AI-Enabled Growth – Execution and Acceleration

In this episode, I am joined by Chris Marin, Associate VP of Digital Marketing at Inovalon, for a deep dive into how AI is moving from an experimental science project to a core piece of operational infrastructure. Chris brings a technologist’s perspective to marketing and shares practical ways his team is using integrated AI systems to change how they get work done.

We explore the critical distinction between automation and acceleration, focusing on how to compress the time between a market signal and an intelligent response without hitting a wall.

Chris explains how they have encoded expert best practices into agentic workflows for paid search and content diagnostics to ensure that as speed increases, quality actually goes up instead of down. We also discuss their Agentic Opportunity Scoring system, a fascinating example of tying AI and signal intelligence directly to revenue outcomes and learning loops.

This conversation is a must-listen for anyone looking to build a hybrid human AI operating model that prioritizes judgment and eliminates the “drudgery” of the boring stuff.

Episode Highlights

• From Automation to Acceleration: Chris explains that true acceleration is about compressing the time between a market signal and an intelligent response, rather than just “creating noise faster”.
• Agentic Workflows in Action: Discover how Inovalon uses specialized AI agents for competitive intelligence, persona-based feedback, and Google ad quality scoring to move from a blank page to a high-quality first draft in record time.
• The “Punchy” Standard: Learn about their custom diagnostics tool that evaluates content for persona alignment, reading level, and “punchiness,” driving internal competition to meet high-performance scores.
• Agentic Opportunity Scoring: We preview their system that combines LinkedIn data, historical conversion rates, and persona alignment to provide salespeople with a “story” for why they should pursue a specific lead.
• The Hybrid Operating Model: Chris shares his vision for a world where human judgment is freed from “the drudgery” to focus on creativity, while bots handle deterministic tasks within clear guardrails.

Check out the previous episodes in our AI Pipeline Series:

1. The AI Pipeline Growth Gap
2. AI and Signal Intelligence
3. Orchestration – The Decision Layer of AI Growth

If you are interested in discussing this or any other topic, let’s have a chat.  Reach out to me directly to schedule a no-obligation discussion. This isn’t a sales call, but rather an opportunity to talk through your questions and challenges.

Follow me on LinkedIn.

Subscribe to The Healthtech Marketing Show on Spotify or watch us on YouTube for more insights into marketing, AI, ABM, buyer journeys, and beyond!

Thank you to our presenting sponsor, HealthcareNOW, 24/7 expert shows, interviews, and podcasts, powering healthcare leaders with innovation, policy, and strategy insights.

Full Episode Transcript

Chris Marin: [00:00:00] And the acceleration part is fascinating because you can go really fast in the wrong direction or, you know, hit those walls. And so it needs to be coupled with intelligence. So it’s really about, you know, compressing that time between that signal, as you alluded to, and an intelligent response, not just any response.

Adam Turinas: Hello and welcome to the Health Tech Marketing Show. I’m your host, Adam Turinas. I’m the CEO of Health Launchpad. First, as always, a big thank you to you for listening. I hope you the subscriber. If you haven’t already subscribed, click the plus button on your apple. Hit the follow on Spotify. And you can also find us on YouTube as well.

Also, a big thank you to our sponsor, our great friends at Healthcare Now Radio. This is an online [00:01:00] network that brings together some of them, the smartest voices in healthcare, and we really appreciate the support they give us. Today’s episode is one I’ve been looking forward to for a while. I’m joined by Chris Marin, who’s from Innovalon.

Chris is a technologist at heart, but he spent the last couple of decades embedded in the marketing team, and he leads digital marketing. He leads analytics and sales enablement, and he’s been building a deeply. Integrated AI system that is really changing the way Innovalon gets work done. This isn’t a hype conversation.

This is a really practical one. We’re gonna dig into the difference between automation and acceleration [00:02:00] because they’re not the same thing. We’re gonna talk about agen workflows in things like we’re paid search, and then how to best encode practices into AI systems. So that you are not compromising on things like quality.

