Building an AI-native sales motion with Heath Barnett

0:00There's a massive shift going on with the whole AI and, you know, should an org be AI native or a product rather or a company? Should you, you know, bolt AI onto your existing product as the go-to-market org? Should you be doing more vibe coding and building stuff? Should you look for solutions that, you know, create a container or bumper guards for you to do this and everywhere in between? And I'm seeing a really big shift with, you know, I've done a lot of vibe coding at Outbound Squad. Just for our delivery team, we could not find a great solution to manage all of our clients and the schedule and our capacity and all this stuff.

0:44We would use a combination of project management tools and Google Sheets. And in three hours, I vibe coded something that we use every single day on the delivery side. And it's been awesome. But I don't think that that is a scalable way for a go-to-market org to do things. You shouldn't be building everything besides your CRM from scratch. Things are moving into more of this headless environment where you can start to do some of that stuff. But I don't think that the majority of orgs will ever be at a point to where that really makes sense, nor will they have the talent to be able to do that. You got to have some extremely capable tech-savvy folks.

1:27The happy kind of medium that I'm seeing is tools coming out that really create a container for this. So an example of that would be instead of you vibe coding and clod and doing a deal room, creating your own deal rooms, let's say, it'd be using a tool that creates the deal room for you. Instead of creating your own proposals, it's a tool that creates the proposals for you and that leverages AI within their container so that you're leveraging this DIY approach that allows you to make things very customized very, very quickly, but you're doing it in a way that's secure. You're not going to break shit and all of those other things.

2:04So where am I going with this? I bring this up because the person that I'm interviewing today, who I will introduce here in a second, is very much on the side of, as a VP of revenue or sales, being very hands-on with this stuff and being able to custom create a lot of these solutions and workflows. And he's going to walk us through how they are doing a PLG motion and be more sales-led throughout that motion, right? Engaging enterprise customers with reps that have signed up to use their solution and engaging some of the sales leaders and doing more of a sales-led motion within their PLG motion.

2:46So before we get to that, though, which I'm really excited about, my name is Jason Bay. I'm CEO at Outbound Squad. We're all about helping you turn complete strangers into paying customers. So the number one thing that we help with is pipe gen. So if you have teams of account executives, for example, that would be how do we make sure that they have the pipeline coverage that they need to hit their targets? It's very focused on top of funnels. So, you know, outbound programs, referral programs, how to secure the cross-sell and upsell conversations. PipeGen is really the big thing that we focus on.

3:21SDR teams, you know, building out that SDR motion, refining it, getting your managers like really tight with their coaching, scaling out that motion. And we're doing a lot of work right now with account management teams and CS teams if they own the upsell or the cross-sell, but again, securing those initial conversations. And we've done it for some really cool companies as big as Shopify, iHeartMedia, Amex, etc., all the way down to your startups and your Series A's that are just getting their first sales team off the ground and shifting away from founder-led sales. So today's conversation is with Heath Barnett.

3:58He was VP of Revenue at MixMax at the time of recording this. We did this in person when I was in Dallas. He is no longer at Mixmax and he's doing GTM consulting. So make sure to connect with Heath. It's Heath Barnett on LinkedIn. Check out some of the stuff that he's up to. But he's going to walk through how he thinks of PLG in terms of how we get context into that user, into that account, common misconceptions with PLG, how to think about the aha moments throughout the user journey that create a natural time for a conversation to take place, how to think about your ICP, how to provide context to the reps that are engaging with these folks, how to be AI native as a go-to-market org.

4:45I mean, just a bunch of really great stuff. This is really a masterclass on that. So before we get to that, two things, make sure to check out our free outbound masterclass. It is studio quality content that we recorded in course firm. There's an interactive workbook. It's free. It's at outbound squad.com slash masterclass, like literally everything that you and your team would need to do tactically to prioritize the right accounts, narrow down your ICP persona based messaging, cold calling, cold email, using LinkedIn, like everything that you need to know to secure that first meeting. outboundsquad.com slash masterclass, go check it out.

5:23And then if you listen to this episode or if you've been a listener for a while and you're like, you know what? I would love the team at Outbound Squad to help us with our pipe gen. Our average client within 90 days increases qualified pipeline by 20 % or more. We just with a recent client worked with around 50, 60 account executives and helped them increase their self-sourced pipeline by 121 % within 90 days. So they went from 70 % of their team, literally having weeks where they put zero on the scoreboard in terms of meeting creation, op creation to nearly a hundred percent of weeks where every single person on the team was putting something on the scoreboard.

6:01So just send me an email directly, Jason at outbound squad.com or connect with me on LinkedIn. Let me know. We'll be happy to chat with you. All right, let's get to the interview.

6:18It's been good to hang out in person. First off, uh, I was trying to think this, the first time we met in person, I think was in Austin at that sales, that top chef dinner that we had. How did you do in that? By the way? Um, I thought I did great because and I say that it wasn't just the dinner. It was like the people that I met. Like that was the first time I met you was the first time I met Scott. Yeah. I met Adam, um, Chris, I'm a sell better show. And like ever since then me and Scott have an amazing relationship. We've built a great friendship. Um, yeah, man, like the, the gifts that I got from that one event has just been continuous kind of pollination.

7:00Yeah. How are your knife skills? I mean, probably not good. Not good. Like realistically. And you know what? I think it's because all your fingers, I still have all my fingers. I'm good. Like, man, I can cut. Um, the secret is, and I know it's just cause I managed a restaurant while I was in college. Okay. The secret is just keeping your knife sharp. Yeah. That's what I'm horrible about. And I'll just like cut harder knowing it's just dull. And I'm like, just go sharpen it, man. Yeah. Um, so I'm super excited to dig in. There's, you've done been working on so much literally in the past couple of weeks, but especially since last year when we met and Mixmax especially has got so many crazy interesting things going on and the thing I really want to kind of pick your brain on and we'll kind of start with a few of the things like in your career like when I think of Heath I think of like like you're the guy that can like be very strategic as a VP but then also get like extremely tactical which I think is a rare skill and then you're also like extremely AI forward you're not just like throwing a couple prompts here into Klon and doing stuff.

