Introduction to Eric Williamson and CallMiner

Alex Gluz: Alex Gluz here. I'm the host of the Revenue Engine Podcast, where I talk with top leaders in business, marketing, sales, taxes and more.

This episode is brought to you by T.A. Monroe Digital. At T.A. Monroe, we build world-class revenue engines for B2B SaaS and tech companies. What that means is we do everything behind the scenes to generate you tons of qualified leads, opportunities and customers.

We build highly effective marketing systems and scale them profitably through our demand generation strategies. We think scientifically how to scale your demand and revenue. We bring decades of experience and expertise.

Email me at alex@tamonroe.com or visit tamonroe.com today to schedule a strategy session, or learn more about our mission.

We have a great guest for you today. Hey, Eric, nice to have you here.

Eric Williamson: Nice to be here. Thanks for having me.

Alex Gluz: Can you do a brief intro of who you are and what you do?

Eric Williamson: Certainly. Eric Williamson. I'm the chief marketing officer for a company named CallMiner. CallMiner is an intelligent CX automation platform.

To articulate what that means, we work with generally enterprise-level companies across the globe. Our platform captures and analyzes 100% of all forms of customer interactions: calls, chats, surveys and so on.

Then it uses those insights to inform our automation products, which would be voice and chatbots, as well as triggered surveys. All of that works together in an infinite loop.

We've been around for a little while. Probably most well known for conversation intelligence, but our automation tools are our newest offering.

Eric's career and CallMiner's move into automation

Alex Gluz: How did you get started? What did you go to school for, Eric?

Eric Williamson: Political science was my undergrad, and I think at the time I was like, "Maybe I'll go to law school afterwards."

But thankfully, whether you call it serendipity or whatever, my first role out of college was a BDR, or an SDR, as sometimes they're called. Especially if you're going to work in tech, I think that is probably one of the best first jobs you can get, whether you want to be in sales or marketing or anything. I learned a lot from it.

After that, I ultimately ended up on the ad agency side. Pretty much the first half of my career, I worked for large ad agencies: The Martin Agency, MullenLowe here in the Boston area, working on really large consumer brands. Everybody from JetBlue to Microsoft and Royal Caribbean.

I had the opportunity to switch in my career into the client side, and I chose tech. I've been working on the B2B marketing side in tech since 2016.

Alex Gluz: That's a great career journey. What brought you to CallMiner?

Eric Williamson: Before CallMiner, I was at a company here in the Boston area named Acquia, which is in the DXP space. They primarily compete against companies like Adobe, but they're the open-source option. It's built around an open-source project called Drupal.

We had a good run, and then we were acquired by Vista private equity. Once we had been acquired by Vista, there were several different opportunities at the CMO level.

At the time, this would have been about 2020, I was looking for something that was squarely in the artificial intelligence space. Even though AI was around then, it wasn't as pervasive as we talk about it all day, every day now. But I knew that's the space I wanted to be in.

If you really think about it, what's the best way to understand your customers? By analyzing 100% of all the conversations you have with them, any kind of interaction.

The fact that's where this company was rooted was interesting to me. Then the fact that we were able to analyze that at scale for enterprise-level customers, I felt was a very good proposition.

I'd done my research on the space, and it is a very hot space currently. This is why you're seeing a ton of M&A in the space.

I joined in 2020, and since that time we've gone through an initial rebrand. Over the course of my time there, we've acquired two different companies. In 2025, we acquired a company called VOCALLS, which is where our voice and chatbot products come from.

We also released our own product that we built, which was a triggered survey platform. I wouldn't call it a full rebrand, but it caused us to need to do a repositioning, since our suite had evolved from being about intelligence into really more of an automation player.

Alex Gluz: A small observation: when you say 2020, and this probably shows how old we are, we're like, "That was just recently." But that was actually six years ago.

Eric Williamson: I know. Especially in the tech world, a CMO staying that long in one company is kind of unheard of. But we've had a really good run, and we have a really great path forward.

