don't be threatened by AI. In other words. Right.

And now I'm looking at software development and things like that. And it's it's transforming and changing. I still need high quality people to do that work. But now let me take it back to VastAdvisor. What you've described to me is you have invented a very robust process that takes an entire team of people who are actually AI people, if you will.

That wasn't possible before because only AI can do this. But but nobody's buying your product. Getting a single agent that has this one LLM input. They're getting this choreographed kind of ensemble of, of skill sets and behaviors that you had to create. I'm sorry to promote that.

Ian Karnell: 24:57

But a fine tuned model at the enterprise level and at the Aria level, that's really important. So it's the orchestration of the agents through the workflows, but it creates what we call the advisor intelligence loop. So it loops back into a system and retrains these models and improves upon these models. And it gives us the ability to be able to enable a new model release and then track lifts in, you know, and conversion rates or a drop in your cost per click or your cost per acquisition, or your ability to be able to more precisely target an audience, we'll be able to track that. We'll be able to expose how well your model is performing against the cohort median.

Right. So like and it's all about how your model and the signal intelligence we capture from your model helps us to improve that model. So it is about learning speed. It's about compounding institutional knowledge and intelligence. That's why I don't I don't I don't talk about advisor as a marketing platform.

We're far from yeah.

Richard Walker: 26:08

Yeah, you are far from it. And I really think that when we talk about emerging tech with AI, the true value is what you've described. You have built a system that involves so many moving parts and pieces, as well as intelligence that you just can't replicate easily. You know, you have a big competitive moat from your own product standpoint. And second, this wasn't just strap on AI and let's go.

This wasn't just a quick fix and let's go faster. In fact, I'm finding out something else just over this past weekend, to be honest, that with what I've built in terms of software development, that's autonomous software development. In order to perfect it and make it better, it has to actually go slower as a result.

Ian Karnell: 26:47

Yep.

Richard Walker: 26:48

Now the outcome is faster because I'm not reiterating and revising and fixing because I solved a lot of quality problems. But I went slower up front, which is amazing to think, oh, AI is going to go slower. Yes, with better quality results.

Ian Karnell: 27:02

Yeah. Less lines of code, better outcomes from the code written. Absolutely, absolutely.

Richard Walker: 27:07

Yeah.

Ian Karnell: 27:08

Yeah.

Richard Walker: 27:09

Okay, so one other question about your product is AI a is there an entry point in your product where people are actually talking to AI and sharing their data with AI, or is it all behind the scenes to give them this intelligence and help them make decisions? Or is it?

Ian Karnell: 27:23

Great question. So we've designed with Invest Advisor AI to do a lot of the heavy cognitive lifting, right? Throughout all of the different kind of component parts of or the features within our platform. I haven't really talked about specific features yet, but but we've designed AI to do the cognitive the heavy cognitive lift. That being said, it's critically important that there are human in the loop elements from the start to the end, and every aspect of every use case that our platform solves for.

And so, you know, when we, for example, when we preload in ICP, we are, you know, we're putting the advisor in a position to be able to modify or add to those ICP when we when an advisor and then we provide intelligence overlays like really insightful intelligence with respect to audience sizing, which is the best platform to reach an audience. What type of cardiac efficiency can they achieve by targeting an audience in a specific way? And when they when they begin the process of actually building a campaign again, you're going to have AI on our platform doing a lot of the cognitive lift. It's going to say, okay, great. Which ICP do you want to start with?

And the human is a critical element to all of this. So we're asking the human to I mean, we could have AI do all of this. We could say nope, because it's all data driven. We we have with every single ICP, we have a fit score assigned to that ICP. And that fit score is how well, based on the profile of the firm and your service offering and your fee rate and the and your geographic profile and the specific ICP associated with that, we have a fit, you know, a fit score that says this ICP fits better than this ICP.

And so we could have AI just look at the fit scores. Automatically pick the ICP and then automatically begin the campaign development process, which is, you know, great. We have an ICP. What's the what's the value proposition we want to articulate to that ICP? What's the services offering.

We want to present that ICP. What's the emerging marketing trends that we see. You know, either as headwinds or tailwinds in targeting that ICP. And then what are the campaign themes and then what's the what's the channel mix and messaging strategy, and then what's the daily budget, the optimal daily budget that drives the optimal, you know, kak multiple hour AI does I mean, so so in each one of those stages, it's a five step process. You have AI generating all of the outcomes.

But we ask the human to select. All right great. Which campaign theme do you want. And we give a resonant score for each campaign theme. Great.

So which what's the channel mix and a messaging strategy. Our AI will pre-select the most optimal channels, but you can still add in any additional channels that you want or take away from a channel. And then when we present the optimal daily spend to target that, and we give you the optimal cash multiple, you can dial that. You know, you can dial that budget back or up based on what your constraints are. And then, you know, and then we get into our platform, building out automatically all of the campaign assets, everything from the copy, the creative.

It does a complete compliance review. That's part of the workflow, not an not an afterthought or something that you do as a last step. It's embedded within the workflow. And so and so the human in the loop element is such a critical component to the experience somebody has in the app.

Richard Walker: 31:05

I love that you've concluded you've realized that humans are the ultimate gatekeepers. Yes, you have to. Yeah. Because I've seen this. I did this one enhancement to my framework for building software, where I brought in what's called Red team, blue Team, and the red team is a different team of people with their own personas of of behaviors, etc., where they critique the work that we're doing and they give feedback.

