Get Operationally AI First Or Get Left Behind (No Hype)

You're not optimizing a funnel. Your app, you're optimizing a specific activity. But what's creating the opportunity for the rep to send a better email?

Cold outbound doesn't work. You can't cold email. Email is dead. No one picks up the phone. iOS 16, with call screening, is going to kill cold calls.

And if you're lazy, you're going to suck as a founder. I don't care if you're getting venture cash or not. It's like you got to go learn how to be operationally first.

Welcome back to another episode of the Bridge the Gap podcast powered by Revenue Reimagine. Today's guest is Nick Zeets, founder of Smoke Signals and one of the sharpest voices calling out the nonsense—and there's a lot of nonsense in the world of go-to-market AI. His background spans SaaS growth, sales leadership, and advising GTM teams. And now he's on a mission to help companies stop chasing the hype and start building real signal-driven strategies that cut through the noise. This episode is going to be about clarity: what's real, what's not, and how go-to-market leaders can actually use AI to win. Hint: it is not asking ChatGPT to write your cold outbound email.

Nick, welcome to the show, man.

Oh, thank you very much. I love all of that. I should just package that up.

Listen, we say all the time, you could take it, package it, reuse it, make it your walk-on. We'll add some music behind it. It'll be a good day.

That's beautiful. Yeah. One of my favorite humans on the planet is this guy Andy Mau. And Andy is a huge fan of LinkedIn voice notes. And I'm going to clip that and send it to him. Be like, "Here, buddy. Here's one."

Andy from Whispered.

Yeah. Yes, sir. I love that guy.

Andy is a good friend of mine. He's a good human. Great human.

Love him. We just had—

Oh, good. Oh, good. Well, that makes me feel even better about myself.

See, full circle. Now you know you're on a real show if Andy came. At least we know it's real.

Before it was just two dudes sending you an invite saying, "Hey, do you want to come chat with us for a little bit? We have two listeners. Why not?"

Yeah. We're not going to post this anywhere. It's just—

Oh, we're just going to record it in chat.

I like that a lot.

Come on, Dale. Ask a question. Awesome. So, Adam, Adam kicked it off really well: AI nonsense and go-to-market. What do you mean by that?

Yeah, I think it's a great question. This is the whole foundation of where I think we've been getting go-to-market wrong for a long time. And my gut tells me that AI is just exposing how bad we have all been at go-to-market for a long, long time.

Look, there are some interesting things that have happened over the course of the last 20 years. One: marketing grew out of just running field events, right? I mean, back in the day it was like, "Yeah, marketing does the field parties and all that kind of stuff." But if you wanted to scale it, it was a linear decision about BDRs and AEs, right? That's what we were going to do. And at the time it worked because you could send a single email and get a response from a C-suite executive in the Fortune 1000. Right? You could make a phone call and get a CFO of an F100 on the phone and actually chat with them.

And then what we did is the moment that we decided marketing mattered was also the moment that we gave them automation tools, right? We're like, "Oh, yeah." And so what we did was we indoctrinated the mission of marketing with a fundamental concept of lead volume, right? And squeezing that out. And again, before the channels got devastated by volume, it worked. It did work. It was more productive, right? And there was this whole excitement around what was happening with marketing because it was actually pretty easy to do a pretty decent job, right? I mean, you could do an Outlook mail merge and generate 100 leads in a two-week period of time, right? So you get your hands on Marketo and, wow, like look at this. This is insane.

Obviously, I don't mean to diminish the early-day marketers entirely, right? Like they were working with what they were given, and they were given a market, largely for most industries, that was open and receptive to getting these types of comms. Although I do pride myself on having been a fax machine bandit for a number of years early in my career.

Fax machine, my man. Ain't nothing like it. You fax. And here was my golden move: fax the facts to a pizza shop near the executive and have them staple it on top of a pizza and have the pizza delivered to the office.

