The Transaction

TT - 055 - Jordan Crawford
===

Jordan Crawford: [00:00:00] never in my life has there been a more intimidating box than the chatgpt, the Claude, the Gemini entry box because for the first time in our lifetimes, and maybe human history, we are the limiters.

We're trying to put a legless robot on a horse. you still have the slowness, you have modern technology, but the horse still shits everywhere.

you have to pick a segment,

A segment forces you to make choices.

I invented this framework, called the PVP, the Permissionless Value Prop.

The list is the message, if you know why someone is on a list, you know what to say to them.

they did the one thing that I advise everyone to do, and they're the only one that I know that has done it.

They shed a RR, they got rid of 170 k in a RR because it wasn't the niche that they were going after.

How,

Craig Rosenberg: about this shirt for you?

Matt Amundson: dude.

Jordan Crawford: oh,

Matt Amundson: I could go. I, could go for

Jordan Crawford: I'm into it. I went to, I went to Mizzou. That's where we had, we had those, there

Matt Amundson: You had Mr. Pickles.

Jordan Crawford: Yeah, yeah, yeah, yeah.

Matt Amundson: Nice.

Craig Rosenberg: Oh my God, that's a fun fact. [00:01:00] Um,

Matt Amundson: Insight number one.

Sam Guertin: Yeah.

Craig Rosenberg: just so you guys. we get it out of the way. I pocket dialed my son's friend, the che Matt

Matt Amundson: Oh yeah. Oh

Craig Rosenberg: called me back. Yeah. Jesus Christ. Um, so one is,

Jordan Crawford: Uh, this is great. I You are like a Muppet, like I've never seen. It's just like that doesn't, this is so amazing. Like, I don't have headphones. I'm like, I'm like, I'm not talking down the people that I know. I'm talking down the people that my kids know how to that get in your phone. This is amazing. Oh.

Sam Guertin: he's pocket dialing the people he knows.

Matt Amundson: Yeah.

Craig Rosenberg: pocket dialed Sam like one in the morning on Saturday. I actually, I think Matt, I didn't, I FaceTime pocket dial. You

Jordan Crawford: Wow. Wow.

Matt Amundson: Yeah. From, yeah, just inside your pants pocket. It was great.

Craig Rosenberg: Okay, you guys, this one is crazy. Ready for this. So when Covid first hit,

Jordan Crawford: Don't blame this on covid. Be like, oh, I was, I had [00:02:00] covid. Your phone can't catch C. Like, let's just be like really clear.

Craig Rosenberg: I've people for years. Um, but, uh. But like I do Jordan, the point is I do epic pocket dials. I mean, like, you can't even believe it. I once, yeah, so I'm, so, I'm like, you know what? I'm gonna go running and I'd never listen to a podcast. So I'm like, I downloaded the podcast app, uh, put, just put the phone in my pocket, you know, had the headphones and I start running and I'm so outta shape, right?

I'm just like, huh. So I'm listening to the podcast and all of a sudden I hear my buddy go, rose. And I'm like, what? And then another guy comes on, while running and huffing and puffing, pocket dialed a, like a FaceTime group of like 15 guys who were talking with video. They literally, during the po, I don't even know how this works, they just jump in [00:03:00] and they're like, dude, Rosenberg's like, well, I don't want to tell you what they said I was doing. It, it was, it was epic. I mean, it just, I mean, I dunno.

Jordan Crawford: I can just, I can just see, see Tim Cook is like, we've looked into this, uh, long and hot and we have no idea how this gentleman did this.

Craig Rosenberg: no, that's true when I used to blame, blame Apple Scott Al, my old uh, uh, co-founder, used to go Apple, it's Apple's

Jordan Crawford: I was, yeah.

Craig Rosenberg: Old used iPhones, dude, like it's definitely Apple, not you.

Jordan Crawford: Okay. By the way, your name is squad caster, age 5 93, so you didn't even put your name on.

Craig Rosenberg: what? Oh my God. Look That is,

Sam Guertin: is every episode.

Matt Amundson: Yeah.

Craig Rosenberg: you know what, we should just, Jordan, why don't you come on every month and we'll just do a roast.

Matt Amundson: it's better.

Jordan Crawford: Yeah. Yeah.

Matt Amundson: uh.

Craig Rosenberg: That actually would be a great. [00:04:00] So, uh, so let me, I'm gonna context this with, uh, more stuff that you can make fun of me

Jordan Crawford: No, no, I, I'm done. Yeah, by the way.

Craig Rosenberg: way, just if you can't tell, I.

​

Craig Rosenberg: So, uh, I've been going through, and Matt, you can characterize it however you

Matt Amundson: Okay.

Craig Rosenberg: in the last month and a half I'm going through midlife crisis B2B, go to

Matt Amundson: Oh.

Craig Rosenberg: midlife

Jordan Crawford: Okay.

Matt Amundson: Yeah.

Craig Rosenberg: because realizing that. A generation is gone and it's mine. Like we're just, we have no idea. [00:05:00] so here's the crazy thing is we started this podcast 'cause Matt and I were worked together like a couple years ago and we realized that, oh my God, like the playbook has changed. And it had, and so we said, you know, we started talking about it, so we'd create the transaction so we can go talk to people. But here's the thing. We kept, we were talk, we weren't talking to the people. I mean, we talked to a fair number. I'd say a, a fair number. Have these revolutionary strategic ideas.