Like I said, this is a very hot space right now, both in M&A but also in the capital that is being made available to it. I think that's indicative of how hot the space is right now.

Obviously, anybody and everybody's talking about AI and their agents, whether you're Salesforce and Agentforce and so on. Everybody's chasing after the same thing.

Our use cases primarily focus on customer service, which makes us a little different than maybe a Salesforce or a Gong or something like that.

How conversation intelligence and AI agents are coming together

Alex Gluz: To say the space is hot, specifically customer service, customer success, automation, is an understatement.

Just this week, we're working on a proposal for a client in this space, and they gave us a few of their biggest competitors and what they think of them. I don't want to name them, but this company's only been around for a few years and they're doing really well.

I understand why this space is super hot, because there's big demand for it. And then also, to serve this, it's a relatively good space for LLMs to step into. Maybe you can elaborate on this.

Eric Williamson: You have a lot of new entrants. With the rise of OpenAI and Anthropic and all the different models that are available out there, the barrier to entry in this, especially on the intelligence side, has lowered.

But what I'll say is a lot of the newer entrants here, you pick whether it's Anthropic or OpenAI, really it's an LLM with a thin layering of product on top of it, as opposed to being a broad analytics platform that provides deep insights.

What you also have now is the convergence of intelligence companies and what I'll call AI agent companies: chatbots and voice agents as well. Typically, those have been two different categories, but now they're merging together quite quickly.

A lot of that is due to the M&A that I was talking about. I think what we're finding is the players that came from the automation space, most of them started out mostly as chatbots, and then they moved into voice agents.

There's been one or two players, and this is why we chose VOCALLS, that started with voice. They started with the harder part, the voice agents.

But all of these automation players, if you look at, let's say, NiCE acquired Cognigy, we acquired VOCALLS, all of the intelligence and the automation are starting to merge together. What you've got is still a lot of players that just deal with the bots and have no intelligence.

When we did acquire VOCALLS, it quickly required us to step back and think about our positioning. Our superpower is our intelligence, but we're an automation company.

It became that connection between intelligence that comes from 100% of all the conversations that are happening out there that we're analyzing. That's what informs our AI agents.

Most of these companies that we're playing with in this space, they don't have the intelligence. They just have the automation.

Alex Gluz: Very interesting. We should meet in a year and chat more.

Eric Williamson: It's changing very fast, too. But what I would say is what we're seeing on the customer side, some industries are lagging behind a little more on embracing AI agents and voice agents, chat and so on.

BPOs in particular, the largest ones seem to be a little bit stuck in the past, and they can't get out of the butts-in-seats type mentality.

But when you look at financial services, when you look at healthcare, even though those are highly regulated industries, they are leaning in big time into AI agents. We have quite a few very large healthcare and financial services customers.

SaaS pricing and the shift to consumption models

Alex Gluz: Let's change this up a little bit. Pre-interview, you said the "SaaS is dead" headline is mostly noise. The real story is how survivors are shifting their pricing models.

There's a few folks I talk about this with. What's the lazy version of that shift that most companies still make?

Eric Williamson: I think it's really good clickbait. How many times have we seen whatever insert "is dead"? I've even seen headlines around the "SaaS apocalypse."

It makes for a good article because it's a provocative headline. I'm sure there's a little bit of truth in it. But all in all, I would say that most of the SaaS players saw this coming, just like they saw when we moved from zero percent interest rates and growth at all costs. You could spend all you wanted. All of this was happening at the same time.

I think the biggest shift is you're seeing a lot of these SaaS companies that are being disrupted by LLMs. They've quickly merged that into their core product, if they didn't already have some type of AI in their product to begin with.

The agentic layer is at the core of their entire product, as opposed to it being a feature.

Then, to the pricing model. SaaS is your classic seat-based, and what we're seeing is a shift to more of a consumption pricing model in addition to that.

We got lucky because we already offered seat-based as well as consumption-based before the SaaS apocalypse started to happen. We were able to easily lean into our consumption model more.

