Jacob Andra: AI Is NOT the Ozempic of Business: The Truth About AI Efficiency

Business. Business. Business. Business.

Welcome to Go to Market, a podcast for RevOps, sales, and marketing leaders to dive deep into industry insights and best practices. Go to Market is sponsored by Fullcast and Silicon Slopes. It is my pleasure to introduce Jacob Andrus, CEO of Talbot West. Jacob, how are you?

Doing great, Amy. Thanks for having me on.

It's a pleasure to have you on today. Thank you. I, I have heard about you before as, as a very sophisticated AI leader. And so it is really an honor to have you on and, and have you talk about one of the most pressing things that we can talk about right now, AI. How efficient is it? How effective is it? But first, before we get started, could you tell us a little bit about your background?

Yeah, so I'm a longtime entrepreneur, self-starter. I did do about a decade stint in corporate, you know, in like various tech roles. But other than that, for my entire life, I have been more in the entrepreneurial space in a variety of ventures. And I started Talbot West to feed a real need I saw in the industry, in the market in general, which is unbiased perspective and leadership for companies on how to adopt AI, because almost all the advice or opinions you're getting out there is extremely biased.

Absolutely. I think that that is really what we want to talk about today. So what is bias? Like, what do you, what is your hot take on this whole quagmire of what everybody's talking about today? AI.

Yeah, exactly. And you were referencing my LinkedIn headline. I think you like that one.

I love that. Can I just read it? Here it is. AI is not the Ozempic of business efficiency because most of us think that it is and we think it's like the cure-all. So, tell us, what do you mean by that?

Yeah, I'd love to, and I wish I could take credit for originating that. I actually heard it from David Stahl, President of Hillcrest Bank, who heard it from somebody else. But whoever originated that, it's absolutely brilliant. And there is a lot of magical thinking and hype around AI. And back to your question of, you know, why information is biased. It's biased for a number of reasons. People who create a specific tool platform or technology tend to over-index on its capabilities. And, you know, because they're so steeped in that one platform or capability, they tend to overestimate the applicability of it to a wide range of business conditions. And then beyond that, they tend to, there tends to just be a lot of misinformation of either on the one hand denigrating the capabilities of AI. So people with maybe a sort of opinion or hot take that it's no good because large language models hallucinate. And on the flip side, people who just think like AGI is upon us, everything, you know, like it's going to do everything for you in the overnight, right? So there's just so much extremism on both sides and biased opinions. And so a lot of business leaders have kind of magical thinking about the capabilities of AI that, you know, maybe I can just subscribe or find the right AI tool and somehow sprinkling it over my business will solve all my problems. And that's of course not the case, right? So for something, for some technology like a large language model or other AI tool to actually meaningfully change business outcomes requires a lot of orchestration, a lot of things to be handled and lined up, you know, data readiness and the right architecture, the right scoping and change management. I mean, there are just so many things that come into play. So it's not this pill you take that will magically make your problems go away.

Yeah, absolutely. So what are a few things that we should be watching out for when we see ourselves or our leaders kind of engaging in magical thinking? What are some cues that this magical thinking is happening?

Well, rather than talking about the, the limited bounded capabilities of a specific technology solution or platform and its direct applicability to specific narrowly scoped use cases in your business, it's more broadly looked at as, you know, this transformative thing in its own right. An analogy I often draw is, you know, could you give ChatGPT a prompt that just says, create me a $10 billion business with, you know, 25% margins, keep me at least 50% stakeholder, thanks, you know, put the money in this bank account, I'm going on vacation. Of course not, right? Like, nobody would actually expect ChatGPT to be able to do that, right? But you know, people just somehow think that like AI, you know, is going to somehow like, the complexity of a business and its operations is equally as complex, right? And so the idea that AI would somehow know how to just come in and, you know, make a change. I mean, this is like a very narrow set of technologies that have very specific functions and they have to be sort of applied in a very strategic way. I mean, the devil is totally in the details. You have to be extremely realistic. Realistic about what they can and can't do and all of that sort of thing.

That's really helpful. What do you do when you've bought a technology and it doesn't do what you think it's gonna do?

What do I do?

Yeah.

I mean, that doesn't happen for me 'cause I make sure to be realistic. I make sure to be realistic.

Which is why you created Talbot West, right? Tell me a little bit about what Talbot West is.

Yeah, so think of us as a consulting firm bolted onto a solutions architecture and engineering practice. Practice. So that means we can do it all end to end, but it really leads with the consulting side of things. What makes our consulting especially impactful is because we do it all end to end, when we scope a project out and say, you should do this, we actually know it's going to work. So that gives us a lot of credibility. And what makes us different from an engineering practice is we have extensive expertise in knowing how to apply technologies to real-world business problems and outcomes and priorities rather than You know, there's plenty of dev shops or engineering firms. It's like, hey, if you, if you tell me exactly what you're trying to do here, you know, we can build it for you. But most business owners actually are dealing with kind of a fog of war situation where the challenge is we don't even know where to start. We don't even know how to tie these disparate pieces together. And so we come in on the consultative side and help with the prioritization aspect. Like, let's take just a lay of the land of what's actually going on in your business, where you are currently, where you'd like to be, what are the opportunities, problems, and What's the bottleneck? And let's have a very clear-eyed, pragmatic look at what different AI technologies and other types of technologies working together can do to make a difference, what that's actually going to cost, what's a realistic timeline and all of that. And out of that can emerge a clear prioritization.