Matt Amundson: Yeah.

Craig Rosenberg: And then we've been like. For like a month and a half ago. We're like, dude, we're missing this. Like we need the guys are doing the, the crazy shit happen like the next set, this is the, you, you guys are in this next crew of people that are revolutionizing go to market. And so Matt's like, we need new guess. I'm like, okay, great.

So I called Scott Alber, who I just mentioned. You may not even know the guy. I said, dude, I, we need new guests. We need people that are doing crazy. He's like, have you seen this dude? like, who? He's like, Jordan Crawford, have you seen this guy's videos?

Jordan Crawford: So I [00:06:00] started watching the.

Craig Rosenberg: videos and basically, man, you're, you have taken thing that, the things I'm addicted to on YouTube of like learning stuff where it's fun and it's like you're doing these things and showing like, my kids have e-bikes and it's like I'm learning how to put these things together by just watching these

Jordan Crawford: How

Craig Rosenberg: you're doing it for AI and for go to market. And I'm like, oh, I need that

Jordan Crawford: Yeah. Cool.

Craig Rosenberg: everyone knows I reached out and he's like, hell yeah, let's

Jordan Crawford: Yeah, let's do it.

Craig Rosenberg: And now I've been listening to your stuff. Now all these guys are, and then Sam was like, he killed it on the go to market now podcast.

so, you are. guest that's gonna transform the show to the next level, which is there's crazy stuff happening. There's a new way of looking at things, there's a new way of doing it, things. Can we talk to the right people? And I will tell the podcast audience right now. We have the right guy.

It's Jordan Crawford, and welcome to the transaction.

Matt Amundson: Yeah.

Jordan Crawford: Amazing. Where do you want to kick this off?

Craig Rosenberg: Well, I want to ask you this. [00:07:00] So normally we ask people to tell us, uh, a, a funny, amazing, ironic, whatever go to market story, like something fun. Um, but we've done a bunch of those. I don't know if you have any others you wanna lay on, but we like to lead with like, uh, something sort of fun, but that's instructional.

Jordan Crawford: Oh my God. Fun and instructional. Um, I don't know that I have anything that, uh, that fits both of those categories. Most of what I do is instructional. That happens to be funny, not the other way around. I am thinking about how best to orient people because, um, generally what I do is I will describe the current model and what tools and technologies are slotting into the current model so people can see themselves in that.

And why? The way that I talk about this is that we're trying to put a, um, a legless. Robot on a horse. Um, you still have the slowness, uh, you have modern technology, but the horse still shits everywhere. So you're not going anywhere faster. But there's like pieces of new technology there of course. Um, so maybe let's start [00:08:00] there.

Um, and so like I'll go over this argument briefly because it is good to know where we have the framework that we're operating in, especially for sales right now. Because we'll understand why slotting tools into that broken framework doesn't work. So let's like quickly go over that framework. I'm doing the politician thing, which is, I'm answering the question I wish I would ask.

Like, it's like, it's like, well, what do you think? What do you think about benefits? Well, let me tell you about our jail systems. Um, so, um, so, uh, yeah, yeah, yeah. Um, okay. So the way that we do a lot of, like our go-to-market today is like, we do account selection. We do persona level messages, like VPs of marketing, whatever.

Then we go find contacts. We have SDRs personalized and we scale. And so this is kind of the general five step motion and the problem is that all of our tools are slotting into that. That existing motion. So it's like, oh, the AI SDR is slotting into this SDR personalization. And what they're saying is, I have no idea [00:09:00] why I'm messaging you.

but what I'm gonna do is I'm gonna write my way out of this hole because it's not tied into account selection. I. If you think about the people that let's, and let's go through these linearly. Let's talk about account selection. So, the sort of very latest go-to-market tools and account selection are trying to sell you signals.

Here's what's changing in the world. Well, the problem is the more people buy signals, the worse they get because signals have nothing to do with what you are selling. there are some other person's heuristic that you are trying to shoehorn in your process and everyone's doing the same thing, right? It is the next evolution of, I saw you raise money.

I like money, can I have your money? Right? That's like, and that like worked for a while, which is like fucking insane. Like, can you imagine the SaaS environment where you could send that message to people? Like, I do need to spend this money. Um, so this is like, I mean, I might as well give it to you. You sent me, you know, it's like the investors be like, this is reckless.

What are you doing? Like you can't just respond to some random. on the persona level messages, the, the most innovation we're seeing here is really on the training side, where there's AI agents that, uh, sales reps can [00:10:00] call and, and like they can be angry customers, which, yeah, yeah, yeah. It's like pretty interesting.

It's like, who, why are you calling? It's like, I'm calling to sell you about my insurance solutions. Like, get the fuck outta here. Like, it hangs up on you. It's like amazing. Um, the find contacts there are tools like Full Enrich, um, which are these waterfall contact providers. So we don't have to rely, and this is like valuable.

Um. on the scale side, uh, so we talked a little about the AI SDR, um, and generally the AI SDR is gonna have problems because it's doing the same, it's doing the same fucking thing that screwed companies in the original thing, which is like you are selling horizontal SaaS to 7 87 personas, 57 different industries, like with 12 use cases, and you've got a million dollars in ARR.