8:08You're like doing stuff with agents and like all kinds of crazy shit. And you're like this like signal led product led growth. Like, like you have this specialty with like companies, especially that are in that stage of growth. So give us a little background. Like you worked at Uber, Notch, Airwallex, like kind of similar approaches, like walk us through some of the stuff that you've been up to. Yeah, I think like, you know, the the first one is around this whole ai go to market what does it mean go to market engineer i think like at the end of the day all we're trying to do is just connect signals to action yeah full stop um i think uber drilled that in us so i was actually on my way over here in an uber yeah and me and the guy were talking and we started talking about sales and i was telling how when i had my startup i did i ubered like i would do the morning uh rush hour from like four to seven work on my startup and then i would hit the rush hour afterwards and i say that because it had me reflecting on like building the organization downtown here in dallas and what just like what we were talking about last night and i was like man this isn't new like it's just like now we're actually giving it dedicated focus and so an example at uber was we were a marketplace and the beauty of a marketplace and i almost i always got so much crap for it from sass they're like oh you come from marketplace like do you understand like yeah it's like 10 times harder i'm like yeah it's a lot harder man like i have to balance a three-sided marketplace right yeah um and so to a great example at uber is we ended up we built a heat map when we would launch new markets or if we were trying to like expand in a different category for us a category we like the cuisine this is on the uber eat side yeah and so we're like well we have we can track the data that our customers what are they searching for we can look at like what are they actually like buying from.

10:02And so we're like, oh, simple. Let's just see where the supply gap is and the actual demand is there. And let's just go take that to restaurants and show them like, hey, we can close that gap and bring you more revenue. That was signal led sales. That's how we would go and penetrate markets. After that, Notch, for example, Notch started as a marketplace called Chef Hero. And they were trying to solve the last mile delivery problem for restaurants because in this day and age still, you have a delivery truck that shows up and it's like, oh, hey, by the way, we didn't have tomatoes. And you're trying to set up for the day.

10:38So what happens is restaurateurs or chefs, they go to the local store, go buy tomatoes, but it just throws it all off. COVID obviously threw that completely on its head. And so a new CEO came in and I came in to help them. And we decided to actually turn around and we're like, the real problem is around the invoicing process and the inventory process that was still very manual they were still cutting manual checks and so that was a great example where it's like why can't just go sell the suppliers i can't go to benny keith and be like hey let us actually be like that marketplace for you they're like no we have the market on lock why would we do that yeah so we had to again generate the demand and so we're like okay cool what do line cooks need to do they're still using a piece of paper to take inventory of like their line, which is inventory in the, in the, the freezer.

11:27I was like, okay, put that in app. Let's just give a line cook an easy way to where they can actually just do digital line check and inventory management. And if I see that they're using it, I can then go talk to the head chef that gets me a door open. If I can then go close a management company that has 10 restaurants. Well, those 10 restaurants combined have buying power now for Benny Keith to get me a seat at the table. Yeah. Right. So we just had to inverse, like, how do we actually go to market? How do we go and do this again? Like that wasn't, that wasn't a thing yet. Um, air wallets is the next one is I took air wallets into North America and you have a company who is their whole job is on global transactions, how to make that easier.

12:10There was no easy way then for me to get, say, I want to go find every U S based company that's doing international transactions or has international employees, like we had to go and actually buy the data. We would go and actually buy a distribution, wholesale trading data. We would then go buy data of like these different cohorts. And luckily, you know, well-funded$8 billion value company. We had a data science team, but that's basically what go-to-market engineering is, is like, we had to actually take that data and be able to figure out what's that data telling of us the beauty of ai is the stuff that used to take me six months just like we're talking about i can do in two weeks now yeah right and like that's just how this has just unlocked so much yeah it was a long answer so there's no i love the context there's kind of an interesting so the frame here in which you're thinking about it is like you're thinking about this like economics or a a signal is demand supply is your solution or whatever it is that's going to like meet that demand and that's really like signals and triggers and like all of this other shit that like is so popularized and has all of this marketing lingo now it's really about like helping that execution layer between the solution and uh the demand right there so it's kind of an interesting way to think about it when you look at it that way so put it in an even simpler way right um commerce has been doing this forever yeah right so if we just think about the way that we shop when i go in and let's say like i'm shopping online right i go and i click on a product by a pair like a jordan's or yeah by a pair of jordan's um i go in i look at 10 different ones that's tracked i then go in and i go to my cart, but I abandoned my cart.

14:00Like I obviously get an email from that. Like, that's all it is, is like, you're just meeting where the customer is at in their journey. We just haven't done a good job of that. And I continue to say this as I'm like, SAS is entering the world of Pepsi versus Cola. You can have a thousand different software products, but now it's about how do you actually impact somebody's life? That's what a Coke commercial is about. We have to start thinking that way. And so to your point, a signal isolated, it still really isn't telling me where's the demand or right. Yeah. What we're talking about. But if I use the shopping cart, if I'm able to know that you like Jordans, that's a signal.

14:39But then if I'm able to know that, let's just say I can't access your private information, but I'm able to understand where you live. Yeah. And that's probably a high income area. So I now validate, you probably have the discretionary funds to buy those Jordans. When I combine all those things together to create the context, now I'm like, that's a customer in demand. Let me go and hit them with the right information to show them, I understand your problem. Here's how my company is actually uniquely positioned to solve that problem. And it sounds like there's a couple of really big key components.

15:14This one, if you're able to do PLG. And there's a lot of companies that have the ability to do quote unquote PLG, but are not doing a lot of the stuff that we're going to talk about in terms of like, yeah, just because someone is in your platform using your product doesn't make them a good fit. There are tons of people paying you money that should purely be like, you know, was it self demand? Yeah. So there's that piece. And there's other folks, I think, if you sell a solution that doesn't have plg it's how are people interacting with all of your marketing content like that is another really big thing and for whatever reason there just hasn't been a lot if those two things have been so siloed so i agree and i think it's important to also separate what we mean when we talk about plg versus product led cells yeah right because plg just in the most eli five I've explained it like in five terms is if a user can buy your product and grow in your product without needing to talk to anybody, that's PLG.

16:19That's product-led growth, right? I'm all about that because I'm like, if you can continue to take care of a customer, give them service, and they just naturally grow in your product, why would I want to touch them? Yeah. Self-serve was the term I was looking for before. Self-serve. Yeah, self-serve. And so that's the problem, though, where most companies get product-led sales wrong is they just start going after self-serve customers. Where product-led sales is, it's identifying there are users who are part of an organization that that user can't go and put a$10 ,000,$20 ,000,$30 ,000 contract on their credit card.