Old school sales rebellion, bro. We got to get back to that. I'm a huge fan of the phone. You got to do stuff that doesn't fit. But that actually is kind of what I think the best marketers are thinking about right now: "Whoa, whoa, whoa. These are humans. How cool would it be if somebody sent me some cool thing stapled to the top of a pizza box?"

That's what I'm talking about. Yeah, dude. I'm hungry. And I don't know who doesn't like pizza. We'll just assume we're going to lose ten percent of the market on a campaign that's pizza-box-driven. But, okay. But I think what's happened with AI is because it's now not only do I have scale delivery in my marketing automation tools, but now I have scale content generation, and I use that word "content" very loosely in this context, right? We kind of can do anything. And what "anything" has resulted in, in most places, is more volume, because AI has given us what some level of conviction that it actually might connect with the recipient.

So anyway, this is kind of what I feel foundationally. And when you talk to senior leadership, they're also like, "I hate this." Doing it right—I hate it, but I'm going to make everyone do it.

Hate that you get it. But then you do it.

Yep. And then you still do it because you're skating towards the wrong KPIs, right? Everyone's skating towards a wide top of funnel and not skating towards a wide middle of funnel, right? And that's where I think everything falls apart.

It was like back in the day with Gary Vaynerchuk when he was saying, like, from a marketing perspective, you will not market on Facebook or social media platforms, but you'll read it every day. It's the same thing. It's just like you're doing the same thing you don't want to have happen.

One hundred percent. Right? One hundred percent. But unlike knowing that people will actually read stuff on social, they are not reading your spam-cannon emails.

No, I mean, I was just having this conversation with someone the other day. I pulled up my phone, and when I looked at my emails, they all—I had 23 that started with some version of "Hi, Adam. Hope you're doing well." And like, they immediately get deleted. There's so much AI out there, and so many tools are being launched every day—both for go-to-market and for not go-to-market. Like, how do you distinguish hype versus reality?

Look, man, I mean there's two pieces of this puzzle, right? The first is the way that SaaS and whatever next-generation naming acronym convention we want to use for everything that's coming up in the AI space in particular—they've been incentivized to tell their own stories forever. And it's like I kind of get it, right? The reality is I do think some of it is utter nonsense, right? I look at some of these tools, you play with them, and you go, "Oh, that sucks." Like, objectively, it's terrible. And it takes you—it takes a real practitioner all of about 15 minutes to be like, "Hey, this is a machine." Right?

Really poorly wrapped GPT wrapper, right?

Cut it out, dude. I can create a prompt that's better than that and plug it into any number of workflows. I've got ChatGPT, so does everybody else. Cut it out. Like, this is not a product, right?

So there's that. But I'm not entirely sure I think that's the narrative, the general narrative in the market about why these tools stink. There are actually a lot of them that don't stink. Sure. But the problem is they're all talking about themselves, and all of them solve one of 19 interrelated paper cuts, right? And so the real effort—the thing that is so difficult—is that leadership is getting pitched on this new paper-cut solution, and they're like, "Yeah, but like, what does this make possible for me?" Right? And that's not the story that they're telling, right?

And so I think one of the issues in play is you've got operational stars who have never designed soup-to-nuts a go-to-market program, and you've got go-to-market program leadership who have never had an operational job. And there's no translation happening between those two audiences. And tools are coming in and telling stories that continue to drive a wedge between those two stakeholders in the organization.

And so for me, my last company—that was a martech company in the HubSpot ecosystem—we exited in April. Our big story was, "Dude, it's not us, not by ourselves. No way."

It's us plus these guys plus HubSpot. If you stick it all together, here's an end-to-end program that becomes possible. And so whenever I chat with other martech sales tech startups, that's the thing I'm like, "Hey, you got to slow down. You got to find five friends that you think are really good at things that are complementary to what you do, and you guys got to start telling stories together."

But it's also why I think agencies have never been more valuable because we have the remit to look at all this kind of stuff. And you know, look, if all you did with an agency was just say, "Hey, can you give me five playbooks and tell me the tools and tell me that they work, and then I'll go implement," I think that's a service that's going to start to evolve in the market because big companies that have RevOps stars, marketing ops stars, they can do the work. It's just the vision disconnect between what's technically possible and campaign design. That's where translation just falls apart for me.