What are you doing? Like this is totally ridiculous. Of course, you're gonna have problems going to market because you haven't picked, and the ai SDR, and this is the thing about AI models generally, is that they don't pick, they're like, I'll coming right up. It's like, Hey, I'm at a burger place. Can I have some sushi? Coming right up.

And you're like, this tastes like total garbage. And it's like, uh, you're right. Now that I've, uh, thought [00:11:00] about it, I don't know how to fucking do this. Right? Like, these are the models, right? And so we need a framework to apply the models in where we can. Get the best of them, the sort of 10 x version without suffering their insufferabilities.

And the AI SDR is like positioned perfectly in the old world, which is like the reason that they're suffering is because we never made a choice. And we're saying the largest language models, you're the smartest thing that humanity's ever created. You figured out. And the large language model's, like, I don't know what you mean.

Sounds good. you guys have any questions about that model before I sort of transition into the, to talk to you about how I think that we should move into this new world?

Matt Amundson: No, but it's like, uh, this is a soapbox that I've been standing on for the last couple of weeks, which is, you know, this concept of, uh, you know, people have been buying solutions, especially in this like, sort of complicated market. I think that's the nicest way of saying a shit market. But like in this complicated market where you're like.

God, like, I don't know. I don't know how to solve this problem. I'll buy something that'll solve the problem for me. Right? Like, and I'll and AI's buzzy. [00:12:00] So like, to your point, AI is super smart. It'll solve the problem for me. But the reality is, is like you have to go out, solve the problem. And in certain cases that might mean making hard decisions like.

Yeah, we theoretically could sell this to all these people, but where, what's the one group of people that we can sell this to? So you gotta figure out how to solve that problem. You gotta figure out what the message is, and then you buy a solution that scales that instead of, Hey, I'm gonna buy something that's gonna solve this for me.

Under the pretense that like AI is smarter than I am and it's gonna point me in the right direction. It won't, it just won't.

Jordan Crawford: I totally agree with you here, and the way that I talk about this is that you write the message manually as if you had unlimited amount of time to research one company. Go find the company that closed recently use deep research. You can use all the AI tools, but you write the one message and I'll tell you, VPs of sales CROs, they can't do this because they're so far removed from the data.

So, so let's talk about the new [00:13:00] world and then let's talk about how you close that gap because it's important to have some functional things. So. The first is you have to pick a segment, and what I mean by a segment is that you have to identify what is the, if these conditions are true, this company not only is, and there's two sort of components here, not only is qualified, but they are the top

quantification of that segment. And so a qualified, um, and I'm not talking about head count, right? Like you might think of, qualification as well. The account needs to have 10 sales reps because we have a tool, we're seat-based, whatever, right? But quantification is, they have to, the more complex their solution, the better it is for me.

So let's talk about a segment for like lean data, right? That does routing their best segment. very, very complex sales team. lots of traffic. Those two things. Just those two, right? And so, um, they basically are like, oh, you have a sales team, you've got a Northwest team and an east team and a, [00:14:00] and it's like, okay, great.

Very, very complex team that is getting just like, hammered with leads. And so you need to figure out, okay, well does this go to the guy in Abilene, Texas or Derby Texas? Like, I don't know, how do I do that? Like, so that's the first thing. So you gotta figure out that segment. A segment forces you to make choices.

So you got the segment, what's the second piece? Data. What is the data to identify the segment? Now, generally, this is two to five unique heuristics. Often that imply tension. So in the lean data case, it's like lots of traffic, complex sales team, you have tension there,

Matt Amundson: Mm-hmm.

Jordan Crawford: right? And so the nice thing about that is that those two things together are the message.

Matt Amundson: Mm.

Jordan Crawford: Jordan, um, you've got 156 people in Texas alone. Um, you have, uh, you know, 20 million hits a month on your website, and you are selling in this case, a storage solution that has 456 options. Someone has to speak with sales. It is highly dependent on the area [00:15:00] because it's a physical location, it's a storage location, and also there's a bunch of bells and whistles that you're going to need.

That means that you have to talk to a person. LeanData can route that. So that's what I call a PQS, a Pain Qualified Segment. Um, so we're still on step three. There is one additional version of this, and this is sort of hinting to the future. I invented this framework, but it's called the PVP, the Permissionless Value Prop.

So again, we're still in this message section. Um, value, I'm sorry.

Yeah,

Craig Rosenberg: by the way, I'm wearing my headset for

Jordan Crawford: because it charged thing for me. Yeah, it's your podcast. You're like, just Jordan, for you. I'm wearing my headset so my podcast doesn't echo.

Matt Amundson: We like to roll out the red carpet here.

Craig Rosenberg: wait, but say it. Say it. So the first acronym was what?

Jordan Crawford: A pain qualified segment, and this is two to five data heuristics that imply tension on the account. Right? And that tension has a timing element, right? Um, and so, like for example, there's a lot of supply chain disruption right now because of the tariffs, right? And so they're, these people are scrambling.

