E4: "Prompting is dead": Why GTM engineers need business sense and creativity | Josh Whitfield

All right, what's up, GTM Builders? It's Jared and Matteo again and welcome to GTM Engineer School where we interview GTM engineers to uncover principles, workflows and tools in this quickly emerging space. So today we're hanging out with Josh Whitfield, founder of Content Marketing Media, the only outbound agency certified across clay instantly Maldoso and Octave. Josh is a true GTM engineer creator, sharing tons of his workflows, best practices and agents on LinkedIn and X and everywhere else. So please make sure to give him a follow before getting deep into the go to market space. Josh spent 15 years in insurance where he led agile teams focused on the development of intelligent automation solutions, Think process mining, API integrations and RPA robotic process automation. So he basically worked in AI and automation before. It was cool. So now here we are, we're excited to unpack Josh's unique POV on go to market, engineering outbound and agents. So, Josh, welcome to the pod.

Yeah, thanks. Happy to be here. I mean, you kind of summed up my background pretty well, right? At the end of the day, it's like we were doing boring stuff in R and Python and data science before Claude and OpenAI. It's definitely a different world and it's nice to kind of bring that institutionalized knowledge to be able to run Marino, run the business and just help clients, you know, in innovative ways. So it's an exciting time. Like it's never a boring day in AI for sure.

So you probably get a lot of questions from people like, who are getting into it now and just finding you, like, what is it? So how would you define go to market engineering in your own words?

Yeah, you know, it's such a broad term. Ironically, like Clay even did a really good job yesterday of coming out with new information on how to explain it. So I'll just kind of, you know, a little bit build on, on what they've said. Like I see the post, right, and people are like, oh, you know, Clay created this job title, right? But I, I can kind of go back to the day of like when Salesforce came out, it was perfectly acceptable for Salesforce to have Salesforce engineers, right? Or Google came out and Google had all kinds of job titles they created. And I think it just speaks to the fact that clay is that much of a revolutionary technology where they're like, look, we are going to define a job title that, as I read earlier this morning, I think there's like a hundred postings right now at really big companies. These aren't mom and pop companies posting for GTM engineers. These are companies like OpenAI and Anthropic and really large type organizations. So I think it just speaks to that. And like, for me, you know, I have four, right. Six months ago I had zero, you know, but now I have four GTM engineers. And it's really because it has that much of an impact on the industry where it's evolving right at the end of the day. And it's someone who can kind of keep up with the tech. Someone like, like me that is highly curious about tech, that maybe has disparate skills across a bunch of different disciplines, but they're able to help a company, whether they're, you know, two founders that just started a SaaS yesterday and built, you know, Vibe coded a lovable, you know, app that they deployed and turned it into something marketable, which we see happen every day. Or, you know, there are 50, 100 team SDRs where instead of asking SDRs to spend 80% of their time building lists and doing, you know, things that GTM engineers can do at 10 times the speed, they should be on the phone. Right. So I think it's, it's very applicable across the organization and I think that's really why you're seeing it take off.

So you've had a front row seat, Josh, seeing things really evolve in a way that most people have not. So in your mind, what, what are the biggest differences in this, in these workflows in the last two, three years versus how you were doing things previously?

Yeah, that's a great question. So I was in a very structured, regulated, organized, cloud based infrastructure type approach. Right. So the concept of Vibe coding and Vibe engineering and all that didn't really exist. Right. You know, if one of my engineers went off and did stuff on his laptop with the data that we were using, that's a fireball offense. Right. And you don't really see that mentality anymore. I mean, at the end of the day it's yes, I work with pretty big clients and rightfully so. They need to protect their intellectual property, their data, their customer's data. And that is a bridge that's closing today. But I think really the ability for what we could do a few years ago it was heavily Python, heavily r, heavily Snowflake type databases, SQL. And you needed jobs or people in all of those roles. Right. You need a SQL engineer. Right. You needed somebody who. The data scientists that I worked with and worked on my teams, these were PhD level actuaries. Right. That understood math to the level that I could only prompt GPT to try to understand. Right. And they knew it and they went to school for it and they were highly educated so they could build these predictive regression models using Python and R. And today I can go into ChatGPT and build most of those same models. Right. And I'm not saying that their knowledge, their education is worthless, of course it's valuable. But the difference is the technical capability has just moved forward so fast. And I'm not exaggerating, for a couple $20 subscriptions, the cursor lovable. And a couple other tools and a little bit of effort and a little bit of YouTube watching, you can build some pretty robust stuff that's rooted in the principles of data science. And it's just because they built the behind the scenes technology and capability at light speed over the last year that just creates an enablement that's never existed before.