Probably the biggest shift for us most recently is the shift from the typical annual contract or three-year contracts to more of a pay-as-you-go. You'll find pay-as-you-go, especially for the AI agent automation players, that's generally how the pricing models would work.

Now we're seeing that begin to become the way of the world for how our pricing model works. We'll offer it as pay-as-you-go, at the very least just to get the relationship going with a larger enterprise customer.

Then, of course, you can get a fairly large discount from the pay-as-you-go levels if you are willing to sign an annual contract. I think those are some of the biggest disruptions.

Beyond that, I do think that certain use cases that are very administrative, a lot of repeatable, mundane tasks that may be a SaaS software model, whether you're in legal or some of the repeatable financial tasks, a lot of that can be easily disrupted with all the LLMs out there.

But again, most of those companies have already put an agentic layer into their product.

Alex Gluz: There is so much to discuss here. But for me, the laziest version I see is what I call the rename. Users become agents, and licenses become entitlements. But underneath it all, the math is the same.

A company announces the AI pricing pivot. New page goes live, big announcement. Then you pull up the old one, and the structure is pretty much identical. Same tiers, same per-unit logic.

The CFO still runs the math the same way and pushes it back the same way. Companies actually shifting the model are rebuilding what counts as a unit, as you mentioned.

The rest are just changing the label.

Eric Williamson: I think it's introduced a lot of the customers that would have relied on a SaaS platform. They're looking at this and thinking, "I can go build this myself."

There's some truth to that. There are certain aspects of it that a company can build in-house. But I think what they're going to find is the total cost of ownership to continue to maintain that, but also what they can actually build and maintain in-house probably is not as powerful as the platform they were paying for before.

There are still gaps there that they can be working with vendors to fill.

I think you're going to see a little bit more of a combination of DIY and, whether you call it SaaS or just AI platforms out there from vendors, that end up providing the solution. Especially for large enterprise companies who have 30 use cases that they're trying to apply their software to.

Build versus buy and AI use within marketing

Alex Gluz: So many interesting topics touched on here.

What do you think is buy versus build? What is the line here? How do you approach this internally?

Also, as a CMO, as a marketing leader, what is the messaging and position you use versus somebody saying, "Can we build this ourselves and control everything?"

Eric Williamson: I'll speak to it in terms of what we're seeing with our customers and in the market. But then I'll also flip it around and speak to it in terms of how, as a CMO, I'm running my marketing department and what we're doing with AI.

First, think about it from the customers and the market side. It's a little bit of what I was talking about before. We are seeing a lot of companies consider that they could build something themselves.

But we are definitely finding that, whether it's a year in or six months after they start this, they realize that they're not actually going to be able to fully replace all the capabilities that they had from their SaaS platform or AI platform before. They are realizing that there are gaps to this.

We're seeing some of those companies come back. The ones that said, "We're now building all this," they'll come back, and they'll need parts of what they needed before, but not the full suite. I think that's happening a lot.

I think there's a lot of pressure on CIOs and all the different Cs coming from the CEO because they're reading, "You guys can just go vibe code all this. Why are we spending $250,000 a year on X, Y vendor when you could just use Claude Code and go build our own stuff?"

Some of that is misperception that the CEO is pushing down on the rest of his or her organization. They find out a year later that they couldn't actually do it.

But there is some reality to the ability to do this. I do think a lot of the platforms out there have made their architecture flexible enough where we're able to integrate, regardless of whether you're using OpenAI or Claude or whoever else. There are 100 different great models out there.

If I flip it around and think about how our company, and particularly marketing, is using AI, our company is using AI in two forms.

One, nearly everybody I can think of in the company is using ChatGPT, or I prefer Claude of the two major ones. For a lot of the things that I need, which is research, some of the early versions of content that I might need to write and so on.

But the problem with that, in a lot of cases for a company-wide scalable program, is we're not allowed to put any proprietary information in there. I can't take that and put customer information in or anything else.