The problem is people are trying to get into the middle of the chain and they don't look at it end to end. So they don't understand that there's a flow from the marketing to the sales piece to the CS piece. And like yes, I can go write a better email as a salesperson potentially if I prompt it properly and give it, like, your LinkedIn profile and a bunch of other things, but that's only like one link in the entire chain. So that productivity—we were just talking about this prior to the show—that productivity paradox may be fundamentally creating a little bit of better productivity for yourself, but as an organization, it's not moving the needle.

You're not optimizing a funnel. You're optimizing a specific activity. And it's like, yeah, but what's creating the opportunity for the rep to send a better email? Because not for nothing, whatever work that's been done by demand generation or the marketing team is the context to the freaking better email. So how much better is the email absent an end-to-end understanding of how that person got there to begin with?

So again, what we're doing is acting like all these islands of activities aren't part of an archipelago. And it's like, no, no, no, somebody has to start having like a Key West vision and building the bridges between all of these things. And that's again where I think, you know, truly visionary marketing, GTM, sales, CS leadership along with their operational counterparts across marketing ops and RevOps have to start coming together, and they got to slow down to speed up, right? Because there's just a total lack of clarity.

But let's talk about that lack of clarity and let's talk about signals, for lack of better terms, right? Like this term "signals" has been used more broadly over, call it, the past six to twelve months. You know, cold outbound doesn't work. You can't cold email. It's dead. No one picks up the phone. iOS 16, you know, with call screening is going to kill cold calls. And like, it's the—I actually turned it off because I do get calls from unknown numbers that I do need to answer, and I don't want to screen every single one of them and put people through that pain because then, like, whatever.

But when we talk about signals, what is a signal? Like, what is a signal-based strategy? Because I've heard everything from, okay, we're going to email every time someone changes a job. That's a signal, right? Because they're ten times more likely to buy new technology because they need to make an impact.

I love that. So this is my favorite thing on planet Earth right now, except for my family. My family.

All right. So here's a couple of things that I think are really important. One, if the signal that you're working off of is easy to collect, everybody else has got it, right? So you know, people talk about GTM alpha and all this kind of stuff, mostly because the VC world needs to like, you know, act as though growing businesses is this weird kind of intellectual academic pursuit, and it's not, right? It's a natural thing with humans, and it's like, whatever, and I've got my beef with venture. But you know, you look at it, and one of my favorite questions to ask when we get started working with a company is going to the sales folks on the front lines. It's like, you got into a discovery call. What is the thing that if a prospect were to say it, you're like, "Oh my god, I'm going to get this deal." What is that thing? And the thing that's wild is that more often than not, there is a way to get that before that person ever gets into the discovery call, right?

And here's the thing: people are thinking way too small about what signal is and what it is for, right? I'll give you a really good example. We have a customer right now, and it's really a profound, it's a provocative solution, and they're really kind of on their own in the market. But they basically provide what's known as a cashless option for employee stock purchase programs, right? And the net is, normally, you know, I have my salary and I'm an employee and I'm middle of the road or kind of lower tier. I need to take home my whole salary, right? In order to live my life. But I could use my stock to buy my stock. But you know what? What if these guys actually paid to buy the stock on my behalf and I got a share in the upside of the business?

Now we're starting to increase participation. The companies love this for a whole bunch of different reasons. I'm jacked about these guys. Like they are big time.

Now the Securities and Exchange Commission demands that these programs and everything related to them gets filed, and then it's public, right? The SEC has a public API. It is called the EDGAR database. You can get any of it for free. Everything from the earnings call transcripts to their 10K filings to the specific document related to these ESP programs called an AK, right? Like, unbelievable treasure trove of information.

Now we know specifically that if we look in that information—and this is a place where building a very specific LLM-based, kind of generative "Hey, are we hearing signal here that indicates that this company has a real opportunity to touch on a nerve that has been exposed here in public?" Awesome. That's a great time to reach out.