Fantastic. Well, we're now going to hit you with the lightning round. Are you ready? Do you need a sip of coffee?

Yeah, I'll be ready, but I'll take the sip anyway. Thank you. All right.

All right then I'll just like ask you some questions and in as little time as possible, you just give me your answer and maybe double click on why you picked that answer.

Okay.

So favorite CRM, HubSpot, Salesforce or Atio.

Yeah, it's tough because me and my Clients use both HubSpot and CRM. I use Atio. Bring it on.

So top LLM, Claude or ChatGPT.

That's tough because I use the heck out of both of them. I'd say Claude. Especially Claude. Sonnet 4.0 and Opus.

Yeah, tell us more.

You know, it's, it's a master at copywriting at the end of the day. And they're both pretty equivalent when it comes to capabilities, mcp, stuff like that. But Claude is just, I mean if you saw the emails we sent, that would be mind blowing. You would never know they were written by AI. Love it.

Top enrichment tool.

You know, from an orchestrator perspective. By far Clay from an overall contribute to our outbound motion. Octave. Love it.

And the top GTM engineering tool above all.

Yeah, I mean I think it, I think today it's still. It has to be clay, right? Just the ability to bring any API in the world and into tables and rows and do that up to 72 times per row has yet to be replicated.

Never thought of the 72 times per row, which is like pretty outstanding.

Yeah, you want to make 72 API calls for one row, you can that's some scale.

And what's the most underrated GTM engineering.

Tool in your opinion? Yeah, by far, it's Octave. Like, I cannot believe the world. And obviously they're growing significantly, and I'm really happy for the recent funding, but I feel like you get that tool in the hands of people, you just. You don't need copywriters, you don't need researchers. Right. It by far has radically changed our results and given credibility to professional messaging. And that's what it takes, in my opinion, to succeed it in gtm.

Bring it on. Shout out to our friends at octave for raising 5 million recently. Yeah, shout out. Zach and team. All right, thank you so much. That was fun. Now let's shift a bit gears and I would love to chat with you about system design. So, like, you obviously with your clients, design tens and tens of GTM stacks for album motions and more. Can you walk us through the building blocks of your GTM stack and how you approach designing it?

Yeah, I'd say at a core level. Right. It aligns pretty well with our key partners. Right. So, you know, we onboard a client, we bring them into Octave, we're very transparent of what we do. We get them involved in building out playbooks, right. Whether it be the rules, we're going to target their existing customer references, pain points, value proposition, et cetera. And we really spent a lot of time in the onboarding process explaining why and the value. And then we get them involved. We say, you know, what do you love? What don't you love? What data? You know, can we train this on the gong transcripts of all the calls you close? Can we, you know, use case studies and really explain to them that I to call it the brain of the operation for every client, right? Then we start to integrate Octave definitively into Clay, where we use, you know, no matter what, the original data sources of the lead. And that may be things like Apollo, it may be things like apify, it may be Fire Crawl scraping the web, Panda match for lookalike. Like, whatever the source that my engineers use to capture the original lead data, it still comes into play and still feeds with Octave, right? And then from there we build those campaigns. We use robust email verification via API, and then we push that to instance. So, and that is where our outbound messaging goes. Male does so is the core of our email infrastructure. So that's our general motion Octave clay instantly Meldoso infrastructure. But we're not afraid to bring in a bunch of different tools. Like today we're using Manus we're using model context protocol within Claude. Like I very much believe in the mindset from a DTM engineering perspective. It's my job is to get my clients data that other people can't find and really try to look for signals, intent and really find. Clay calls it, you know, the alpha signal and really find that thing that really defines like this person is telling me they need to get or be involved in the solution that's being offered and that they have the means or will to do so. So like if you look at it really quickly from the perspective of like a realtor, you can look at things like does a realtor did they recently just make a price drop on a house? Which means that they're struggling to sell that house. And then in the same data points you can say have they sold a lot of houses in the last six months? And since they get paid on commission, you can make a pretty good inference that they're making a lot of money. So there you've kind of identified here's a need for this particular solution and they, they have the means by which to engage. And so that's kind of just an example of for every unique client it's finding those signals that then take you well beyond the fact that they happen to be a founder in a given role with that at a particular company. Because guess what, that's what almost everybody does and that's what makes us unique.