Well, my first try at this, they went off and had a conversation without me. They just it was funny. It was watching a transcript written and they made a bad assumption. Ian. And they went off so deep on that assumption.

And when they finally shut up, I got to talk. I said, guys, you've made a terrible assumption that would never happen. Oh, then everything we've done is wrong. I'm like, okay, let's let's retool this. You need to keep me in the loop here as the human.

Yeah.

Ian Karnell: 31:57

Yep. Check in and validate before you make a core assumption and move on. Absolutely. Yeah.

Richard Walker: 32:03

Yeah. No, that's really, really awesome that you're putting all this together. Dude, we should keep talking. We're getting to the end, and I have to wrap up at some point. I, I know I'm going to.

You and I are going to go to barbecue with Mike and hang out and talk more about all this stuff. Maybe we'll come back for another show and talk about it further, because this is just fascinating. So let me switch gears before I get to my last question. What's the best way for people to find and connect with you, Ian.

Ian Karnell: 32:27

Oh thank you. Well, LinkedIn, obviously you can do a search on LinkedIn Ian Karnell or you can search VastAdvisor or VastAssembly. You'll find you'll find us that way you can connect with me directly on LinkedIn. Love those connections. You can email me directly at ian@vastassembly.ai, there's a reason there's two different names.

And and those are probably the two best ways. Or you can go to the VastAdvisor website and you can click. If you click on scheduling a demo, it goes directly into my calendar. So that's another way to schedule with me directly. LinkedIn is another way.

My calendar link is right in my profile.

Richard Walker: 33:04

So okay, 30s VastAssembly. Tell me about it.

Ian Karnell: 33:08

Yeah, VastAssembly is a holding company. So the idea here is that, I anticipated that we were going to do some really special work in solving for a very specific use case in wealth, but that it would translate to other adjacent at markets within financial services, banking, insurance, so on and so forth. And so when we started to develop the code, we wanted the code to be written and owned by VastAssembly. It's essentially a venture platform where we can bring in, if we wanted to, outside venture capital to accelerate the growth of that platform and other adjacent markets, outside, inside of financial services, outside of financial services. And we license the code to the operating company, the advisor.

So when we when we end up exiting from the advisor as an opco, the buyer gets to keep that code set. They can fork it, they can do whatever they want with it. But the core code that we've developed to be these AI agents, all of the IP, the data we can then take and apply to the next adjacent market sector or into a completely different market sector, if we wanted to. That's why.

Richard Walker: 34:20

So I told you, entrepreneur.

Ian Karnell: 34:23

AI is the is essentially the venture website that we have for the two companies. Yeah.

Richard Walker: 34:29

Nice. Very nice. Okay. Awesome. Here's my last question.

Who has had the biggest impact on your leadership style and how you approach your role today?

Ian Karnell: 34:39

Well, I mean, I think there it takes a village, right? I think there are a number of people that do that that have had the biggest influence. I mean, my twin brother, Jeremi Karnell, who's still at Envestnet, who acquired our last company as head of data solutions over there, has always had probably the biggest impact in my life in helping to shape my own leadership style. He was the CEO of Trulia, which was the company that we sold to Envestnet and and to be part of that company to to watch how he navigated some really tricky times during the startup, and the scaling of true Linux to the to the time that we successfully exited exited from that company with Envestnet really taught me some amazing lessons in leadership, and so I credit my twin brother quite a bit for that. But there are other people like my business partners Phil and one of my newest, you know, the newest additions to the team are our Chief Data Officer, Dr. Eduardo Airoldi. When you think about the sophistication and the way that we're approaching AI and, and fine tuned models within our platform, and how we are articulating a very different value proposition than most AI solutions out there in Wealthtech or outside of Wealthtech. I credit him. He was a former business partner of mine as well. I met him when he was the head of well, he was a Harvard professor at the time when I met him. And so.

Yeah. So he's had a huge influence. And then, you know, people like Danny Fava and Jason Pyrah, both advisers to and investors, both advisers and investors and vast Danny is the chief strategy officer at Carson. Jason runs Woodgate an aria out of in Canada out of Toronto, a very successful one, and he's a big influencer in this space as well. And so both of those individuals have had an amazing influence over how we've sharpened our value proposition, validating a lot of the ideas that I bring to them, saying, hey, look, this is the problem we're thinking about solving.

Is this right? And then really helping to refine that, to help me mold that so that I don't go down rabbit holes that I learn these lessons the hard way later. So I credit them as well.

Richard Walker: 36:54

Oh, man, these people along the way are so, so important to how we develop critically important. I mean, look, one entrepreneur to another, we're like unbridled horses quite often. And if somebody is not helping us stay straight.

Ian Karnell: 37:07

The scar tissue I have all over my body from not having that in the past, has taught me to make sure that you have that as an entrepreneur. Absolutely.

Richard Walker: 37:15

Oh for sure. All right. I hate to wrap this up. I have to do it. I want to give a huge thank you to Ian Karnell, founder and CEO of VastAdvisor, for being on this episode of The Customer Wins.

Go check out Ian's website at vastadvisor.ai, and don't forget to check out Quik! at quickforms.com where we make processing forms easy. I hope you enjoyed this discussion. We'll click the like button, share this with someone, and subscribe to our channels for future episodes of The Customer Wins. Ian, thank you so much for joining me today.

Ian Karnell: 37:43

Rich, thank you for having me. I appreciate it.

Outro: 37:46

Thanks for listening to The Customer Wins podcast. We'll see you again next time, and be sure to click subscribe to get future episodes.