We do have Copilot turned on throughout our organization, and it's embedded within Teams and everything else. All the different Microsoft apps have Copilot pervasively throughout. Company-wide, that's how we're using it.

Specifically for marketing, we're using what I just mentioned, the internally approved Copilot. But then also, almost everybody in my organization is using mostly Claude. I found that to be the preferred LLM for marketers.

That largely is about content and the things we're creating. But we also have a couple of other things going on.

We do a lot of video interviews as well. Probably not as good as yours, and we definitely don't have your audience. We're using Riverside for a lot of our interviews. We end up utilizing the tools; it does AI cutdowns.

We are using 6sense, which is our ABM, our account-based marketing intent platform. We're also using them for AI SDRs.

A lot of the repeatable, high-volume outreach that normally a human BDR would do, we are now using an AI SDR that's doing back-and-forth, reciprocal-type outreach, mostly via email.

It takes over a lot of that outreach, and then once a connection is made, we have that transition to a human BDR.

Just a couple of examples of what we're doing with AI as an organization. Our engineering team is definitely using Claude and a variety of things.

Dividing marketing work between internal teams and agencies

Alex Gluz: It's a very similar stack for a lot of companies. Similar, not the same.

You came from paid media agencies, so you can relate. That's what we do, the agency life.

When you're looking at your resourcing and looking at your goals, how do you think about, "This is what we do in-house. This is what we build in-house. But we're going to hire an agency for this, maybe for paid, for SEO, maybe something else"? How do you approach this?

Eric Williamson: I've got a demand gen team, an events team, product marketing and so on. A lot of the content creation, building white papers, blog posts and things like that, is done in-house. Plus, we have one freelancer that works as part of that overall team.

If you think of a classic demand gen engine, it needs the white papers, and it needs the different content that we're pointing a lot of our campaigns toward. That's generally done in-house, plus a little bit of freelance.

When it comes to the agencies that we work with, largely that has to do with our paid media. A lot of the paid search, a lot of the display ads, as well as third-party publications that we will run campaigns on.

We also use an agency for our SEO. But at this point, a lot of the SEO efforts have been split between SEO and, I think it's got a couple of different terms, GEO or LLMO. Essentially, trying to make sure that we are showing up in more AI Overviews, as well as in answers from Claude, ChatGPT and so on.

Those are the things that we deem as agency. But internally, my marketing operations team own everything to do with Marketo and the connection into Salesforce, and we do all of our email marketing on our own.

It's really just paid media and organic-type media, which would be SEO and LLMO. That's the split there.

The way I think about this, and some of the major shifts that I think we're seeing in tech: I don't know how long ago it's been, but you think of a company like HubSpot, which essentially invented inbound marketing and your classic MQLs.

Now you're probably reading headlines: not only is SaaS dead, but the MQL is dead. I don't 100% buy into that. I think it's really just more that the MQL is another intent signal. But we still do rely on some form of MQL scoring for our inbound.

The last part about this is, in the age that we're in right now, brand has become really more important than in situations before, where you thought about it as a different line item in your budget than demand gen, which was hardcore driving MQLs.

The lines are not that evenly cross-cut there, where I don't say, "This is my brand budget. This is my demand gen budget." If you have a good messaging hierarchy, a lot of your demand gen should be driving brand as well.

But also, all the efforts that you put into LLMO, all the efforts that you put into your analysts, your analyst relations to do well in waves and things like that, all that's driving awareness. It blurs together quite a bit more.

Alex Gluz: You touched on so many things. I understood you being on a paid media side made you a little biased because you've been behind the scenes, and you understand what a good agency can do and how much value.

Eric Williamson:

In the agencies that I worked on, we did everything from massive TV campaigns all the way down to paid social media.

The one thing that my background has benefited me on is I don't think a lot of B2B marketers have the foundational training that I did, coming from the agency world, especially on the consumer side.

You learn how to write a creative brief. You learn what a brand hierarchy looks like and how that spills into a messaging hierarchy, and how to have something like that so that you're not just constantly chasing tactics.