Love it. So again if I had to believe back at you, you know like there is Octave which is the brain where you host all the grounding context on product ICP and messaging. Then you enrich that with an orchestration layer that is component with both prospecting data from the likes of Apollo and like list builders and then obviously Clay for where you can import all sorts of enrich enrichment standards and less so where that's where the alpha is, you know, like in the apifies of the world or your creativity to understand based on the ICP which enrichments to find or which signals to find. And then you have this distribution slash infra layer that is male dose of order deliverability as well as you know the sequencer basically like a smart lead or instantly and how, what are, what are the main trade offs that sometimes it happens you have to make when you build these GTM infra.

Yeah, it's a great question. Like from a technology perspective I think when you use a wide variety of vendors, nothing's ever perfect. Right. Just as much as you know, an engineer might forget to put the curly brackets around first name. And so now 9,000 emails go out with a first name instead of hey, they're actually their actual name. Sometimes the tech fails. Yeah, it happens. You know, it's a hard conversation with the client but, but you know, on the other side, sometimes the tech doesn't work. Like cold email is notorious. And it doesn't matter if it's Mildosa or Instantly or Gmail or Microsoft where, you know, it's a hard business to be in because one day you wake up and for three months you can't hit Outlook inboxes, right? And that's, you have to be adaptable. So I mean, I think that's a pitfall with the tech. I think the other thing is it's, I like the analogy I can give. It's, it's like trying to hit a dartboard on the back of a 18 wheeler going 65 miles an hour sometimes. Because that's how fast the tech is changing, right? Like these billion dollar companies, like OpenAI, billion dollars is probably a shortfall. Anthropic, Amazon, like Google, Meta, like all, you know, the social media revolution that happened in the last 10 years, now they're all going through this AI revolution and it's like they release models like they're nothing, like they're candy. And it takes testing and a lot of iteration to understand what impact they can have, but they're so powerful that they, in and of themselves, sometimes they knock out SaaS companies, right? And like just a week or2ago, OpenAI is like, oh, we're just going to let you record every call from now on and store it in OpenAI. And there's like nine sasses I've used I can think about in the last year that are probably thinking, oh God, you know, they have 600 million users. If 1% of them start recording their call, we're in trouble. Right? So that's the, the same impact happens to gtm. But I think, you know, our job is to kind of balance that and then every client's different. You know, some clients have Salesforce, some clients have HubSpot, some clients have big teams, some clients have Salesforce but have done a really bad job of keeping up with the data, right? So there's just so many variables to this and I think it's really having good relationships with your clients, being transparent, being honest of what you know you can do. And, and I have many of my clients that come to me and they're my favorite clients. They're just like, hey, I don't know if this is Possible. But can we try to do this with clay? You know, like, yeah. And some of those have turned in the greatest use case. I had one client just quick story that was like, hey, I found this robotic pen that looks so real, like they hand write with a robotic pen, handwritten notes. And he's like, do you think you can dive into the API and see if we can do it via clay? And I'm like, sure. So we wrote the API calls and everything, got it to work. And he's like, can we connect that to HubSpot? And I will build a workflow that populates clay. And every time somebody gets to where they're almost about to close the deal, can you find the address and automatically send them this hander? And that workflow runs today. Like, if I'd have told you that six months ago, you probably would have thought I was crazy. But it's a real workflow and guess what? It worked like. It's, it works like I love for somebody that I get on a demo call with, you know, two weeks later to send me because they don't know it's written by a robot. It looks real, right? So those are the kind of things that clay and clients bring. And you know, it brings innovation and play in these type of motions specifically, they create enablement and value doesn't always come from meetings booked. You know, if that handwritten note closes a hundred thousand dollar deal, like, we're set for life. Right.

I love that you, like anticipated our next question, basically.

Yeah, exactly.

Sorry, I'll let Jared take this one.