[00:16:00] They're like, oh my gosh. if you're selling a tool that sells to manufacturers. You might want to identify who is under that tension. Right? Well, if you're buying stuff from China, probably more tension than you're buying stuff from Germany. Um, and like, well, where is there the most tension? Well, it's probably not computers.

Like, it's probably like, you know, other types of goods. Right. And so you're, and it's like, okay, well it might, it might even be more specific. It might be port related. Okay. Well if you're shipping outta the Shenzhen port, I don't think that's a coastal city. Maybe it's, um, but um.

Matt Amundson: is. No, it's landlocked,

but you're good. yeah, yeah, yeah, yeah.

Jordan Crawford: The, the, yeah, yeah, yeah. Yeah. Fuck. Uh, um, so, so Covid. Yeah. Yeah. Covid. Yeah. I got, I got that Covid. Yeah. Yeah. So, yeah, I don't understand geography because of Covid. the idea behind the PVP, the permissionless value prop is, um, on that data, what you're doing is you are combining two to five data heuristics to deliver a message that is independently valuable to the recipient before you talk about your product at all.

So [00:17:00] let me give you an example of this for a company called House called Pro. and generally it's just using public data. So, um, I have access to the world's largest permit database in the United States. So one of the things that we can do, and they sell to HVAC companies, so, um, the, you know, air conditionings, et cetera. So one of the things that we can do is we can look across all of the permits in the country, and it turns out for a lot of HVAC systems, you have to pull a permit. And so we can go and at the contractor level, I can look at all of the contractors in the country, all of their permit history. I. Okay, well, why is that interesting?

Well, it turns out that HVAC systems die after about 15 to 20 years. AI can identify the system that was installed, its average lifespan. It can also look at all of the permits, uh, for that address, right? The contractors doesn't have access to this, and I can send a message that says, Hey Craig, I know in the past you worked on 1, 2, 3 Main Street, 1, 2 7 Bernard Street and 1 8 5, uh, Leonard Street.

You installed jobs there that were worth, [00:18:00] 10,000, 20,000 and 50,000. All those systems are coming up to be replaced, and by the way, we've looked at all of the historical permits for those addresses. It turns out no other HVAC contractor has worked on those addresses. I thought this would be useful too.

You might wanna contact your existing customers, you probably have about $50,000 in upsell potential here. so now I turn an outbound to an inbound, right? People are like, oh, who are you Jordan? What do you, what are you doing? How, how'd you get that? Right? Like, like it leans people in because they wanna understand.

It's like, look, this is the type of thing that you can get if you have House Call Pro, you can. You can, and it relates to their product, right? Their product is a CRM for these HVAC companies, right? And so you can take that public data, weaponize it and provide it as unique insights to the customers, and that defines your segment.

So we're only at, we're only at three, but let's, let's pause there. Yeah. Yeah.

Craig Rosenberg: By the way, who is Matt? Just if you guys don't mind, it's not because I got made fun of the most, but is Matt breathing heavily in his microphone?

Jordan Crawford: I don't know.

Craig Rosenberg: No. [00:19:00] Okay. All right. Just now he just, I, that was, he's in a cave too.

Jordan Crawford: I.

Craig Rosenberg: got that. All right. So that is, I love that, the peak. Okay. Keep going.

Jordan Crawford: Yeah. Okay. So, so we have, remember we've got the segment. So we've identified the segment, the data, it's, the data is like two to five heuristics. And we keep this simple because once you get, like sometimes I'll, I've, I've been at customers and I look at their like account scoring model, and they're like, here are the 368 factors we take into account.

I was like. There's no fucking way you can make sense of that. Like how would you, it's like, well, we, we have a sign function and a co-sign function, and then we go to the port city of Shenzhen to be able to calculate this. Uh, you know, they have plenty of calculators and then we use an abacus. It's like, it's like, no, you don't, you have no idea.

Like you have no idea to be able to measure the efficacy. And it turns out that that sort of constriction is important because you have the opportunity to say, is this really valuable? So, okay. Segment data message. We've got the first three. Well, then you can [00:20:00] channelize that message.

So unlike intent data, which is like, we have determined that you should sell to this person. We don't know why you should sell to them, but they searched some keyword on some website because we cookie someone, it could be an intern at the Salesforce Tower, it could be Mark Benioff.

We got no idea. But why don't you just deploy a shit ton of marketing dollars to them? And by the way, we'll tie into your CRM and and prove to you that it worked. It's like, okay, well I. It's a black box. How am I supposed to do anything? Um, Facebook. Facebook is like, don't worry. You don't know who you're targeting.

We've got machine learning and experts and we're, we'll sell you the click. We'll make sure that we get the hell outta clicks. Google is like, we know the exact things people are searching for. If like you sell an electronic toothbrush, you of course will pay for that, right? But all of these methods require you to outsource your thinking about who your best customers are to someone else.

All, every single one of them. With cold email, You have to pick a person. Why that person? It is the most relentless channel, and the nice thing is you can't go. You can't dig into the [00:21:00] intent data, you can't dig into the Facebook data. You are blocked. But with cold email, you can do the other way, which is, you can say turn this, because it's a single person that is inside a segment that has data heuristics about the problem there and now.