In fact, quality should go up instead of down. We’re gonna talk about how Chris and his team have built what they call AG agentic Opportunity scoring, and that’s a system that ties AI directly into revenue outcomes. We’re gonna get into a lot of other stuff as well. So with that, let’s get into the show and meet Chris.

Chris Marin: Chris, welcome to the show, my friend. So good

Adam Turinas: to have you on here. I’ve been dying to have you on the show for quite a while since I, I spoke with your boss, Mr. Nick, the amazing Mr. Nick Panai. He painted such an incredible picture about [00:03:00] what you guys are doing at a novalon. And I’m just looking forward to digging into some of the specifics.

But before we get into it, tell everybody a little bit about yourself.

Chris Marin: Well, first of all, thank you for having me on the show. Appreciate it. And yeah, we are having a lot of fun at Innovalon. I’ve had the privilege of working with Nick for a long time and we have a great partnership. So my background is as a technologist, and that is my passion, but I’ve been embedded in marketing now for decades, and I’ve run digital teams.

I’ve run sales enablement, analytics, brand social media, but my heart and my love is really using technology to enable all sorts of fun and productive business outcomes. So, so been, you know, using machine learning and ai, you know, since its inception, and it’s just amazing to see how much progress has been made in recent years Here.

Adam Turinas: Yeah, and it never seems to stop. It’s just relentless, you know? Was that yesterday? I’m one of the 30 million people who read the Matt Schumer LinkedIn [00:04:00] post about the new anthropic and open AI models and everybody’s, well, he’s created a lot of swell. I, I dunno, I have to sort of process it a bit, but we’re not gonna get into that today.

’cause otherwise we’ll spend the whole episode talking about that. So what we’re gonna get into today is this notion of. Execution and acceleration. So the premise here is signal intelligence helps you figure out who might be in market, what their intentions are. Orchestration is making a decision about what your next action’s gonna be.

You know who does what. Now we’re gonna get into the what and the execution and acceleration. So when I talk about execution, what I really mean is now that we have a decision, what’s the action gonna be? Right? Acceleration is about shortening the speed to value. Reducing time to value, if you like. And so it’s a little bit of a nuance there.

Chris Marin: Mm-hmm.

Adam Turinas: And particularly, and I think it’s particularly important issue in [00:05:00] healthcare where it’s, uh, best way I could describe it is, it’s a little bit like an analogy about sailing, which is that sailing is sort of perpetual hours of boredom followed my blind minutes of panic. And I, reason I that in healthcare is, is that, you know, sales cycle might be.

13 months on average, maybe 18 months, and you can go months of not hearing anything or having a few conversations, and then suddenly everything is right now and needs to be done now. And

Chris Marin: you better, you better be ready

Adam Turinas: and you better be ready. Exactly. Absolutely right. So I mean, you know, from your perspective when it comes to the use of ai, what do you think about this notion of ex execution and acceleration?

Chris Marin: I mean, I think it’s about the distinction between AI as a science project as a. POC that somebody’s playing with versus where it’s a core part of your operational infrastructure. And [00:06:00] so, you know, most teams have AI embedded in a dashboard somewhere. Everybody has chat, GPT or some equivalent. You have models that you can interrogate.

But if the motion is I’m gonna go ask the model something and I, I need to interpret it. Uh, and then I need to go manually enter something for anything to happen that’s useful. It’s better than nothing, but it’s really decision support with a human bottleneck attached to it.

Adam Turinas: What a nice analogy. Yeah.

Chris Marin: And so when I think of, you know, the future state that we’re all headed towards, it’s about having that ai.

Really embedded in those workflows and those core workflows for everything that we do. And that obviously helps with execution and acceleration.

Adam Turinas: Yeah.

Chris Marin: And the acceleration part is fascinating because you can go really fast in the wrong direction or, you know, hit those walls. And so, you know, it needs to be coupled with intelligence and with those guardrails because if not, you’re just, you know, you’re, you’re just creating noise faster.

Adam Turinas: I love that notion there’s a difference between automation and acceleration. That’s quite something to ponder on. I think one of the examples, best examples of. What you’ve described there, and I think one of the things that you’ve built is the ENT workflow

Chris Marin: mm-hmm.