No, that's fantastic. Josh. When you're starting out with a new client and they're really ground zero, what is the first play that you run with them? How do you work with them to figure out what that first play is to get them off the starting line? And if you have a specific company or project where you've onboarded them and you want to use as an example, that'd be great.

Yeah, so it's really changing. In the beginning it was very much like, hey, here's this three page Google Doc. Fill it out and tell us what everything you know. And for us, Octave changed that because we found out like with Octave, we probably know more about their TAM and ICP than they do. And so now it's become very collaborative. Right. Where on the sales calls that I do, you know, I, I demo. Like I have my team go in and build a clay table, build the actual campaign, build Octave playbooks for a demo call. Right. So that's very Intentional, because number one, I want people to show up and like, like these guys are serious. They've done their homework, right? And secondarily I want them to know, like, this is what you're investing, right? You're investing in some of the best tech in the world. And I want you to see it. And we're so serious that we've already built a little bit for you. And so that helps because the transition into onboarding becomes more straightforward, right? Where it's like we now do it all in Slack. We have them fill out the information in Slack as we run campaigns, we update strategies in Slack, we send weekly reports in Slack. So we've kind of built a cohesive un it around communication. But to your question, how do we choose the first play? It's, it's different for every client because sometimes the clients come to us because they want to execute a specific playbook or a specific motion. Other times they have no idea. And we just, we talk through our analysis of what we think is best. And like the most recent, I don't call it a hack, but the most recent thing we've done, which has added superpowers to that, is we go into Manus AI and we do an hour of research and anybody can do this. You just go like, hey, here is my customer's website, or here is this lead's website. Now go build me out a robust playbook on how to get the best results for their ICP. And I'm telling you, you're getting like PhD level research coded with Python with a thousand web. I mean it's unbelievable what you can do with these models. And even if that's all I did, the clients would be like, wow, you know, but it gives us a good reference point and a frame of point and we just look very competent. And it's because we do our homework right. But it's not like we go out and spend 50 hours on the web. We just have good tools and we know the questions to ask using the.

Best tools in the world and what questions to ask. Now what has been the biggest impact for the clients and what's the biggest been the biggest impact for you? Is it conversion rates and how have they changed? Or is it the ability to send fewer emails and get the same result or tell me more.

I think retention is probably the number one impact for us, especially with our clients. Most clients for us in the past have started out with, they come to us and they're like, hey, I saw you're an expert with Insulate, expert at Play. You know, I think I want to execute this motion. I want you to send a lot of emails, right? So many of our clients now, especially the ones that have really kind of stayed on for a long period of time, they care less about the email and they care more about the motions. Right. Of supporting their team. And the retention aspect comes into it is because the tools have just allowed us to be innovative, allowed us to come up with new ways to support them beyond sending emails. And especially if you're a chief revenue officer at a big company or you're a founder of a growing tech company, you want to feel like you have an advocate or somebody that is really on the forefront of AI that's supporting you. Right. And that's the, the, I don't say the vibe, but that I'm very open with my clients. I'm like, if I learn something and I think it's going to help you, I'm going to tell you, you know, I'm going to implement it in your campaigns and I think I would love you guys know Julian at Octave, like, he is a trusted advisor to me. Like, I know a lot about AI. I'm well educated and I've, I've run data science teams, you know. Julian, you work for Google, right? And you know, like, my clients come to me and look to be a supporter and advisor because they understand the significance of AI and the things that are happening in GTM on their business. Same the way I look at Julia, right? You surround yourself with people that you support that can support you and make your knowledge grow. And that's really what we try to provide. It's more than just sending emails or building clay tables or using Octave. It's like, hey, we're going to use our collective knowledge of the industry, our collective knowledge of having 22 SaaS clients at the moment and different innovations and plays and sizes and bring it to all of you in a way that we don't violate client confidentiality and strategic motions. But we also know that this could be applicable and it's probably something we should try for.

And so, yeah, Josh, you talked a little bit about using the right tools and asking the right questions, right. Which from what I've been hearing from talking with experts like yourselves, as well as just like being in the industry, you know, like, it's one of the probably most critical skills that GTM engineering is bringing right into the marketplace or the talent marketplace. Right. Like, you know, prompt engineering. You also, like, mentioned adaptability as one of, like, you know, an important thing. But why don't you tell us more about, in your opinion, what are the really most important skills that aspiring GTM engineers should cultivate?