When I shifted into B2B, I found a lot of it to be very formulaic, lacking any emotion in terms of the creative or the copywriting. Hopefully I've retained some of that, instilled some of that into my overall approach.

I think a lot of my career has been serendipitous, but that was a serendipitous shift, to be able to come in with all those tools into B2B. Especially when we're doing rebrands, this is what I spent a lot of time doing before. It's been beneficial.

Agency benchmarks and human judgment in AI content

Alex Gluz: This is an ongoing conversation. Now it differs: do we hire somebody in-house, or do we build something with AI?

Eric Williamson: AI has been an incredible productivity multiplier. But at the end of the day, and this is what I meant about really understanding the foundation, if you don't know what good looks like, or if you haven't spent your career learning how to do some of these things yourself, then all you're really contributing to is a lot more AI slop.

There is a ton of that out there right now. While we may use it to get to a V1 or a V2 of a blog post or a white paper or whatever, we will never publish anything or send something live without having a lot of human intervention from the people who know what good looks like.

Alex Gluz: That's becoming a bigger and bigger issue, the AI slop.

Eric Williamson: Go look at LinkedIn. You can almost spot them.

Half of them, clearly they didn't edit anything. They just took it straight out of Claude or ChatGPT and dropped it in there.

Alex Gluz: I'll tell you something I just learned today, Eric. The writing, hands down, we all see that. Apparently, for pictures, a lot of people are using Nano Banana.

When you see somebody lying on the beach, saying, "I'm lying on the beach right now while my AI agents are running the show, and I'm making billions of dollars," this guy is in an office, probably working 12 hours.

I played with the tool for voice, and it's kind of scary, honestly, how close it came. It's still not there. But I sent some samples to my friends, and they said, "I just know what it sounds like, but if I wouldn't know, I thought it was a real person."

Voice agents and the handoff to human customer service

Eric Williamson: We have a voice AI agent in our OmniAgent offering. It's voice and chat.

What some of our customers will do is they will find some of their best human customer service agents, and we will clone their voice.

Also, if you think of a large enterprise company, they will have the voice agent have a slightly different tonality, but also accent. If the voice agent is talking to a customer in Texas, it may say "y'all." But if it's talking to somebody who's in the Northeast, it's definitely not going to say "y'all."

It's interesting what it can do. This is still early days, but we're seeing some really interesting capabilities, and scary.

Alex Gluz: And scary, yes, for sure. If I call customer service, as long as my problem gets solved, I don't have an issue with that.

But that's definitely changing the work model and the financial models, especially for these big enterprise companies, which are serving quite possibly millions or tens of millions of people, maybe each month or a year.

How's it changing the overall structure of the company, going from that human at the other end to now possibly replacing at least some of it?

Eric Williamson: I think we're going to see the shift. Let's say it starts off at 100% human customer service agents. I think we're going to see a shift to at least 50% of that being automated fairly quickly over the next two or three years. It's already happening.

The key here is companies understanding, think about the customer journey, which touchpoints to automate and which touchpoints and use cases to leave with the humans.

What we've found is this is why intelligence is the superpower here, and this is why we think it's our key positioning. If you have the intelligence of millions of other customer interactions and you've analyzed that, you could say, "Which interactions should be automated?"

It's going to tell you, based on millions of interactions, "These are the ones that are most promising from an empathy and everything else standpoint for you to automate. Would you like to automate?"

You can click it, and it will go ahead and create the bot for you.

I think this is where the shortcoming happens when you're like, "Great, we've just bought a voice and chatbot platform," but it has zero intelligence. You're basically then making your best guesses as to what to automate. That's really not a good model.

Alex Gluz: I agree. This is a great way to put it. I think this is a fine balance. Finding this fine balance on many levels, at the business level, also on the human level, because there are other humans at the end, with employees.

Eric Williamson: The other thing I would recommend is, regardless of whether you used intelligence to figure out what you're going to automate, start small. Don't try to automate everything right out of the gate.