16:56Yeah. Right? That's product-led sales where I can go, oh, somebody just signed up from Cursor. Yeah. right it's an AE it's an SDR I'm okay with that I want to now understand what problem is that SDR trying to solve and then how can I extract that problem into a strategic problem so I can now actually go to that VP and say hey I think this is a problem you're probably dealing with because you have reps that are using our tool now tell me more about your organization right who else is potentially impacting like that's where I consider that still warm based outbound is like the product warmed up the account by just getting one person to use our solution.

17:36Yeah. And I think it's important because I believe you can still have a product led sales motion without being PLG. Product led sales at the end of the day is just taking how your customers are using your product to understand where are they not getting value, where are they getting value, and then how can we close that gap? I mean, that should be most of your expansion plays. Exactly. Right? That's really what Stripe does an amazing job. There are companies who have been world-class. Most of them, they call them growth AEs, but that's it. They're just like, we're going to go, we have a great demand gen.

18:12We're going to go acquire everybody, just sign up, and then we're going to be really good at looking at the data to understand who actually should we have that rep to try and go have a conversation with. And in my experience, there's more companies that don't execute this well or just and it's not a knock on them i think this is an incredibly complex thing to do um especially prior to the advent of how a you know yes like you're leveraging ai yeah so let's get into this dude so we were talking last night and and i want to add one more you hit a good one about the content the marketing side the customer journey because i think plg are not.

18:49That is a gap and it's critical in the close. And what I say by that is to me, the journey, and this is where signals product led aside, but when I can, let me give you a real concrete example. Somebody just raised funding is usually why we reach out. Right. The moment company raises fun. Like I hate my inbox. Cause I'm just like, I know I'm about to just get a million of congrats on the raise. Congratulations. Buy my product. But when you take the funding and I know that you have a sales organization. A great example for Mixmax is we know that there's a high correlation that if they have a contact sales call to action button on their website, they have a B2B sales motion.

19:28Easy to pull, right? So I say that because when I take someone and I understand that they just raised capital and I have an understanding of their business motion, and then I also maybe know that they're hiring for three reps right now. Oh, and they're also hiring for rev ops. Just go with me. What does that tell you? You're about to scale. You're hiring for it. You don't have the infrastructure yet because you're still hiring for a rev ops. Yeah. Right. Like that now tells me a story that I can at least come into you with a hypothesis. Yeah. Now where I need to close that gap is everything I relate to or link to in an article has to be related to that journey.

20:06Yeah. And then the follow up, the retargeting that we do for marketing has to be related to that journey. Yeah. The problem is today is we just say, hey, you raised money. Let me throw you some random article with a bunch of random data. And then let marketing send you a sign up now type of email. That's what actually happens today. PLG or not, that has to change to where we better actually like. Or the SDR calls and says, I saw you guys raised a bunch of money. Do you want to chat?

20:34So it's this. Again, the word sophisticated comes to mind for me. This is a very sophisticated go-to-market motion. that I think is like companies are now like I think the like when I heard about stuff like this like seven eight years ago and people were talking about a you know account based like ABX I think is what they called it right and they were detailing like how they would string it together it's very manual dude it was extremely manual I think these really large organizations were able to pull it off but we're at the point right now where you can pull off pull off a pretty sophisticated motion like this with like, and be really lean actually.

21:14So let's dig into that. So there's kind of a six step process that you've outlined when building an AI native product led sales motion. Let's talk first before we dig into that. When you say AI native, what does that, what does that mean to you? That's a great question because we throw it around a lot. But to me, AI native is not about a prompt. It's not about deploying an agent. It's about building an ecosystem to where like whatever you do deploy, it continues to learn from itself and get better. So I'll give you a great example is any agents that I have deployed, even if it's through co-work right now, meaning device dependent, I have a QA agent that sends me a weekly summary every week.

22:01Yeah. And what it basically does is it goes and analyzes what all the other agents did. What feedback did they get? What were other reps asking for? And it will make suggestions on like, hey, we should go and update this skill on this agent. And all I have to do now is just say approve. And so that to me is an AI native go to market to where it's how am I not just ingraining AI through our tech stack, but how am I also making sure that like it continues to get better and smarter. Yeah. And it's almost like you're building it instead of this like AI assisted human led, you know, kind of thing. It's really AI first, right?

22:40So let's talk about the six different steps and just to kind of go high level, do you want to start high level what those are? And then let's dig into each of those. Yep. So if we go over the six, you have like, especially for product led sales today, right? It's first and foremost, no matter what you do is understand your icp yeah right and so we're laughing about this because i told you like a couple weeks ago we were talking i was like i'm about to just get on the phone and start just hitting up our product leads to learn more and learn better and start up that dialer dude and just the dialer i did i mean i had like power dialers ready like let's go yeah um but i paused for a second because i was like well wait like how much can i actually learn today that i didn't know yesterday.

23:22And I think that's where the power of AI has really helped was I was able to actually, like, we've always had a good understanding of our ICP. One of the biggest challenges when it comes to product led of like a self-serve motion versus sales led is it's the common frictions. Do we have two different ICPs? Yeah. It's the dividing line between those two. Yeah. Because I would imagine for Mixmax, it's a very different. It's actually not. Oh, really? So what is really interesting is when you dig into it, it is just the size of the company. Um, and I say that because what the data shows us, yeah, we have verticals that we're, we're focused in right now and stuff like that.

24:00But when I remove the verticals, I'm like, well, what are they using us for? Um, so one is validating is like, yep, ICP is like this customer profile. So like, basically like if it's less than a hundred person company, like it's very much a self-serve motion, And like, that's what the data has shown us. Yeah. So basically the easiest way for us is self-serve, regardless if it's a solopreneur or an AE, they are the decision maker to actually solve the problem that they're hiring us for. Yeah. When they come sales led, it means they're making a decision on behalf of a larger pool of people. That's really the main difference.