You know, it's funny because if you'd asked me that six months ago, how to set prompt engineering, today, I would tell you I think it's creativity because, you know, I don't write any of my own prompts anymore. I just ask the models to write the prompts for me. Right. So that's what's changed. Right. It's like you used to have to be really good at prompts. Now you just need to ask the models to write the prompt for you. And especially when they're connected to big databases and stuff like so, you know, I think, I think it's creativity at the end of the day and balancing business understanding with creativity. Right. Because the capability is only growing. If anything, we're kind of in a situation right now. It's which tool do I use? Because there's so many. So, you know, I think it's more of like the, the future, true, impactful GTM engineer has enough institutional knowledge of the business, knows how to go find out and fill in the gaps of what they don't know. And then really, whether it comes from their own brain or their ability to have conversations with these models and learn things, to be creative, it's. It's applying that creativity. You know, Jared asked, like sending smaller campaigns earlier. Yeah. I mean, you know, six months ago it's pull a list, send emails. Today it's build orchestrated signal tables that fire off every time something significant happens. Right. But it takes creativity to not be like everybody else and pull news and funding and job changes. It takes creativity to say, look, I'm going to go out and I'm going to figure out every time someone inserts a geocode radius outside of a conference location in San Jose. I made that up. But I'm sure it's possible. It's, you know, it's that type of creativity that gives strategic advantage to whether they're supporting multiple clients or they work at a company. Right. And it's, it's just that balance of creativity and institutional knowledge. Because I think the tools at some point are going to be. The entry level is going to be so low that it's whoever can orchestrate and come up with the creativity and the unique value points for there. Is it going to make all the difference in the world?

I mean, like, obviously I can't. I definitely agree with you, but how do you define creativity to our audience? Maybe in this context, and how would you nurture or develop creativity in this space?

Yeah, I think I oversimplify sometimes some things, but I would just say like, you know, in the, in the realm of like, how do you know, how do you kind of acquire and get those skills? Is like, you know, for us, we service multiple clients, so I think it's a little bit harder. Like we have to be good at really sourcing information because some of our clients do really complex stuff, right. And there's so many of them. So I've got to learn 25 businesses. But if you're a GTM engineer at one company, right. Like the best example I can give you, and we just did this for ourselves, is like, heck, just train an AI on all this institutional knowledge and build up a brain like we talked about earlier with Octave. Like I built a Delphi app of like 560,000 words that I've said over the last two years that have been recorded, whether it be on social media platforms and stuff. And then I connected it to Slack and I'm on nine calls a day, so when my team can't reach me, they talk to my AI and they get probably better, at least more humorous answers than I would give them. But it, it's practical and it works, right? And it's creative. But, and I posted about it because it's like it says, it definitely has a better sense of humor than me, but that's how you acquire knowledge. Like it doesn't matter what method you go about doing it, but you've got to acquire knowledge, right? Like, who cares that you can code something? Because in reality I can probably find an app that I can prompt to get that same level of coding, if not better. It's more about like companies need engineers that are coming, they're willing to learn, they can gather a lot of information, they can decipher that information down to what's meaningful to revenue at the end of the day and then take that information. You don't have to have all the answers anymore. Like, I don't have the answer. I'm not, you know, I'm not that smart. I just know where to go get the answer, which makes me look smart. Does that make sense?

Absolutely.

How much of your time would you recommend people spend playing with the new tools, exploring so they could be that partner for creative aspect of this so they, they have a wealth of tools to go to.