Adam Turinas: That you created for paid search. Mm-hmm.

And I think it would be helpful if you kind of drill down a little bit and explain that in that context.

Chris Marin: Sure. So this was actually one of the [00:08:00] first Agen workflows that I built for Innovalon and I, it was close to two years ago that this was done. So this was, this is the archaic times to speak. But I think, you know, there are a lot of learnings that came out of that.

And essentially, I mean, the approach is. You know, how do you encode all the best practices for any given set of tasks? What do you, if you can elicit all of the knowledge that the experts do when they do it right, how do you encode that in a system that enables you to perform that at scale? And so when I looked at SEA map creation, there’s things that, that you ought to be doing every time you take that off.

You should be doing some competitive intel. So you all, what is out there in the market right now? Ideally, you’re generating multiple responses within ll, right? You wanna have different options to choose from. So there’s agents that do that. Competitive intel. We’re leveraging multiple AI models to generate responses.

I fine tuned a model on the best performing ads in [00:09:00] our space, and so that’s part of that mix. I have some agents that are spun up when the person that you know enters the information, specifies their target audience. We spin up entity, you know, agents that are evaluating those results through the lens of that persona, and then they provide feedback.

Then very importantly, you need to regenerate your results based on that feedback. So that’s called the reflexion pattern. And then after that, those ads are all scored well, also spun up an agent that represents the Google ad quality, you know, point of view that also provides input. All those scores are aggregated, they’re sorted.

And then what we give to our users is, is not just a rank ordered set of ads. But also the feedback that’s associated with it, the lens to which the, those agents are looking at it, and that gives them that first draft for them to pick and choose and modify as needed, so it compresses that cycle to that first draft.

Adam Turinas: And you know, using it for two years. What has been the [00:10:00] impact? Efficiency, better performance?

Chris Marin: Yeah. I think it’s mostly about efficiency and it’s also, it’s about, again, it’s about encoding those best practices because the time to do all of those things that I mentioned isn’t always available to teams that are just pressed for time.

And so when I look at AI in general, I mean just for all marketing teams. You never have enough time to do everything that you want to do. Or at least not at the level that you would like to do it. So it buys you time that you can invest in other things. And that’s why I’m, you know, always preaching to my team and others is automate the boring stuff so that we can focus on areas where we can deliver value, where we have judgment and can make a difference.

Adam Turinas: Automate the boring stuff. I think that should be on a t-shirt. That’s amazing.

Chris Marin: That’s not my, there was a wonderful book. It’s on Python, the Python program language called Automate the Boring Stuff. So I’ve stole it from there. I love it.

Adam Turinas: Yeah. As a non-developer, I hadn’t heard of it, but that’s really great.

I think that’s a very interesting way of explaining the [00:11:00] impact of it because it’s, there’s two sides to it. There’s the first side to it, which is you’re shortening the cycle because it takes less time to do the whole end to end thing, but you’re also eliminating the shortcuts. So you know, the old saying is, you know, you can go faster, you can have better quality, fast speed of faster execution.

Or a bigger scope, but you pick two, you can’t have all three. What you are describing actually means can get all three in a shorter timeframe.

Chris Marin: The Iron Triangle, is that what it’s called?

Adam Turinas: Yeah. There you go. The Iron Triangle. Yeah. There you go. You should be writing this podcast. You’re doing it much better.

You know? Any other examples of. In what you’ve done in the last couple of years in reducing time to value, or may also improving the quality of execution.

Chris Marin: I do think that just in terms of quality, you know, having these systems that are beck and call that can give us 10, 20 [00:12:00] variations for us to pick and choose from.

I mean that that has to elevate quality. I mean, we know this in our daily interactions with these systems, and if you imagine. Having each of us produce 10, 20 different variations. Like that’s just not gonna happen. And that’s obviously, that’s true. Not just in writing. That’s true in image generation.

That’s true Across all of these. And so if you couple those with systems that can automate the, the, you know, some of your own guardrails. So we have, for example, a contact diagnostics tool that I built that we host in Monday where our marketers live. That tool evaluates writing generated by humans, or LLMs or some combination.