And so you can write an email to them. You can write a LinkedIn message to them, you can, uh, write a cold call script to them. It turns out that you can actually match a person to an ad id, which just means that I can upload a list of individual people just like I would to a sequence tool to Facebook or to LinkedIn.

And so suddenly I can say something to them. And the nice thing is AI is really good at reformatting ai, AI's not inventing anything because remember we did that in the data section. We actually. We use APIs and we use technology to be able to determine like, is this true about this account? Um, so you basically can use AI to channelize that message and deploy it to different channels.

And the way in which you scale is you do another segment. And so you just do this for plumbing and you just do this for, you know, if, if you serve a different area. [00:22:00] And so you segment by segment by segment, you can run all these automations and suddenly you can get all the benefits of ai. It's doing things like changing the message, et cetera, and maybe there's even AI agents as part of this.

You get none of the downsides. 'cause it doesn't have to guess why it's contacting someone. It already knows. 'cause we did the data work, right? We did the segment work, we did the thinking work upfront. And you can, just basically replicate this same model. And, but it requires understanding, like it requires understanding and that that's the one thing that, That we don't yet have the tools to do understanding at scale. Gong has been not doing a great job of this, honestly, like, and there's some other tools that are trying to do this. I'm building a gong integration to help with this. but we have to begin with the understanding and the, and the segment identification.

And if you're at a large organization, really the only people that know it are the SDRs that get fired if they don't do their research today. They don't know the heuristics, but they can tell you all the individual data points. So you have to say, okay, well when you look at these seven things, that means [00:23:00] this concept and those are the concepts that you can tie into the data. Sorry, just like went on a big, long, I, it's like a podcast. It's like more of a monologue. Sorry guys. It's like the podcast of the single guy talking.

Craig Rosenberg: no, I love it. And I think you had an amazing quote with me, which is, what was it again? The

Jordan Crawford: headphones. I think I said last week

Craig Rosenberg: Well, no, that wasn't a theme,

Jordan Crawford: that was,

Craig Rosenberg: uh, you have

Jordan Crawford: you said like the list is the message.

Matt Amundson: Yeah.

Jordan Crawford: Yes. Mm-hmm.

Craig Rosenberg: And you just basically said, but, but do it again so that we could do a 32nd snippet on this,

Jordan Crawford: Okay. Okay. Yeah, there's you just fuck it. Don't edit this. Just ship it. Like just who needs editing? Right. Just.

Craig Rosenberg: by the way, we don't

Jordan Crawford: Oh, okay. Yeah, that's, yeah. Yeah. you're saying. I

Matt Amundson: Sam might beg to

differ, but.

Jordan Crawford: three, there's three of you of course, though. Why would you edit? Like there's not, yeah,

Craig Rosenberg: he edits, but not, we like

Jordan Crawford: So.

Craig Rosenberg: and downs. so the, the quote is

Jordan Crawford: the list is the message, which [00:24:00] is if you know why someone is on a list, you know what to say to them. And so all you're doing is re-describing the targeting back to someone. And I'll give you one example of this owner.com. They, they sell into, restaurants that are paying a ton on DoorDash and Uber,

Matt Amundson: Yeah. Yeah.

Jordan Crawford: so think about the list they could build. They know, they can find your review count that's public and they can, find your menu prices and they can compare that into the menu prices on your website. So you know the delta that they're charging and the, the review count is a proxy for orders.

So. If I can say, I'm just gonna take the folks that have really high DoorDash and order reviews, right? They have a lot of reviews and also a high delta between their menu prices and the, Uber prices. It's like all I gotta do is describe that back to you, Jordan. Like you have had amazing success on DoorDash and, and Uber.

and my guess is based on your review count, you are probably sending about $46,000 a month off to these [00:25:00] platforms that you get no piece of. When John's tacos right down the street from you, he's just a mile away. He removed himself from these two platforms and it turns out that he didn't lose a single order and he kept that $46,000. All I'm just doing is describing why you got in the list in the first place,

Matt Amundson: Yeah.

Craig Rosenberg: Yeah.

Jordan Crawford: that's a pain qualified segment.

Craig Rosenberg: that killer or what?

Matt Amundson: That's amazing.

Craig Rosenberg: think of it? I mean that, that might

Matt Amundson: Don't ask. Don't ask Sam's opinion. Don't ask Sam's opinion. He's just in the, he's just,

Craig Rosenberg: uh, sorry. All right, hold on, Sam,

Jordan Crawford: not gonna edit the fog. He's like, he's like, screw these guys. Like, hmm.

Sam Guertin: think it's great. I think these are some wonderful examples, uh, that are very tactical and actually helpful in the way to actually think about how to do this.

Matt Amundson: that was very I, do this live

Jordan Crawford: every 9:00 AM Friday. I do this live cannonball go to market where we will take a brand, pluck it out of a hat, and we, we basically do all of the synthesis of this knowledge live and we'll try to get to a message that people will pay to receive, um, without knowing anything [00:26:00] about a brand.

And so, um, like there's no reason that you can't take. Like the thing is, the methodology is really important because AI can slot into the methodology and you have checkpoints along the way. So it's like, is this really the best segment? Is this really the best data? Can we find better data? Like, is this message really going to resonate with this segment?