24:33And so that's just the size of the organization. I mean, we stress tested this to where I even extracted from Amplitude is like, take the domains and regardless if we have them classified as an AE or SDR, if they self-classified, I want to go find what vertical are they from. And so that's an interesting one is where known RICP is. We have real estate companies, insurance companies, but when I actually look at the way they use our product across the board, direct sell, self-serve, non-SaaS vertical, the way they use our product is the exact same way. What does that tell me? They're hiring us for the same exact job to be done.

25:09Yeah. Right. So once I understand that, then I can actually go and understand, well, what are they hiring us for? And I say hiring, cause like as a software company, you're being hired to actually solve a problem. And so that's the next piece. I'm like, well, what are they trying to hire us for? That's the core understanding of your ICP of who are they? And like, what are the questions that they're asking the LLM today? That's what I ask myself. What are customers who are looking for a solution like Mixmax? What are they typing in ChatGPT or in Google search or whatever it might be, right? That's the first most important one.

25:43Because if you don't understand your customer, you're going to go in action on the wrong data, the wrong customers. Like everything is based off of really understand your ICP. Yeah. Let's actually just while we're bringing it in, let's just dig into each of these. So what are, so if we pull back from Mixmax for a second, what are some of the common mistakes that sales orgs are making when it comes to ICP identification? Because oftentimes it's, God, I just want to say this is such a straightforward thing to do, but it's like most companies aren't running. And it's so easy to do now. Just a simple closed one and closed lost report of the last 12 months of deals.

26:25and just looking at basic, basic patterns and the size of these accounts, industries to your point, the products that were sold to these folks, what are some of the mistakes that you see? And is there anything that's maybe not so obvious to look for also when it comes to patterns and things like that? So I pulled this up because this is literally like our ICP profile. Yeah. Okay. So where most get, I think they mess it up. you can ask most people what's your icp they're gonna go oh like they're software companies um they've raised about 10 million or more dollars and they're less than 100 million in revenue um and they're probably a 200 person organization yeah it's still pretty vague super even though it's like firmographic data yeah right it's it's like it's more specific than some people will say when they say anyone can use our solution but there's like even from a firmographic standpoint there's b2b versus b2c there's the industry vertical that they serve there's their buyer there's the size of certain departments maybe there's um do they sell multi-product multi-persona stuff i mean there's so much nuance in there as you know just even on the firmographic side so that's your first layer yeah right so think about it um i learned this uh when i was in go to market engineering school from um what's his name over at clay but he explained to the best he was like you're taking a piece of stone and you're trying to create a statue out of it yeah that's how you are trying to figure out your icp so the firmographic that's just like taking a square block and like making it into a human figure yeah right then from there this is a jobs to be done right is okay so if that's the company profile the next question i'm asking is well if the vp of revenue or VP of sales is like my persona, what type of environment are they living in if they're in that type of company?

28:27Right. What do you mean by that? Okay. So let's say if like my sweet spot is like, they just raised, uh, they've raised 10 million or more in their last round and they're under a hundred million in revenue. Um, if that is my ICP, I should be able to know that they're scaling cause they just raised money. They're probably super lean and efficient because like They're not a big revenue generating company. I can compare that by their head count. So that VP of sales is probably being told what I've been told. You have a quarter of the budget and you need a double revenue next year. I know you need five roles, but you only get two.

29:02Go figure it out. Yeah. Right? And so when I go and I take that, that VP of sales in that environment, the next question I'm asking, what is he or she searching at night while they're laying in bed? Yeah. What are the problems they're trying to solve? So if I'm understanding, so ICP is firmographic. Is it psychographics? Is that what you've described next? So there's like the buyer and their mindset and their state. And like that is such a nuanced thing that you would not get from LinkedIn sales navigator analysis of your ICP. This is like you're pulling in call recordings and transcripts. Like you're painting a picture and a story, telling a story, right?

29:46Yep. And that last layer, what was the last layer that you just said? The last is what are they searching? So that's jobs to be done. Yeah. So what are they searching for help with, like the jobs to be done piece? This is this whole – and we're just talking about the ICP. But it's the most – it is the fundamental piece to get right. Yes. And too often we don't give it enough attention or time. Yeah. We just blow by it. And then we go and we do all this like lead scoring and enriching and we wonder why we're still not hitting numbers. And it's just based on firmographics and job title basically.

30:23Yeah. We like to show ourselves positive signs of like, ooh, like I have 5 ,000 customers in my book that are all ICP. Yes. It's not about that. They're ICP, but how do you prioritize that ICP? Yeah. Right? So that's the goal of the ICP pillar is prioritization. Yeah. And we talked yesterday, I love to call it disqualification. Like, how are we going to disqualify 80%, 90 % of the quote-unquote accounts that we're working? Anything else we need to know about ICP before we move to the next one? No, but I'm going to give you an example of like, here's what I, when you ask me what my ICP is, okay?

31:06Let's just go to the sales leader. I have it separated by a direct sales and self-serve world. So I go, sales leader, the sales leader is measured on a number that depends entirely on other people doing the follow-up. Their job is to make selling repeatable. The problem in the room for a direct sales VP, it's Thursday before the forecast call. They're staring at a number they can't fully defend, not because pipeline's thin, but because half it lives in the heads of 15 reps, and the other half are in a CRM that's three days stale. They don't have a pipeline problem. They have a did you follow-up actually problem.

31:38somewhere in that spreadsheet is a deal rep swears as warm blah blah how they solve it today and then how mix max wins yeah so i just gave my rep not just you sell to a sales leader i'm like this is the type of sales leader that you're selling into yeah right here's what they care about this is like when you can describe in the language that you did there as a seller i mean that's like immediate credibility too that's what separates you from just selling another product to an actual like strategic thought partner. So, and just to be clear, what you're sharing here is this is org provided insight to the rep.

32:17I am as a rep, not having to go find all of this stuff. Like you're of the theory of this needs to be centralized in how we do this. And like, we want a lot of control over the inputs and like all of this stuff we're pulling and we should be surfacing these insights for the rep so that they don't have to. I think we should be surfacing those insights, but even more is turn those insights into a narrative. Because even that, just the example I gave, if I throw a rep like, oh, they're hiring two sales roles and they have 180 salespeople at a 500-person company. It's all facts and figures. I'm just throwing spaghetti at you, and I'm like, now go sell.