But we have particular thinkers that we love, like make it punchy from Ms. Stratton. And so that system evaluates the writing in terms of its punchiness and gives back detailed responses. But it also does things like, it looks at alignment to our message banks, looks at alignment to our personas. It looks at the flesh kincade reading [00:13:00] level because we’d had a tendency to write a little bit higher than the average American can comprehend.

Um, and then there’s an agent that synthesizes all those results and says, okay, based on the totality of all these. Here’s your actionable next steps and if you want the additional details that’s, and so we’ve used that quite a bit in our work to get to, you know, better place faster.

Adam Turinas: And how much of that is now baked into the standard operating procedures.

Because the reason why I ask is that sometimes, you know, put a lot of effort into crazy, something like this, and it’s just hard to get people to change the way they work. Yeah. Despite the benefits.

Chris Marin: I mean, I have to say, with that diagnostic tool in particular and the punchy tool that sort of kicked it off, it’s a GPT called, is it Punchy?

It’s, it’s actually available to anybody that has Jet G pt. The reaction to that one was phenomenal. I think it, in part because there’s still a lot of agency on the part of the writers, so they, you know, they’re producing the work and then they can get responsive back to it, you know, so there’s a sense of [00:14:00] ownership there.

But there’s also a sense of competition and there’s just something about putting a number, a score just, you know, really drives certain people. And so actually just last week I heard, I, I will not hand in anything that does not get a score of, let’s say, an 82 or not. If it’s not 82, I’m gonna, I’m gonna keep rewriting it until it hits that punch in this level.

Oh,

Adam Turinas: wow. That’s good.

Chris Marin: That makes me happy. That makes our editor and then the head of content very happy as well. And, and then, you know, we have a similar voice and across all the, all the pieces. Yeah.

Adam Turinas: I’m curious when we last. Chas, I think you said that you guys had adopted Jasper. Are you using Jasper?

Mm-hmm. Yeah. How does that fit into the workflow?

Chris Marin: So we’re actually right in the process of setting that up. We’ve used a lot of, you know, custom tools that, you know, that we’ve developed. But, you know, I think, you know, they’re doing some really interesting things there. I know it’s, it’s, I, if not the most popular, one of the most popular tools in this spec.

But obviously you know the center, it’ll be a center of gravity for our content hub team. We’re gonna imbue it with, you know, [00:15:00] all of the, we call it, we built a knowledge block library for all of our AI projects. So we have markdown files, so it’s easy for the LMS to comprehend, but we have our messaging banks, our personas.

Information about the company, information about our products,

Adam Turinas: yeah.

Chris Marin: Voice, all of that stuff. So that’s getting ingested there. And, and it’s a way to standardize that voice and those outputs and those preferences when it comes to tone and Yeah. And that sort of thing in one, easy to use, you know, place.

Adam Turinas: That’s great. ’cause you’re providing guardrails. That’s really clever. Yeah. I wanna come onto guardrails a bit later on.

Chris Marin: Sure.

Adam Turinas: One of the things, promises of, of all this, that. I find really exciting is the notion of learning loops and continuous improvement. Yeah. Have you been able to, you know, to, to instrument that in any way?

Chris Marin: Well, let me tell you about one of our latest projects, and it’s one I’m kind of excited about, and it’s a custom system we built called [00:16:00] the Agentic Opportunity Scoring System. And so the idea with that one is, you know, we’ve had a lead scoring model in Marketo, like so many do. But we wanted to, you know, kind of take it to the next level.

And the idea was have it taken all of the available signal intelligence that we have at our disposal. So we have the people level data in terms of. Persona alignment. We have demographic information, we have LinkedIn profile information. We have company level information, so we can look at ICP Fit. We have data from outside of our digital walls, from demand base, Revit, and then we have really good historical data, um, conversion rates by channel, conversion rates by sales teams and so on.