Because I can't say, is this a message that's gonna resonate with everyone that uses this product? 'cause it's like financial services and healthcare and, and AI does a really bad job at that. You really need to narrow ai and you can do that in the context of segment. You can't say who's the best customer for copy ai because they serve anyone with two teeth and a a computer.

Right?

Craig Rosenberg: Yeah,

Matt Amundson: Minimum two teeth.

Craig Rosenberg: wait, two teeth. So like they,

Jordan Crawford: Yeah. Not, so that's a, that's a qualification criteria. They have to check if you have to. Yeah, if you don't, um, there's no, I'm, I'm pretty sure they have no customers in West Virginia. Like just total guests like, yeah, yeah, yeah, yeah, yeah, yeah.

Craig Rosenberg: so one thing you said I think is really important, which is kind of [00:27:00] offline. I, I feel like, um, you said we can find out the data we need by just talking to the SDRs. right. And finding out what things resonate and work with them in driving demand. And then we take that data we bought, we, we sort of find a level, uh, you know, data point there that we can put into, create segments from,

Jordan Crawford: Yep.

Craig Rosenberg: uh, is that what you said? Because

Jordan Crawford: Yep.

Craig Rosenberg: that I, people try to find scale in that. And what you're saying is, well, no, they know. They're the ones who will get fired if they can't figure it out. Is

Jordan Crawford: Yeah. Yeah. And usually, usually in advanced companies, they know and, and they will, uh, the STRs will give you breadcrumbs. So usually the sa, the sales leaders don't know. It's like, well, what makes a, a best customer? And they'll say, like, over $50 million. And I'm like, so you want to go for rich customers?

Like, absolutely I want people. And it's like, well, who makes a better customer? Well, is their wallet even bigger than I [00:28:00] want 'em? And it's like, okay, I had a, this is like really funny. I had a CRO, walk in was meeting with. Like the person that does work at the company. Um, and, uh, the CRO came in and, and he is like, pull up that chart.

And I'm like, alright. Right. So he pulls up this chart and he's like, Jordan, here is my bar and here's where I need to get to. And I'm like, I understand your job buddy. Like, I get it. Like, I like we're working on moving that bar up. Like can you tell me why? And it's like. We need to just move it up. And it's like, okay, so, so like, really you need to, the C-suite, the, the CRO, the, the director, et cetera, they understand the context of the business.

They don't understand the context of the lead, and they need to really connect those two things. And so the SDRs will be like, well. So like one of the customers I was chatting with, they're like, we look at, the hygienist, like they sold this sold dental ai. And I was like, well, why do you look at the hygienist?

And it's like, well, um, because doctors are over their skis. I said, well, okay, well, like, [00:29:00] let's dig in this a little more. It's like, well, it turns out that they have a dental AI product that can identify, diseases without. A dentist looking at the x-ray. So what they were saying is that if you have a small number of dentists and a lot of hygienists, that means that you are likely missing a lot of things because the hygienist can't look at those x-rays.

So that's the heuristic, right? That is the heuristic. The fact that they have hygienists actually is not really useful. It's like, okay, well what do I do with that? But it turns out they have a lot of them relative to the number of dentists. That's a useful heuristic. Um, and so. It's the blending of those two things that can turn that thing into a segment.

But you have to connect those two worlds. 'cause the SDRs like, I'm not gonna get fired. Okay, they got this, they got this, they got this, and then they're just trying to put together a message at low scale over and over and over again. So they know, but they, but they don't know how to connect that to a segment.

And then, then once you can do that, then you can operationalize everything. Um, but otherwise what happens is that horizontal [00:30:00] B2B SaaS CROs. The, their best tactic is a whip, which is like, all we need to do is we need to go to the people that will get fired and be like, you know, you're gonna get fired.

Right? Like, and, and like, by the way, this works too. Like, it's like activity matters. Like it does matter. Um, but you are, you are basically saying, I have. I have screwed you all the way to this point. I'm not gonna give you any tools. I don't know why you're targeting. You have, like I had with this one SDR R that was like, I target oil and gas.

The company had written one single document about like a case with Chevron and it's like this guy has to like talk to executives and oil and gas, like get outta here. He's like totally unequipped to do that. Some 22-year-old, and of course he has the. Machismo of, uh, you know, of a mic or a Chad. And so he's like, nah, well, I, I gotta do this, you know, and so, no, but it's, it's like, it's a problem.

Craig Rosenberg: Oh man, that was awesome. And then when you get the heuristics to create the segments, are you, is the app you would [00:31:00] use like clay or are there other things or what, what, where does it go from there to,

Jordan Crawford: Yeah, generally I'm using, generally, I don't do list creation in clay. I do list automation in clay, and so clay is like not really good for list creation, and I think that that's kind of an intractable problem. So for one of our customers, for example, they wanted to scrape a website that literally had all of the

donation data for companies on it. So like, here are the donation programs, here's, do they do employee matching? Do they do And so you can, you can get a lot of information. They have sold software to help manage that. Um, so you can just get a lot of information because there's all, there's like some corner of the internet that is already structured this data or like, um, uh, this is like a fleet card for trucking companies, right?

Well, it turns out the Department of Transit knows how many miles people drive. They know how many trucks they have. They know how many hours they are on the road. Like, and a lot of that's public, right? So if you're thinking about a fleet card, you'd wanna go scrape that. [00:32:00] I forget their website.