32:53yeah but when and that's the beauty of where i see ai is really good is when i can provide all these data points and i can say this is the type of output i want you to provide to reps i want you to take this data and turn it into a narrative and tell them what action to take yes to your point my team rev ops like their job is to build the lead routing the enriching the scoring and the prioritization all my rep does is they show up and they say number one's the most important one for me to hit let me click this button oh here's my brief here's my context boom ship that in a sequence love it so product is next product is next so what do we want to think about with product the only difference between product we don't have to go deep into it yeah um it's the same exact thing it's just like do you what are the signals your product tells you all that all the the thing between that is that's data mapping um is actually tracking the event when you say yeah so the events so these are the moments in which in a customer experience is like delight in your product?

33:51Is that what it is? What we call is aha moments. Aha moments. Yeah. Okay. So for example, we know that, um, our sequencing tool is like the number one lever. Like if someone uses our sequencing tool, our conversion is ridiculous. So I'm like, okay, first aha moment, right. Did they actually sign up and onboard? Did they use our sequencing tool? Because if they're not even getting value from the product, why am I going to tell a rep to like, go talk to them? Like we're just trying to reps into support. And I've done that before. I've made that mistake yeah um and so for me it's about the event so a great example is in amplitude because there's so much data where it gets down to the point where it's like oh they opened the meeting summary but did they actually like action on the meeting summary like there's 10 different fields about a meeting summary yeah this is the other thing with ai is like i used to have to go to engineering and say hey can you take these 10 fields and group it and create one new event field for me and engineering was always like i don't have time yeah now if ai i can build and i built this is an i call it uh it's our amplitude mapping yeah so it's like take all these events and then if these four are met we're going to call this this action yeah so i'm creating i'm turning it into like rep facing language but again it's not taking external signals it's me understanding my product what actually works my product what drives value now let me actually map that out to where sales comes in at the right point.

35:15And I want to say this out loud because it's important. This does not just impact sales. When you understand your user journey and have those aha moments, this impacts PLG because you can now place paywalls in a better way. This also impacts marketing because you can now do in-app campaigns when a customer is actually ready for it. Nine times out of 10, we don't. We send them an in-app campaign and it's like they haven't even set up their first sequence. Yeah. Right? So it impacts way more than just like how does sell go and sell. Yeah. I love that, man. So with product, we want to make sure that we know the aha moments so that we're engineering our sales motion around that.

35:54The next part is signals. So this is like the topic du jour, if that's the right. Yeah. So the signals, all the signals are, when I talk about signals, um the signals are the inputs that creates the actual context which is the output yeah okay um and so for me for example is with our product side one is a great one is like we track it to where i know if they're using our core product so a great one because we just pushed out a lot of ai different tools and there are some of our plans that they don't get access to it yeah um But we obviously like PLG, we ran a trial, we let people use it. And so we tracked all that.

36:39And so now it is like, hey, let me go and find people who are using our core product to our ICP who actually trialed this AI, but they're not using it now because they didn't buy it. Yeah. If they're a self-serve, I'm like PLG team, like let's go and try and convert them, do in-app, whatever. But if they're part of a large organization who's ICP, that's a great opportunity for my rep to step in and handhold it. let that rep be set up for success and then be like oh by the way who manages your sales force yeah oh tanya does is there can you introduce me to tanya i just opened the door yeah so with signals this is both internal and external signals and the idea here again is that i'm spoon feeding the rep the objective things that are happening i'm also giving them the subjective maybe it's not subjective is maybe not the right word, but the story and the narrative around those things that are happening so that I'm just envisioning like, yeah, I'm a Mixmax AE.

37:41I go to reach out to ABC company and I'm like, cool. I already know what they're using, where they're getting value from. I know that they're ICP fit. I have a really good idea of the people that I should be talking to. And I'm like, just freaking hand it to me on a silver platter right there. So the thing about it, obviously we're a sales engagement platform. So CRM matters. email provider matters so when we think about the signals if I bring back full circle I understand my ICP now right I understand what they're trying to find what problems are searching for and I also understand what that looks like so outside of what we talked about that ICP they have a b2b sales motion yep so now my goal is how do I go find all the right partners yeah data providers that can help me actually answer those yeah before i even have to like score it to give to my rep yeah so for example um we use deep line which has just been amazing because they man they're democratizing the data world right now and it's wild yeah what's deep line do so deep line is a lot of things but basically like our entire snowflake crm everything uh is connected into DeepLine.

38:55So we can pull all of our product data, everything. But then we also are integrated out of the box with over like 60 different data providers, different email sequencing tools, our own sequencing tool. And then they built it in a way to where I can create workflows where I can say like, for example, that's what I'm doing right now with the scoring model now is like there's a workflow where lead comes in, it goes to DeepLine. DeepLine then sends it out to all the providers I use. so for example i use full and rich i use fire crawl um who was the other ones uh their stack to look at their tech stack built with to look at their tech stack uh full and rich to do the entire buying committee so like i used to have to go and get plans with each of these individual ones i have access to all that out of the box it's kind of like the clay for this it's it is clay i would say it's clay on steroids yeah clay on steroids but like they built in a very like uh easy way like i know kyle over at owner uses deep line yeah we started working with them before they even became deep line yeah um but i say that because like that's an example where regardless of who you're using is when you know what signals actually matter yeah then go find the providers i think we actually do it the other way right now is we just go find a bunch of providers and we just go pull a bunch of data and it's you've used the word sophistication and i'm gonna say it's the word of intention.

40:20Yeah. It's really not a sophisticated. It's just being intentional because when you're intentional and go back to the ICP, there's usually three core jobs that you're hired to do every sales, every software. You're usually hired for like core three core things to do. Know those three core things, build your entire funnel, your demand gen, everything around those three things. Yeah. It's just being intentional. Yeah. It's really this first principles thinking because it's the same mistake people make with AI. They start with, what is all the shit I could possibly do with Claude versus what am I trying to accomplish and how can this tool help me?

40:55So same kind of thing here. So, dude, this is crazy. So I think also another thing to preface is like you've Vibacoded most of this, if not all of it. All of it. So, and then you're working with these APIs of these tools so that you're able to set up like the workflow and stuff. So we have ICP product signals. Now we're looking at scoring, which, again, this is something that I just have not seen with any level of sophistication or intention. At most of the clients that we work with, outside of RevOps has scored tier one, tier two, tier three, purely based on firmographic data. And usually it's not correct.