And so the system is built in such a way so they can take in all this information. We have some. Agents that are doing things like looking at the persona alignment. We have some be math that occurs under the hood to calculate a probability that a given lead will convert. [00:17:00] And that we provide that, you know, that textual surround, so here’s the story for this particular lead and why you should prosecute it to your salesperson.

Adam Turinas: Mm-hmm.

Chris Marin: And so the way it ties back to learning is there, there is a learning loop as part of this. So as we stamp these records with these probabilities. With the given attributes that we’ve assigned, you know, the, you know, terms of personas and all these things, it goes through our various systems and then it’s fed back in to compute those next set of probabilities.

And so the system is gonna get smarter over time as more leads progress through it.

Adam Turinas: Fascinating. That’s really interesting. So essentially. Like with an earlier signal, you could start to say, Hmm, this one looks like it’s gonna be one of the good ones, or this one looks like it’s one of the ones that doesn’t convert.

Maybe. Is that, is that a possibility?

Chris Marin: Exactly. And it allows us to differentiate between, let’s say, you know, an executive performing a certain. Connection versus someone lower on, you know, lower in the organization. And just being [00:18:00] a little bit smarter about how we treat it. And then also then as we hand off these leads, there’s less research that needs to occur on the other end because we’ve, you know, handled that automatically to the degree.

Cause.

Adam Turinas: Talk about guardrails and governance. How do you keep the system safe and how do you embed best practice and do no harm into the system?

Chris Marin: Yeah. Well, I mean, that’s essential if you don’t have guardrails. I mean, you have the, you know, you have the flashy sports car that goes incredibly fast with no breaks, so you build those in.

Adam Turinas: That’s fun. Fun for about 30 seconds. Yeah,

Chris Marin: yeah, yeah. So, okay, so going back to that particular system, uh, I mean, obviously there are. The constraints programmed into the prompts of the various L and agents. We also have monitoring with evals with a system called OPEC that’s very popular. So you can see it at a sort macro level, how are things going and make adjustments as needed.

So we’re using prefect [00:19:00] for workflow orchestration, which gives you another layer of visibility. But I think at the core of these decisions, and this is something I wrote an article about, is we have at our disposal sort of. Three strand of, you know, it’s three substrates that we can assign tasks to. We have humans that are good at a particular set of things.

We have AI that’s fantastic at certain things, but you know, can fail miserably in others. And then we have this bucket of. Deterministic code automation. So this is your Marketo, your work autos, your zas, your Python, your math, things that will always do the same thing every single time. And so as the owners of these systems of orchestration, it’s our job to figure out which ones are done by, you know, which strand.

And so you wanna be careful about. You know, not putting things in the LLM bucket that need to, you know, that really should be in one of the other two that, that are a little bit more regulated though, a little bit more brittle, for example, if you’re in the deterministic branch. So I think that that’s part of it.

If somebody had [00:20:00] asked me, Hey Chris, go build an ai. Speed scoring system. The easiest thing to do is just spin up an agent and say, Hey, go do this. The way we implemented is a little bit more work, but you have more control and predictability because of those architectural decisions.

Adam Turinas: Yeah, that makes perfect sense.

Yeah. You mentioned AI in humans, and so that prompts me to ask you one question, which. It is a really interesting area and it’s the notion of AI SDRs.

Chris Marin: Yeah.

Adam Turinas: I think a lot of people, I, it’s funny, I demoed a system you guys have used called Synth Flow to a group about a month ago. The reaction was sort of generally like, well that’s really cool, but it wouldn’t really work with us.

But you guys have got, you guys have been using it. I think you’ve had some quite interesting experience with it.

Chris Marin: Yeah, and we started with what I think is the hardest use case, which is external outbound, SDR, and you know, it’s a fascinating space and it’s [00:21:00] one where. Again, you can see the, you know, changes in technology affect the end user experience, even course a month.

So I think when we started working with them last year, they had a latency that was, I don’t remember how many milliseconds it was, but it was long enough that it became noticeable to people. So if you speak. And there’s a pause that is an uncomfortable pause, and then they respond. It just breaks that flow, right?