It's like, but the Department of Transit has this, or we scrape like a, a website that has like, um, that has arborist. And so like we wanted to understand who are the certified arborists. And so you generally. And a lot of the work that I do is in vertical SaaS and technology companies sell.

This is really hard for technology companies selling to other technology companies. This type of thinking is really, really hard to deploy because they are selling, they sell a complex solution to a complex customer, and so finding the point of connection is really hard because they're like. I don't know what I, I don't know who I am best for and I don't know how that actually connects into this other person.

And so, um, this is really, really hard. A lot of the examples that I gave you are vertical SaaS and that, and. That's because it is so much easier to know your customers. If you're selling to a plumber in Alabama and a plumber in San Francisco, it's not like they're like, but you work on pipes. We don't do any pipes.

Like, you don't do pipe. You're a plumber. You don't do pipes. It's like, no, pipes not here [00:33:00] in San Francisco. We just deliver wi, we do wifi, we deliver plumbing via wifi. Like, so, so it it, it's like this is a much harder exercise if you're a horizontal B2B SaaS company selling into a horizontal B2B SaaS company.

Matt Amundson: Hmm. Hmm.

Craig Rosenberg: Yeah. Got it.

Matt Amundson: so let's talk about like some of the companies that you work with. Like, we don't have to talk about them in specifics, but like, typically speaking, once you start engaging, where, like, where are their primary pain points? Is it just like, hey, we've got a team, you know, they're, they're doing a lot of activity but they're not getting many results, or they're kind of like, nah, I don't really know how to target.

Is it more their pain points are around, I don't know how to segment. Like when you find. A group where you have a lot of success, like what does that environment look like? And then how do you start to transition them out into this, like into your model that you've been using,

Jordan Crawford: It's a good question and I would, I would answer it differently. Six to eight months ago. Six to eight months ago. Well, I should, I should answer it today. Um, the way that [00:34:00] people are thinking today are like, I know I need to use ai, and I have maybe gone through one or two rounds of like, the current AI tools and I don't know how to think about this.

Um, and so I will give them the frame and then I now have like a two prompt process that can develop my scope of work usually. And I say, great, here's like the type of things that I would do from you. Here's the data sets that we'd look at. Here's how we would look at it, here's how we'd structure it.

eight months ago it was more like, people would ask for things in the old model, can you do account scoring? Can you do a TAM score for me? And it's like, the problem with the TAM score, total addressable market for, for those that don't know acronyms, um, I, I don't know most acronyms. This is the only one I know.

they'll say, we'll score all my accounts. It's like, okay, this like helpful focus on these, not these accounts, but the question is why, why these? And it's like, well, we've got a complex scoring model, so you hand it to the SDR and it's like, this one is better than this one. And so \ all you're doing is you're saying, I have put a bow around this pile of dog shit for you.

It is better than without a [00:35:00] bow. I mean, it just, the bag was open before, so now, but now you have to turn that into revenue. That year's. Like, what do I do? It's like, do I throw it? Like whatever. Um, and

so. Yeah, yeah.

Matt Amundson: front

Jordan Crawford: Yeah, yeah, yeah. I put it in, I put it in a bag. Like if I put a bag light on fire, like someone's gonna laugh and if they laugh, then I can get in the door.

Um,

and so, so really now people are kind of recognizing they need a new framework. They need, like a go-to-market engineer. People are sort of saying, we need this thing. That's not quite rev ops, that's not quite. Tooling. It's like, uh, it's like strategy. ZoomInfo. Apollo aren't quite the thing that we need.

Like, and they don't even know how to like, wrap their arms around it, but they kind of know they need it. And Clay has accelerated that transition. And in some cases Clay has made it worse because, um, I, I, I, I never, I never really recognized this until I had a lot of conversations, but, um. Like a lot of go-to-market leaders are like, I need clay.

And I'm like, why? I don't know. I just saw it like a lot people are saying like, I need it a lot. And it's like, okay, well [00:36:00] why? And I think that in the sort of the previous world, um, these like intent tools like sold into that, they're like, you don't know what you need. I know what you need. You need intent.

And it's like, why do I need it? Because everyone needs intent. You, do you have any good ideas? I don't have any good ideas. Like I've got good ideas for you. It's intent data. And so, um. Clay is like, clay is capturing a little bit of that. Um, but I think that unlike with intent data or whatever, there are actually people that can translate this like, I need clay.

'cause clay is basically python for go-to market people. So it's the same thing when you say like. Um, and he said, oh, like, people like, great, can I get Python? I'm like, yeah, you can get Python if you want. It's like, well, well what now? And it's like, well, what do you wanna do? And it's like, well, I got the Python, so now, now that I have Python, what do I do?

And it's like, you have to have an opinion about what to do with Python. I was like, what do you mean? It's like when I bought Apollo, I just got a list. I don't know it. And Clay is like, no, the whole point is extensible. And it's like, well, I don't know what to do. And so now they [00:37:00] kind of recognize, they need to know like the framework with.

What to do with these new tools. Um, and so they're receptive to, to learning about that usually. I dunno if that answered your question.