41:44So from a scoring standpoint, how are you approaching scoring? You showed me some stuff on a spreadsheet that was really badass too. But like, yeah, how do you think about scoring and like, what is, what does good look like from an account scoring perspective? Man, it's, it is one of those like scoring is a thing that used to take six months. It was really, really hard. And so DeepLine is a great example. DeepLine actually provides like kind of a local database. so like they will actually take like a snapshot of our crm and then that's our playground and so what i was able to do it's like a sandbox or something for you to mess around so what i was able to do was actually go and then be like hey so i built like a v1 right yeah and then i basically told claude i was like i want you to find every reason why this doesn't work so keep in mind this is the power of AI.

42:38I've built my ICP. I use that in Claude. So I have a whole context now on that. I built the whole context of like what they're looking at us for. I've built now the whole context of the signals that matter. So it's not that these are now isolated. All this is now already feeding into this next conversation of like, now we need to figure out how to score this yeah and that's where what what we did is we took a bunch of historical closed ones historical closed losses to create our baseline and then we basically just stress tested it so like on the self-serve i took like 6 000 domains from signups and i was just like tell me about these domains how do we score them um so yeah it was a lot of trial and error i think like even in two weeks i'm probably on like v8 of the scoring model and it'll continue to get better and fine-tune beauty of it.

43:30So are you using the same criteria then for the ICP? Is that the criteria for scoring is the firmographic, psychographic, and then the, I keep forgetting the third layer. So the, uh, the jobs to be done. Jobs to be done. So the buy, I actually separate ICP versus buying signals. Okay. Um, so I basically have it to where it's like 30 % of our composite score is based off of ICP. 40 % is based off of product usage. And then the other 35 % is based off of signals. Oh, duh. Yeah. You would have product usage in there. Yeah. And so like our product is always going to be like crop because like they're using your product.

44:04Yeah. And so when I look at just the ICP score and I think that's the problem, too, is we tend to build a score that's like all encompassing. Yeah. For me, it's like think of it as like with AI. Think of scoring the same way as you have these like sub agents that go and do things. You should have a score that is like this score is to determine are you ICP or not? then this score is determined, are you using our product or not? Then this score is determining, are you actually in the buying window, which is the combination of fundraising and hiring and job posting and whatever else that might be.

44:39Then when I combine all three of those, that's the composite score, which is now basically filtering out not just are they SCP, but here's the priority list of like, this is why this is a strike it while the iron's hot. This is why this is a hot, a warm. This is why this is a nurture. yeah fuck man okay this is so badass um so the next layer is context context so what we've all talked about right so if i piecemeal today again icp you understand your personas you now understand what data you actually need to be able to enrich that account yeah so think about it this way is an account just steps into your CRM.

45:24Yeah. Right? So now what we're doing is we have a model that says, take that account and go enrich it with all this information. Yeah. Because I know all of that data, when I combine it, will tell me the right story. How fast does that happen? If I'm doing bulk, I mean, so for example, I did a bulk run last night of like 500 accounts. It took like two hours. Yeah. um but it happens pretty quickly i would imagine like when a lead comes inbound exactly and that's the thing is like it that was me doing cohorts this is now built to where as it comes in and gets stamped it then automatically gets enriched damn right so take that blank canvas you enrich it with all this data that's now in salesforce now you're trying to figure out well how do i actually like know if this one is more important over the other account that just got enriched that's what the scoring model now is going to do is it's going to say hey i have all this data i'm now going to score it to prioritize it yeah the context is just regardless of the score it's just taking let me actually understand now what is this telling me okay so i have an example that i'll read to you where i thought it was a funny one um so for example like with amplitude we have our raw events where it's like sequences smart send ai sequence builder like all of these things but when i actually group that to understand sequencing is part of building pipeline and then we have specific features that are for like helping prevent pipeline from going cold yeah now when i actually bucket it that way it's like oh this person's building like really using you for building pipeline but they're actually not using these tools that are going to help them on like preventing pipeline from going cold, right?

47:09It's our AI action items and stuff like that. That's the context. A rep doesn't, if, and what we usually do is we just say like, oh, this is used, this is used, this is not, this is not. And a rep's like, what does that tell me? When I can take that and create a narrative for the rep where I just tell a rep or the context when they get their score and we have a context field and it says, hey, they're using engagement co-pilot. They're only using two of these AI features. Here's how you should reach out to them. That's all a rep really wants. Yeah, because it's like, if you think about identifying white space, quote unquote, that is something that a lot of sales leaders will talk about with these expansion opportunities, right?

47:51It's one thing to identify white space and say, we have five things and they're using three of them and here are the two, go sell them, versus you're doing, we call it eating complexity. The complexity for the rep is all of the steps that they have to take between knowing what they should do and actually doing it. Like that's essentially what you're like significantly cutting down. On top of the fact that there's a massive amount of intelligence that you're adding to it as well. But yeah, that context and that story, that's really what I want as a rep. That's the skill gap. I think it's a big part of the skill gap that's missing from a lot of the organizations is that context, I guarantee you there's patterns.

48:35There's got to be patterns in that context. There is similar stories that we're going to tell across our ICP because these are repeatable patterns that we're finding. That's the thing that you want to train your sales org on is those repeatable patterns and how to take action on them. Yep. And you hit a really important one because all of this that we're talking about, it started at the brief. Yeah. I sat down with one of my reps and I was just like, what's eating up a lot of your time still? And they just walked me through like, well, when I get a lead, I go to LinkedIn sales nav and I do this.

49:11I then go here and I check this. I then go here and I check this. And I was just like, man, you're wasting like 15 minutes just to get context. And I was like, let me go solve that for you. That was the first solution. And then I was like, well, how can I solve this in a scalable way? Well, in order for me to solve this in a scalable way, I really need to make sure we're not wasting tokens and just doing this on every lead. Like, that's how all this came together. It started with the brief, though. And I think that's where we're missing for most of it is we start at the top of this big, like, what's going to be cool for us to present to everybody and show everybody?

49:46It's like bottoms up thinking. It's bottoms up. Yeah. It's like, let me take the reps problem. Let me solve that. Then let me figure out how do I solve that in a scalable way? Yeah. The other hidden cost I'm curious your thoughts on is like, because you're like, yeah, that burns up 15 minutes. The cost also is this like, hey, in a given day, man, you only have so much like cognitive power to be able to like exert into a high level task. And for most of us, it's like we have maybe three or four hours of really high level work that we can do over the course of a day. And when we're eating into that with someone that's extremely skilled and having them do mundane work that could easily be automated, we're eating into that.