They were able to shrink that, you know, to a third of what it was when we started, which made it that much more usable. So we actually did see results from it, but we paused those outbound operations because in terms of the balance of risk and ROI, it just, yeah, yeah, there is some risk there for inbound though.

Our sense is, you know, that is so much more common than it was. I mean, calling in, especially, you know, calling in late Thursday night, you know, having, you know, some sort of AI agent respond to particular type of queries. That is not uncommon now and. They’re so good in terms of being able to [00:22:00] imbue them again with that knowledge block set of, you know, here’s everything about all of our products and services.

That is a first pass. I think it’s, you know, it makes a lot of sense.

Adam Turinas: Yeah. I think that everybody’s a bit squeamish in our space about the humor, the AI agent talking to a prospect. But SDRs, you know, they send a lot of emails, right? And they do outreach through LinkedIn. Are you able to get it working in that area in a way which is less risky?

Chris Marin: Yeah. I mean, so, but always again, with a human in the loop in the right place. So we use a tool called Tega that is helped with that. And there’s, you know, any number of other tools. We have virtual visitor, which offers that functionality, which we have. Enabled. I think it’s one where the individual members, where it’s being sent on their behalf, there’s a period of time where they need to be able to say, no, I’m not okay with this, or I’m okay with this type of action.

Adam Turinas: Yeah.

Chris Marin: Uh, and completely take away the agency. There’s huge risk, not just for the individual, but for the organization as a whole. If it, you know, if [00:23:00] it’s not constrained within the right to the guardrail.

Adam Turinas: Yeah, that makes perfect sense. I’ve tried out virtual visitor and I know exactly what you mean. Is, is that, and I think they’ve been really smart about this.

Mm-hmm. Which is okay, this is what it’s capable of doing. It’s capable of sending out emails on your behalf. Before we do that, let’s just make sure you’re happy with it. So you got the ability to. Intervene until you don’t want to. Which I thought was like, oh, that’s clever. But they’re very, it’s very well thoughtful.

Chris Marin: And I just, one, one thing I’ll note, like, so that’s why we use reduced time to first draft. Our poor marketers here, the one are probably sick of hearing this, but that’s how we wanna frame most of these outputs. This is the first draft. And in many, if not most cases, we still need that human in the loop to make sure that it meets our requirements.

And so what we tell our people is you own. The outputs ultimately, whether it’s you or a machine or some combination thereof, you still own it. And at this stage that, that still matters quite a bit,

Adam Turinas: right? Yeah, absolutely. [00:24:00] Yeah. It’s, I think it’s getting people to think about managing the AI as if it was a, an entry level team member who’s absolutely brilliant.

You know, he is brilliant, incredibly bright, and incredibly productive, but you can’t let him loose without supervision.

Chris Marin: Exactly.

Adam Turinas: Yeah. Very interesting. Chris. Hey Chris, I mean, as you look out, I was gonna say, you know, three to five years, but honestly that’s, that’s, that’s almost ridiculous. Three to five months might be a better, better way of thinking about it, but in terms of where you are going, you know, pick a horizon that you think is useful.

Paint a picture of where you think this is gonna go for you guys and others should be thinking about.

Chris Marin: Yeah, I mean, you hear this term of, you know, overhang where the capabilities of the model are, you know, outpacing the capabilities of society, culture, business, operating laws and so on. And, and so I do think, you know, per that article that went viral yesterday and so many [00:25:00] other, you know, similars, essays.

There has been a qualitative change in the models, particularly with cus 4.6 and, and Codex by 0.3. And even, you know, the Gemini released from yesterday, they can do things at a level that they could not do three months ago. And so if you’re doing your five year planning cycle, I mean it’s, things are accelerating so quickly.

And I hope as a society we can, you know, find a way to deal with that level of change. But for assuming the current pace continues, you know, really the ideal end state for me is the hybrid human AI operating model. Yeah.

Adam Turinas: Yeah. Uh,

Chris Marin: and so what we are doing is. Some of the things, I mean, they may not, so just being very clear about here’s our processes.