Yeah, I spend 25% of my existence. I'm running a, I'm running a team of 10 and a decent sized business at this point because like, I'll just leave you at this. When I was 17 years old, I used to work at Denny's and this guy would sit across. I worked in the smoking section where they had the little bar, you know, where four seats are. This guy Tom used to come in there and sit, and I would sit there and refill his coffee and stuff. And he would stay late. I worked the night shift and he started telling me about this thing called the Internet. And he was like, there's going to be this thing that was 30 years ago, right? And he's like, there's going to be this thing that connects to the Internet and all these businesses are going to be building. And I thought he was smoking crack. It sounded like the future because you know, we barely had aol. We were still listening to that stupid dial tone from AOL at that time, right? So what he described to me. And so you know what I decided this time? Like, I'm not going to be that stupid, right? 30 years later when AI came along and it literally, like when you understand that Sam Altman is having conversations with Hollywood producers because some guy in Bangladesh in a year or two is going to be able to prompt a Hollywood level movie and kill a billion dollar industry. And even six months ago, Sam was having conversations. Just look at VO3, right? And what it's doing in the last two or three weeks, right? And when you think about like that to two hours is only processing capacity, right? More Nvidia chips, right? And more storage. That's all it's going to take to get from there to there. AI is going to have such a look at what I mean, it's killing SEO, like, and I'm not an expert at it, but like, you know, it's just dramatically changing search. Like I like to joke that my kids, they're gonna laugh at us when they're like, used to go to this like thing and type in words and look at a bunch of blue check marks. Like it's, it's a totally, it's, it's a totally impactful revolution. So it just kind of gets me back to like if you are starting off, if you're learning, if you're building a business, you know, we all have, I, we have families, you know, I have five kids and you know, a business to run. But I still spend a significant amount of time doing it because it's happening so fast. And the size of this, in my opinion revolution is like 10 times more impactful than the Internet. And you know, I'm Not a political person, but even in the last six weeks, like foreign countries are giving us billions upon billions of dollars in investment in AI. And this is after, you know, Amazon, Google, Meta, Facebook, every. I mean, every one of them are already investing. So, you know, when you take the 10 rich companies in the world and they're all focused on the same thing, you know, that's a clue that you probably should be paying attention to it too, in my perspective.

Great perspective. You've given our listeners already a bunch of, I think, practical tips. Any, any last tip that you would give our listeners that they can implement today.

Yeah, I think the number one thing that anybody can do today is, is just go in there. It's almost like we were just talking about just go in there and do it. You know, I've learned so much from, in the last year or two from just asking AI the questions, right? Like, you know, and it's just. I'll just use OpenAI as an example. Like, you know, a year or two ago, they had maybe 20, 25 million users. Today they have 600, 700 million users. The capabilities of what these models and this tech can do. I truly believe this. You could, this could be the first time you've ever heard the word GTM Engineering. And if you spend enough time even just OpenAI with web search, you can, it can teach you how to build clay tables. It can teach you how, like, you don't need, like, I'm scared for my kids because they all want to use ChatGPT to write their papers, number one, and they're already doing it at half the colleges across the United States. But, like, I wonder what education is going to look like because you can go to this thing on your phone and almost ask it anything. And you know, a year or two ago you're like, is it hallucinating today? That margin is so small. And so just use. ChatGPT is free now just use the free version. And you know, if you want to be a GTM engineer, like, you should go learn formal education. You know, like you guys at school, right? You should learn that you need to have practical skills. You can't actually build a clay table in ChatGPT, but if you want to understand conceptually the skills you need, the things you should be focused on, how to be creative, et cetera, all of that is at your fingertips. And the device that, you know, you carry around and stare at 12 hours a day, so why not, right?

Fantastic advice. Hey, Josh, there's a lot of places that listeners can find you and follow you. Can you name some of them?

Yeah, predominantly LinkedIn is where we, we do. You know, I post a couple times a day, which is, you know, I promise you, I, I don't have a copywriter, I don't have anybody. I wake up every morning at 6 o' clock and I decide what day I'm gonna, what I'm gonna talk about today. LinkedIn is predominant, but we, you know, we're on YouTube, we're on Facebook, we're on Instagram, we're on X. So.

Okay, awesome. Thank you. So Josh showed us how go to market engineering is revolutionizing go to market strategies if not changing the world. From using AI tools like Clay and Octave to creating innovative workflows like robotic handwritten notes that close six figure deals. Josh is proving that modern GTM is about creativity, cutting edge tech, and finding those alpha signals that really transform the outbound motion. It's not just about sending emails anymore. It's about being a strategic partner, creative strategic partner in the AI powered sales landscape. So thank you for sharing both practical tips and sort of big picture concepts, Josh. This was really valuable. I really, really enjoyed this.

My pleasure, guys. Appreciate it. Appreciate you both having me.

Thanks so much, Josh.

All right, stay in touch.

We'll do. Thank you guys.