But that's a tiny piece of what signal here is. What we're now doing for them is taking the last five years of all those filings and creating a gigantic meta-analysis of everything that impacts these ESP programs. Something that the CEO, CFO, chief human resources officer, and every other stakeholder in this decision is going to understand exists, and they're going to want to read it. It is a profoundly valuable resource for navigating this problem.

So now what are we doing when we reach out, when we see that signal? Delivering the Bible that helps them solve for it from an organization that gives them an easy button to address the entirety of the problem. And this is where I think people are really not growing the value of going super deep on signal.

Sure, maybe, right? If I have a former customer and they loved me and the utilization was really high and they move companies to a place that also looks like my ICP, sure, the most common signal there is great. Let's go. Like that's lowhanging fruit and super simple.

But if you want to get to go-to-market alpha, if you want to truly generate new demand for your business, those are not the signals that are going to drive you. Those are not the types of things that are going to give you foundational strength.

And if you think about it hard enough, the signal itself becomes the message, right? And so now it's a self-fulfilling prophecy. Hey, I can see that you're really struggling with the thing that we're good at. By the way, here's how to navigate that from a resource that is the best on the planet. And the reason that we built it was so that we knew that you were in this pickle. Right?

And so you become the authority on the signal itself, right? And so I think that this is some of where—because it's newer and again because you've got GTM engineers out there that are like, "Oh, I can find out when somebody's left their job." It's like, dude, have you led a GTM org? Do you have no experiences as a VP of marketing? No, you don't. So all you're doing is picking apart this nonsense, and it's like, all right, you're a clay cowboy. Good for you. And I'm huge on that, but like, this is not actually transformative to the business. This isn't actually changing the fundamental capacity of the organization to go generate revenue, right? It's got to be a lot deeper.

And I think that's why people that will be most successful in this evolution of go-to-market will be the people with the knowledge that are more business people that then intersect with how to go start developing their own AI automations just to play with it. So when they go back to a GTM engineer or a developer, they can articulate.

Properly. And you can iterate through it. It's the iteration. It's not the first part of it. There was the Diary of a CEO podcast. I was listening to what he was talking about from a neurological perspective, and they're saying like ChatGPT or generative AI are making people lazy because they're not strategically thinking about it. They're asking a question. They're getting a response. They're taking the answer. They're putting it into an email. They're sending it to their boss, whatever they end up doing, instead of taking the answer and being like, "Okay, I don't know if I agree with this piece. How do I make it better?" and they iterate through it just like we would. Just like in this conversation, that's really how you have a conversation with AI—so that you get to a perspective that's your own and not AI perspective that may be someone else's perspective.

This is the fundamental problem I think, even with people just learning how to write an email. You have so many people right now that use ChatGPT or Claude or whatever to do everything. You're really good at asking a question. I won't go so far as to say people are good at prompting, but you're really good at asking a question. But if you have to think critically on any level, you have no idea what the hell to do because all of your answers are coming from your AI, and you're going to get exposed really quickly in some interview where someone's like, "Great, we're going to be in a room together and you have no laptop, you have no computer screen. Let's figure this out."

Yeah, that's what I like to do. Turn the laptop off and then answer my question.

I mean, even my 14-year-old, his go-to answer now is, "Oh, I'll just use AI." Like, dude, you're not going to be able to use AI on your test. I promise you that.

That's right. Nick, let's switch this up a little bit as we go through this because if not, we're not going to get through all of our stuff. This one I find super interesting: Founders are under tremendous pressure to show that they're AI-first, AI-native. How do you advise them to navigate this ask from the venture community without, I'll leave it there for a second. I'm curious what you say.

I will say out loud here what I say out loud everywhere: The best way to avoid that is don't take venture money. I'm anti-venture, big time, and for lots of reasons. But I'll tell you that we are experiencing a period of time wherein if you are operationally AI-first—I'm not talking about your product. If you are operationally AI-first, you can build product-market fit and go to market with product with two guys, two gals, two humans after work. You don't need that money. And I'll tell you what, everybody's super excited about revenue, right? I just don't care. Being AI-first, I think being operationally AI-driven, there's no excuse at this point.