And these processes are not just documented, but they’re actually embedded in systems, in our case, like monday.com, that are accessible and actionable by both humans and bots. And so then we can say, for this particular span, let’s keep it at the humans, but this one, let’s swap it out. This one makes sense to have the machines swap out.

So just being [00:26:00] very. Actionable and clear about what those boundaries are and respecting the strengths and weaknesses of those three strands, and just knowing if you fully adopt it, those people that fully adopt it can be so much more productive. And again, hopefully to free up. Time from the drudgery to do things that are exciting and creative and allow you to, you know, provide judgment, the things that humans are good at and flourish with it.

So I think the organizations that can shift to that

Adam Turinas: yeah.

Chris Marin: Uh, have a higher chance of survival than those that, you know, close their eyes or the ones that embrace it fully fire everyone. That’s where you get ai, slop, and all sorts of failures. There is still a place for humans in this mix.

Adam Turinas: Absolutely. To do what we do best.

Chris, thank you. I really appreciate the candor and the transparency that you provided,

Chris Marin: of course,

Adam Turinas: and clarity. It’s so impressive what you’re doing and where you’re going with this, and I think it, you know, you, what you [00:27:00] described, I think paints a really clear picture of how you can use ai, not just to create content more easily, but to shorten sales cycles without.

Cutting corners and to do that in a responsible way. And there’s clearly an efficiency impact. And I think the way that you described the lead scoring AgTech opportunity scoring a OS, let’s call it that, I think that clearly will have an impact on revenue. And I think that’s where you can start to tie.

AI to revenue in a very clear way.

Chris Marin: Yeah.

Adam Turinas: Great. Chris, thank you.

Chris Marin: It’s no problem.

Adam Turinas: What a fantastic conversation with Chris. My brain is whirring. It’s exploding. There’s so many insights in this one. Let me try and simplify this and give you four big takeaways. Take away number one, automation [00:28:00] is not the same thing as acceleration.

Just because you’ve automated something doesn’t mean you’re actually moving faster in a meaningful way. True acceleration is about compressing the time between signal and intelligent action, and doing that with context and guardrails built into the workflow. Takeaway number two. Agentic workflows can dramatically shorten the time to first draft, whether that’s a blog, whether that’s an email, thought leadership, a presentation, a sales script, whatever it happens to be to do, and it can do that while actually improving quality.

So whether it’s. Paid search content creation or aligning your personas with messaging, whether it’s encoding best practices into systems, it’s doing it without sacrificing speed, and that’s a powerful combination. Third, [00:29:00] Chris’s a Ag agentic Opportunity scoring system is an amazing example of tying AI directly to revenue, and that’s something that we’re all striving to do.

I think it’s a brilliant example of this. So what you’re doing is you are combining persona alignment, ICP fit. External signals and then historical conversion data, and then you are feeding that back into a learning loop. So you’re not just generating leads, you’re prioritizing the right ones with greater precision.

Fourth, the future isn’t humans or ai, it’s humans plus ai, a hybrid operating model where we deliberately decide what belongs in deterministic automation, what belongs in ai, and what still requires human judgment. So, and that’s, you’ve gotta combine that with strong guardrails and that’s how [00:30:00] you’re gonna be able to scale responsibly.

Chris and the team at innovalon are showing what it looks like to move beyond AI as a science project and into AI as a core operational infrastructure that helps you improve your revenue generation, and that’s where the impact happens, shorter cycles, better execution, and measurable business results.

If you enjoyed this rest episode, please make sure to subscribe to the Health Tech Marketing Show so you don’t miss the upcoming conversations. We’ve got a couple more episodes in this series on AI and I’ve just got back from Vive and I’ve recorded a ton of stuff with 15 health tech marketing executives that I’m gonna be editing and publishing over the next few months.

Some amazing conversations. So if you found it valuable, I’d really appreciate it if you’d leave a rating or share it with a [00:31:00] colleague who’s trying to figure out how to, how this AI stuff can impact their organization. And by the way, if you’d like to. Continue the conversation. You can connect with me on LinkedIn or if you want to chat one-on-one, nothing for sale here, you can book time with me through the link down below.

So with that, thank you for listening and I’ll see you on the next one.