Interesting. So if I were to give you a problem that I have and I say, hey, here's my business problem. I need to grow 25% with 80% gross margins this year. I want to do it using this technology. Is that the kind of problem that you solve for me?

Yeah. And in that case, I would, I would not say it's a necessarily a problem, but more of an opportunity, right? So, you know, but, but yeah, that's a great opportunity and it's a great goal. And so we would talk about how realistic that is. We would actually take a deep look at your sales, you know, your business development efforts, what technology you're currently leveraging, your leveraging processes, systems, people, how well are they trained, you know, and there's probably some amazing opportunities to enhance your business processes through a combination of workflow redesign. Which we also do training and some AI enablement mixed in. So it's not going to be— almost no business goal, problem, or opportunity is going to involve just a technology or just one thing. It's going to be very multifaceted.

Yeah. So we, we always think in terms of consulting and especially RevOps consulting, we think of people, processes, systems, and data. And it sounds like that's what you, that's the lens through which you approach it as well. Is that right?

Yeah, absolutely. And it sounds like you already have the right lens and approach. So let's just say we were consulting with Fulcast on this, you know, we wouldn't try to recreate the excellent analysis and understanding you already have. So if you've already mapped that and have a very good sense of it, then we could cut right to the chase, which is you've got this already mapped out. Let's talk realistically about what AI technologies are out there and how they can be bolted into the understanding and road mapping you've already done of your systems, processes, and workflows.

Mm, I love that. Okay, that makes a lot of sense. And how do you differentiate yourself between what you do and other firms? Is it, is it an AI lens through which you work?

It, it can be, absolutely. So if we're differentiating ourselves to more traditional consultancies or legacy incumbents, then we would absolutely use the AI lens that we're AI native, built from the ground up. We leverage our AI technologies internally. We have a deep understanding of the entire AI landscape, not just generative AI and large language models, but cutting-edge ML and all types of other technologies related to or adjacent to the AI space. So that's how we would differentiate ourselves from traditional incumbents. But vice versa, if we're trying to differentiate ourselves to like, let's say an AI startup that, you know, there are plenty of those obviously, or even something like OpenAI or Anthropic, we would differentiate ourselves by saying we have deep expertise, decades of experience in business operations systems and processes. We understand change management. We understand how all the pieces have to come together, data structures, data readiness, systems integration solutions. Architecture and all of that sort of thing.

That makes a lot of sense.

Yeah.

So what's your, like, I know that our time is short here, but what is your final kind of word on the street on AI? What, what are some tips that you have for our listeners to evaluate different technologies as, as you're thinking about them? Do you have a couple of best practices you could share?

Yeah. And I love Claude and look, large language models are cracked up to be the be-all end-all and they're not, but Claude is actually amazing. It's the best out there in terms of all the commercial large language models available. I would create a, I would get a Claude subscription, the kind of paid enterprise or, you know, teams Claude subscription, get all your people on it. And there are a lot of ways to use Claude really effectively. One thing I would do is create a Claude project for anyone who doesn't know that, just Google it. It's very easy to do. It's like a CustomGPT and ChatGPT, but create a Claude project and basically say, you're an advisor and I want you to advise me from a very pragmatic perspective, grounded in the actual reality of what technology can do and don't. You're not to sugarcoat things. You're to help me, you know, navigate this effectively and then give it a ton of context about your actual business. You can upload a lot of files, documents to it, and then that'll be a great advisor for you. Now you still shouldn't trust it implicitly. You have to have some discernment about. Its outputs and how relevant they are, but it's going to output a lot of things you've probably never thought of. And if you synthesize its guidance with your own understanding and intuition, I think you'll get a lot further than not leveraging it.

I love this. What excellent advice. I, I really appreciate that. And again, you know, it, it goes back to say AI is not going to run your business, but AI is going to help you. And I appreciate the guidance on which ones to use and and what to say. So, I think that's really exceptional. All right, my final question for you today is clearly you are at the top of your game. You are a true expert in your craft, and how do you stay that way?

What do you do? Yeah, so constant experimentation internally for Talbot West's own operations, constant being, you know, trial by fire in the field with clients, both with the implementations we are doing for clients, as well as going in and seeing what's failed, what's not worked for them with other attempts, other consultants, other implementations they've tried. So constant learning in the field and internally in our own operations. A lot of reading and synthesis of, you know, what's happening in the industry and just taking all that in.

That's wonderful. Well, thank you again, Jacob. I really appreciate your guidance and I look forward to hopefully working with you in the future.

Absolutely, Amy. This has been great.

Thank you so much.

Guess that's business. Business. Guess that's business.