Drag and drop used to be a little difficult. There's relay.app. This guy Jacob Bank runs relay.app. Drag and drop, right? If you aren't at least looking at workflows and then saying, "Oh, this is a point within the workflow where there's lots of context, maybe I'll write a prompt." And if you can't write a prompt, it's because you're lazy, right? And if you're lazy, you're going to suck as a founder. I don't care if you're getting venture cash or not. You got to go learn how to be operationally AI-first. This is the thing. We've got a customer. It's a hardware company, right? Cool hardware company. They put servers on top of industrial hot water heaters and use the expelled heat from the servers to heat the hot water. It's called green compute. It's so cool.

Interesting. Very interesting. But it's a piece of hardware, and operationally they're AI-first, right? And they're saying, "How do we look like we're 100 people and stay 15?"

Right.

Let me flip the question on you for a second. It's not easy when you're starting to build a company. What if you have a $10–15 million ARR company that you started two years ago and now you're in this place of like, how do I articulate they may need funding, they may not need funding, whatever. How do you articulate and go through the process of the AI mentality, AI-native, whatever?

That's really interesting. So definitely a separate scenario, and oftentimes at that point you do have venture and you've got a board and it's like you're in the pickle that you're in. And it's not just $10 million. It's billion-dollar topline companies are also sitting there and going, "What do I do about the AI thing?" Right?

I think there's a couple things here. One, anybody who's paying attention to market reactions to afterthought AI feature introductions should take to heart the fact that the market goes, "Dude, like no." So that's one of the first things that I would say is that if you are going down that path and you're feeling the pressure to do it, you need to think completely differently about what you are and where you're going, right?

Do I think HubSpot got everything right at Inbound? No, I don't. But were there some introductions where it's clearly a version 0.9? You're like in its current state not usable, but I see where you're going, and this is a fundamental shift in how you think people will do HubSpotting, right? Yeah, they did it. They really thought about it. And unlike Salesforce that bought Einstein, left it on its own database structure, and then kind of half-assed the relationship between that functionality and what could have been possible. And as much as they like to go out there and blow a lot of agentic smoke, none of it's real. It's nonsense. Their customers and anybody who's real in that market is like, "This is not a thing. I'm out here buying all these other tools now and I'm stuck with Salesforce. My world is built into this very complex back-end database with all of our custom objects and relationships, and whatever. So I'm just going to pull stuff out of it, right?" Which is an opportunity in the market, but I think it's a clear differentiation between doing it correctly—which is a fundamental reimagining of what you probably are going to have to be, which I think HubSpot did nicely—and looking at Salesforce, which has done acquisitions in the AI space and really, I think, blown a huge agentic opportunity. And they'll continue to blow it, I believe. I just don't think that they're built very well to kind of innovate, and that's a whole other structural issue over there. But these are the big things. It's like you can't bolt it on. It's not a feature set. You got to go offsite with the entire organization.

I do think back to the operational pieces. That's really where we look at as well: how do we get AI injected operationally? So if we zoom out a little bit, you know, AI is all the craze right now, right? Like Dale said, how do you prove that you're AI-first? And what's the agentic AI and how are you using AI? It comes up in every interview, every podcast, every LinkedIn post. I would love to see someone do an analysis of the amount of conversations that don't reference the word AI. I think there's a lot of hype, and some of it may go away, some of it may continue to expand. But zooming out, 18 to 24 months from now, what role is AI actually going to play in go-to-market, and what's going to go away and just not be a thing anymore?

Yikes. Put your Nostradamus hat on.

You know what? What a brutal question to answer. I was on a conversation with a prospective customer on Monday and we were talking through their environment and their market and some other things. I was saying, "Hey, I kind of got on and took a look at your sales organization. It's pretty big. Like it doesn't make sense. Your TAM from a unit perspective just ain't that big. If we work together, I'm going to tell you one of the implications of us working together is that you're going to have both the insight and the capability to drop probably half that team."