50:28How do you think about that? Because that's like the hidden cost, I feel like, of a lot of manual work. Yeah, I think that we don't give enough attention. I think like we talked about this last night, right? Like the data point, it's always been there and it hasn't changed somehow of like around rep productivity and stuff like that. Um, I think that I actually believe that hidden costs is going to become more obvious now because we're able to, I mean, we even had a meeting about this yesterday at Mixmax. Like our whole mentality is like move faster. Like how do we actually do this? And I say that because the way you move faster is not actually moving faster.

51:08It's just being way more effective and efficient with your time. Yeah. and so it's less about let me find the the frat like the points that are actually hard for reps but just like with the brief and you're talking about it's going to get better right about being ai native think about our agent right now has generated maybe a couple hundred briefs yeah imagine once it generates a couple thousand over the next year yeah and it's learning it's learning every week it's just going to get richer and richer at making sure it gives reps the right feedback you guys are doing a lot of what like you guys and like you know kyle at owner talks about it a lot this is the um uh god what's the name of the company i'm forgetting jack dorsey he's the ceo of square square yeah he's got a couple yeah well one of them he's not

52:04So he at Square, they introduced this kind of like context layer, like they're really kind of popularized. I think this whole idea of like the org chart is not going to be this thing that's like, you know, really expensive at the top. Really like a lot of ICs and managers at the bottom. And the whole org chart is about like, how do we push information up to people so that they can make better decisions? It's like that gap has significantly closed with AI where, you know, we can get the information almost instantly. And he's talked about this not like flattening the org structure. It's more like building the org structure around the data and that intelligence layer.

52:41And this is really what you're talking about is like you're going to have this intelligence layer in the middle of the organization that everything feeds into. And then it works back on all of these little flywheels where it just gets smarter and smarter and smarter. And it's really badass, man. That's like really true AI when you think about it is when it's, it's not just this automation that's happening. Let's talk about action. And then I want to save some time. I got some other stuff I want to jam with you on. So the next part's action. So on the action side, I mean, it's plain and simple.

53:15It's like, this is not a set it and forget it. Yeah. Right. Like this is about learning and making it better. Usually any score model I've ever built is like six months of fine tuning. I think it's going to be a lot faster with AI just because of the QA. But I think the most important part around the action is, again, it's the intentionality. For example, just like today, my head of RevOps was like, are we going to go rescore our entire CRM? It's like, no. Not the entire CRM. We don't know yet, right? So it's like, but we're going to do campaigns and get feedback from the reps and fine-tune and tweak it and make sure it works.

53:54So the real, in my view, the real work is just not getting started on it. Yeah. Right. So I'm curious, related to action, your perspective on this. There's a Derek Sivers quote that I really love. And it's, if information were the answer, everyone would have six-pack abs and they would be a billionaire. So if I was playing devil's advocate with everything that you've talked about right now, information, awesome. We have more information, more context, all of this stuff at our fingertips. But this quote is essentially the idea is that the information in and of itself is not enough. It's the execution, obviously, and that's this last step, the action.

54:33So once we have all this information, again, if we're playing devil's advocate and if someone were to say, well, lack of great information is not actually the problem, what would you say to that? So I agree. When I talk about action, and I'm glad you challenged this because there's a couple of ways of making sure we actually do the action. I agree. me giving a brief and I've learned this right me giving a brief to my rep like has no way of actually like knowing like did they actually action on it was it actually valuable for them or did they just read it and go about their day yeah um so there's two things one is so we use octave octave is another powerful tool thing about octave is like if you're a coder and listening like github for your icp right github is like your major repo for any product that's what octave is it's the GitHub for all of your ICP knowledge.

55:26So it hooks up to everything, Gong, CRM, MixMax, all of it. And all of our playbooks are in there. And it's constantly learning and listening to our calls and making updates of like, oh, your actual RevOps persona is this. So I say that because the way that our briefs are wired up is it actually already says like, oh, this is the Octave agent ID or playbook that you should use for it. So what that does is two things. One is if it is low tier, meaning it should not be cells led. So there's cells led, cells assisted, and no touch. So like if it's either cells assisted or no touch, well, great. I actually now can just tell Octave to go and execute that sequence.

56:09Right? So once those are all done, I can now through Claude just be like, hey, so every tier four that matches this Octave playbook, I want you to go ahead and enroll it in a sequence. So are you saying part of the answer then is to just not have the rep you? some of this? It's about prioritizing because on the rep side in their brief, I actually have two buttons. One is context button and what it basically does is it generates a prompt of all that information that they can just copy and paste into Cloud and it generates the sequence. We haven't launched it yet, but we're testing internally. It's like our MCP, you can actually now push and launch sequences through the MCP.

56:45So that's how I've solved that with my rep. I gave you the prompt to just copy and paste and you can execute. yeah um i also have a button that we're testing now that's just launching mix max yeah and what it does is it takes all that context um again remember this has all been enriched with full and rich so we have the most recent emails the buyers so like i can now actually automate the entire sequence enrollment process right that's my goal for the rep is to actually be like hey we actually already went and enrolled this in this sequence like task is due tomorrow or today like whatever that might be yeah so where do you feel like AI and this approach fits into the execution gap on a skill level so that's when I pick up the phone to make a cold call that's my ability to execute that call that's when I am running an intro call my ability to like execute on all of the stuff that you're talking about I won't solve for bad execution yeah plain and simple right um ai is dumb without context yeah and i say that because i can i can create the most extravagant and lavished type of like ai enriched brief for a rep but if my rep doesn't still know how to actually like have a conversation with a human my rep doesn't know still how to disqualify or qualify they don't know how to do proper discovery or proper negotiation deal management yeah ai won't fix that yeah and so i think that's it's probably the two most important ones and one that um we just hired new head of sales for this um is like my first focus was know your customer yeah and now my focus is build the actual like playbooks and the engine around those customers now yeah and um gb who i just brought on like she's an amazing coach and that's exactly it is like I want her to step into her strengths, which is leading and building great sales teams.