Then make sure that there is a very simple and graceful handoff between AI agent and human.

What you don't want is your very human customer yelling "human" 10 times before the AI agent finally transfers you to the human. Make it simple. Make it easy, or your brand will definitely be punished for it.

Alex Gluz: I think this is becoming the most popular word now: "Talk to a human, talk to a human."

I just did this a few days ago somewhere. I said, "Talk to a human," because you're not going to be able to solve my problem if it's a complicated problem.

Eric Williamson: I already know a bot probably can't, but usually I'll give it a shot first. That could just be because I'm in the industry.

But if I know it's a complicated situation that needs explaining, I'll just go straight to the human.

Brand, buyer research and visibility in AI search

Alex Gluz: Great chatting with you. I have a few more questions for you. If you had to bet on one growth lever going into 2026, what would it be?

Eric Williamson: I'm going to stick with my area, which is B2B tech. I talked about it briefly earlier, but as much as you need the demand gen engine to handle your account-based outbound and your inbound, I do think there's been a big resurgence of brand.

I don't mean a line item in your budget: "I'm going to do billboard ads." Although I've seen some tech companies do some pretty cool billboards, which nobody would have spent money on five years ago.

But I do think brand is an important piece of the overall job of marketing now.

Largely, I say this because there's some research out there that says essentially 80% of the buyers out there, whatever your target audience is, they've already made their decision, or they at least have a favorite, before they even are willing to talk to a BDR.

They want to do as much of this research as they can on their own. A lot of that means you have a lot of work to do to get on their radar. That's just simple awareness, but also get into their consideration set.

Largely, it's ensuring that your messaging hierarchy that infiltrates across all your demand gen assets is tightly connected to your core positioning.

Don't get sucked into using so much corporate jargon that nobody knows what the heck you're talking about, and you could slap any logo on it.

You still have to keep in mind that you are marketing to other humans. You still have to trigger an emotion that you're going to solve a problem for them, or something that keeps them up at night. Don't forget about emotion.

We're just spending more money on certain things that definitely would be considered brand, and I would include LLMO in that. But also thinking about the things that feed LLMs, so G2 reviews and different things like that have become more important than they were just two years ago.

Alex Gluz: I definitely would add to that thought leadership. Thought leadership is super important, like what we're doing right now, just talking to a human and explaining everything in these simple terms. If needed, we can go deeper.

For LLMs, Reddit is also super important nowadays.

Eric Williamson: YouTube is a huge feeder. We weren't spending much money on YouTube, so we're doing a little bit of paid, and it's not a demand gen campaign. It's all awareness spend. It's all about impressions.

But we're also just paying more attention to YouTube now because of LLMs.

Mentorship, brand advice and closing remarks

Alex Gluz: Great speaking with you, Eric. I have one last question. Who were your mentors, and the best advice they gave you?

Eric Williamson: Hands down, the person that was my mentor, but also had the biggest impact on my career journey, is a woman called Kristen Cavallo.

She is currently the CEO at The Martin Agency, which is where I met Kristen. She wasn't the CEO then, but she's had an incredible career, and I worked at two different agencies with her.

As far as advice, since we were talking about brand, I think probably the best advice I learned from her is the best brands know one thing really well: you can't make everybody happy.

You can't pander to everybody. That's a recipe for a weak brand.

The best brands know who their customers are, both from data, but also because they've talked to them regularly. They know that they will make certain people very happy, and other people they know they can't make happy.

Alex Gluz: Great advice. We've been talking to Eric Williamson, chief marketing officer at CallMiner. Eric, where can people learn more about you?

Eric Williamson: Go to callminer.com, or you can find me on LinkedIn.

Alex Gluz: Great. For our listeners, if you'd like to learn more about how we can help you build marketing revenue engines and paid media that empower growth, visit tamonroe.com or contact me on LinkedIn as well.

Eric, thanks so much.

Eric Williamson: Thank you.

Alex Gluz: Thank you so much.