And look, I don't like that part of what AI is bringing to market.

Yeah.

But I don't know what to tell you. It is right. And I'm talking about both BDRs and AEs in this scenario, right? You way too many of all of this stuff. Now, where are they spending all their time? Are they all just kind of like on the beach half the time? Dude, I don't know. I have no idea.

# Transcript

Know where they're spending their time, but there's no way you need a team that is that big for either of those roles. Both of them could be cut in half in, you know, eight weeks, for sure.

Right. What did they say to that conversation?

Honestly, the guy was like, "I gotta tell you, kind of feeling the same way."

And at least they were real about it.

I mean, I think for all the PR, you know, public facing, "Hey, we're here. We're a team. We're going to evolve and AI is going to open up new opportunities." Dude, privately in the boardroom, top to top, they're absolutely talking about how the organization is going to get smaller.

Sure.

And we have proof, right? Accenture in the UK dropped 4,000 early career analysts—just the UK division, 4,000 early career analysts. And then you look at there's a guy who founded super.com and he's built this really interesting AI-first bootstrapped company index. There are now dozens and dozens of companies in this index of companies that got founded 12, 18 months ago and are at 30, 40, 50 million in revenue.

Sure.

They never raised. They've got teams of eight, nine people, right? I mean, these are not outliers. When Instagram did it, it was an outlier, right? They had incredible viral growth. It made no sense that they got to billions in valuation with 17 people at exit, right? But these companies now, you can't ignore it. The board is having the discussion. And I do think that as we get more efficient, we're going to lose a lot of these people. But I tell you the thing that actually is freaking me out a bit is I do think that we're moving to a genetic arbitrage, right? My organization knows what it needs. Your organization has agents that represent its capacity to serve those needs. I don't want to be the guy who says that the sky is falling, but pretending as though this isn't feasible within two years probably—that's, you know, it's plain ostrich, right? And it's just not the right move. I think you got to be really thoughtful about what the world is going to look like. And I do think agents are going to be selling to procurement agents that are given RFPs that are automatically generated based on business data. Right? Our whole business is sitting inside a Snowflake. I don't even need to write an RFP anymore. I need a procurement agent to go talk to all the viable sales agents and then rationalize, do the arbitrage, and then implement. I think that's a place that we could actually get to. And I don't know, I don't feel great about it. I hope not. I think humans still have to be somewhere in there, but it's going to be a lot less of them.

So which jobs are maybe going away? Maybe all of them. That is a very interesting take and a little controversial one that we will have to sadly end it on because we're almost at time. But we can't do that without a little bit of rapid fire.

I'm very curious to see the feedback we get on that take. I like it. I like leaving on a hot note.

All right, Nick, what's one GTM buzzword that you'd ban forever? Could never say it again?

Man, this is a BS bingo game. God, I'm gonna go with all of the stage acronyms—the TOFU and BOFU. Get it out of here. This nonsense.

Love it. Love it. Don't call Dell. He doesn't like it. What's the hardest lesson you've learned as a GTM operator?

As soon as you think you're there, you're probably three depths of click away from actually being there.

I like it. What is one universal signal that you personally watch for in the markets right now?

Yeah. I think one of the things that's really interesting that you can consistently watch for is homepage messaging. It is a thing that you can always get. It is easily scraped in. It is parsible and it's fascinating to see how organizations are changing the way that they talk about themselves. And every single company that sells to a business should understand how that's evolving over time.

I like an obvious win.

I love that. As we wrap it up, last question. Dream vacation destination.

Oh, yikes. Oh man. I'm going to say Japan. I've never been and it just seems wildly awesome.

Nice.

Love that. I got to go too. My son's got to go. He keeps talking about it.

Gotta do it, man. Gotta do it. Nick, thank you so much for joining the show. We appreciate it, man.

Yeah. Thank you guys very much. This is fun.

Thank you, Nick. Appreciate it.