58:45What I then want to do is like, how do I use the stack we have and layer in AI the right way to where she can actually move faster and be empowered to help reps more and be on more calls? That's the environment I'm trying to create. Yeah. That's like, I'm so bullish on that part. It's like, I think the big mistake that orgs are making is they are quick to invest in getting better information and context and doing all of this stuff. And it's like the larger problem for them is that you have a team of people that is either not the right fit to execute it, or they just don't have the skill sets and haven't been like enabled, upskilled, coached, managed, reinforced, like, et cetera, that is such a huge component of it.

59:34Well, the, and the problem is maybe as spicy, spicy debate topic, but like, yeah, it's one of my frustrations when I get on LinkedIn, everything about AI is all top of funnel, top of funnel, top of funnel, enrichment, enrichment, enrichment. And so what I think is happening is you have VPs that, I mean, like I've been on this side, like our, your board is like, are you investing in AI? What are you doing? And I think a lot of people that actually don't understand it, they go and they wear the noises. And so they're like, Hey team, like we actually built a great enrichment process and your reps are like, cool.

1:00:08What do I do with that? Yep. And that's where it stops. Yes. If I were to use an analogy, if you use fitness as an analogy, it's like, Oh, it's been so hard to create a custom meal plan and like a workout routine. And now I can do that a click of a button. It's so fucking cool. It's like, it saved me all this time. You still got to go to the gym. That's the problem is that you're not going to the gym. Exactly. And the other problem is when you go to the gym, you don't know how to do the lifts that you need to with proper form or you don't go at the level of effort that you need to. Like that is really the problem, you know?

1:00:43So ICP product, signal, score, context, action. Killer framework. I love it. We got five minutes left. I want to get your hot take on a couple of things. So just kind of rapid fire style. Hopefully this is a rapid fire answer, but we talked about yesterday when we were walking after dinner about like the future of the sales org and like, are we going into a place where it's like, hey, this is a thing where we're just going to chop the bottom 25, 30, 40, I don't know, half of people and just say, hey, sales, because we've been able to increase the productivity and efficiency by so much, we're just going to two or three X the number.

1:01:22because people are going to be able to deliver on that. Like, where do you see the sales work, like adapting with AI and all of the efficiency gains that people hopefully will start to really see at scale? What does that look like, you think? Yeah, that's not a quick response. I know, right? So I'll try to be quick. Number one, I think we're finally, I would say, at equilibrium. I think SaaS was just way too fat for the longest time. Yeah. And we created that problem. VCs created that problem, plain and simple, full stop. Yep. Number two, yes, I think the market has shown us that teams are going to be leaner.

1:02:05And I don't think it's, at least my hot take is, I don't think it's about rep efficiency. I actually think it's about, we can actually do a lot of the things that are not even focused around rep efficiency a lot better and able to move a lot faster. And so why I say that, man, our mentality, like when I was at Uber, even like back in the day, the high, like you hired SDR straight out of college, you threw bodies at the problem. Oh, we have a lot of demand. Let's not figure out if that demand's worth our time. Just go throw more bodies at the problem. Yeah. We're now we're like, Oh, we have a demand problem.

1:02:41Do we really have a demand problem? or we just throw in junk in, right? Like I think we're now actually looking at it the right way because we don't have the luxury of just throwing people at it. And so I really think it's more of an actual mentality shift. I think there's obviously the reaction of like people being impacted, but I think that's just the shift of how like go-to-market is going to look like. Because we didn't have the constraints. Now we have a constraint on the headcount. How do you like to decompress after a long day or long work week? man what does decompress like um man for me decompressing it's uh it's it's loaded it's family time even though my kids are wild um that is important to me um i think the other one is just reading i think like for me is like um yeah i think it's it's gonna be mostly reading because that's like the time i can shut my brain off do you like to is it like fiction or non-fiction like what kind of reading do you like i love self-help books yeah my wife makes fun of me like i have every single type of practical prospecting like everything from leadership to sales books um what is the oh my gosh i'm drawing a blank right now on the book that i just started but it is it's a fully like how to basically unlock like the best version of yourself oh cool um my big thing right now is i am i'm a i say yes to a lot of things i overwork myself i I burn myself out.

1:04:10Um, and so I think becoming a dad, my daughter, you know, I have a two year old and a five year old that's really put in perception to where give you a great example, man, this is me being vulnerable. Our teacher, she pulled me aside the other day. And she, and so Fridays is when, uh, the girls are, they're supposed to bring a bag lunch. Yeah. Uh, we're busy. And so we forget. So they get a warm lunch, like four days out of the week, they warm lunches, but she was like, Hey, like, do you think y 'all can start to pack, lunches because like the kids get really excited about it and I walked away I was like wow I was like my daughter's the sad kid at the table like step it up and like immediately I'm like we're going to Target we're getting new like lunch bags like what meals we want to do so like those are small things though where I'm also now realizing is like I need to take my personal time and be more intentional about it yeah that's like a way you can show up for your family and stuff that's really nice too man I love that last question is if you were a first time VP of sales VP of revenue all over again knowing what you know now what advice would you give yourself first every VP of revenue VP of sales is the first time VP of revenue VP of sales right now yeah no shit right right so like let's just call that out and so I think like my words of encouragement to anybody would be like first principles, question everything.

1:05:36I mean, I even have people when they're like, oh, I used to do it that way. I'm like, well, what year was that? Because if it was two years ago, like there's a different way that we can do it now. That is honestly, so like, regimen number is like, always be curious. I tell this to everybody, the more you learn, especially in this world, number two is like, go talk to your customers. Oh my gosh, if you don't do anything, go talk to your customers. nothing is more frustrating than as a VP. You have all this analysis and I did this where you want to show off all the data and stuff. And I'm like, just go pick up the fall, go do 10 customer interviews.

1:06:10You'll learn more then than you will doing a bunch of analysis. So just go talk to your customers. Love that dude. Dude, this has been great. It's been awesome. Where can people go to connect with you? Are you guys hiring right now? Where can people go to learn more about Mixmax? All that kind of good stuff. We are hiring right now. Stay tuned. There might be some openings on a sales development side. Definitely reinvesting back on the SDR side. If you want to find me, it's Heath Barnett on LinkedIn. Cool. Easy enough. Awesome. Thank you, brother. Thank you, brother. Appreciate it.