Matt Amundson: Uh, uh, I mean, it started to, and then it, it sort of, then we started talking about python and clay. But, um, which is a lot of fun. I, I, I love it. Uh, no, but I think that this is, this is. It's starting, it's a, it's a murmur and it's gonna be a roar here probably, you know, within the next two quarters is this concept of like this GTM engineer role and like having that sort of on demand.

And, you know, that's a little bit of what Blueprint does and like, I don't wanna pigeonhole it is just that, but like this, this idea that like you're finding somebody lost, but where are they lost? Are they lost in the, they're lost in the, I don't know who to contact or they're lost in the, like once I know who to contact, I don't know what to say.

Uh, because I think you've brought up a couple examples of like, hey, the, you know, the, the, the list is the, is the, is the story or the list is the message. Right? [00:38:00] And like that's, that's, that's a very different way of thinking. It's just an inverse way of thinking about it. But I think it's a really smart way of thinking about it, which is before we'd be like, we've got a message, so let's build a list for it, as opposed to thinking about like, Hey, there could be a lot of different lists that have a lot of different stories, and maybe they're not.

You know, massive, but like we can get specific with them and we can go tell a story that like should resonate. And you, you brought up plenty of examples of that, but when you're, when you're sort of transitioning them, like what are you transitioning them from and where are they ending up going? I guess the, from two, right?

Like product marketing speak, sorry.

Jordan Crawford: Yeah. So, no, no, no. Um, so the, the, the from place is in the most sort of advanced cases. They have account scoring, they understand their tam, um, and they're pushing, they're pushing all of that intelligence down to the SDR r to structure it, which is like. Look, here's 47 fields. I dunno what you're gonna do with it.

I'm my rev ops guy. [00:39:00] We, we spent a lot of time putting the, these 47 fields, so we, we really have said to the least well paid and the least capable person in our organization. Like, no ding on them, just like that's where they're on their careers. Um, go ahead and make sense of something that I haven't made sense for, for you.

And so, so the problem is that, and, and the way in which I will know if you're good at this is I'll just fire you if you're bad at it. Um, and so it's this like poor person that doesn't have the wherewithal to be like, this is wrong. Like, you shouldn't do this to me. You should help me understand.

And it's like, why don't you, I said to this Sierra, I was like, you prospect for one day. You prospect for one day. Um, and, and see what it's like to work in your organization. I saw a post the other day where this guy said, I logged my clicks. He's like, by 11:00 AM I did 1300 clicks. Like for his job. I'm like, 1300 clicks.

Shoot me now. Um, so. So that's the world that they're coming from, the world that, that I'm taking them to. You could think about this, like these segments, these are just campaigns. So for, [00:40:00] um, we work at a company that sells to, to, uh, pool cleaners. And what we did is I said, give me your internal data about prices for a pool cleaning at the zip code level.

Don't tell me names. And so the, the campaign that we ran was. Hey Jordan, do you charge $126 for pool cleaning and 9 5 1 2 5? You should. Our data shows that this is the optimal price to charge for a pool cleaning. We have thousands of, of pool cleaners in 9 5 1 2 5, and the prices range between 50 and $350, but this is 126.

That's the number to charge, like, do you wanna know how I know? And so the. The, and they're, they're using that and smiling and dialing, and they haven't, it's taking them a while to action this because things like multi-domain setup for outbound, like not, not yet set up, right? So they're using some other tool.

Um, they don't have tools like Nooks or Orum to do, uh, like parallel dialing, right? So, so there's like a. Channel [00:41:00] distribution problem, which by the way, all the channel stuff that we used to buy, uh, a year ago plus all that stuff is for the best organizations. All that stuff has changed. Um, uh, so they have a channel problem and they have a message problem.

Um, and uh, and then, you know, of course everyone has an attribution problem. So those are the kind of things that people are working through. I don't know, I may know who to target, maybe in the best case, they know who to target, but they don't know what to say, which is actually really a targeting problem.

Um. Once they know what to say, then they need systems to be able to deploy what to say that is also new spend, new things that they have to connect. And then they have to figure, well, how the hell do I measure that? Because they just replaced my tooling to do deployment.

Matt Amundson: Mm-hmm. Mm-hmm. You're, yeah. I mean, what you're diagnosing is a problem that like, honestly, you know, Craig and I worked together, or Rosenberg and I worked together. I like that you updated your, uh, your name on here, Craig. Nice job.

Jordan Crawford: Oh.

Craig Rosenberg: Uh, I mean, take feedback well

Matt Amundson: [00:42:00] Well, I mean, espec, especially

when it's you can hear it and when your headphones are charged, right?

But what I think, what I think you're diagnosing here is a problem that I see in almost every early stage startup, or, I mean maybe not even early stage, like even companies that are moving into a scaling motion, right? Like this is, this is, I think back to my time at Marketo and like.

You know, I ran the SDR team there. Like we struggled with that for a while, right? Like, this is not a unique problem, but what I think you're offering is like the unique solution to it. I wonder, like sometimes I wonder if we diagnose how we got here is like, did we get here through like, you know, the proliferation of sales tools that really took a lot of the skillset.

Away from both, both sales and marketing, right? Where we just said, Hey, like let's just play a numbers game instead of playing a qualitative game.

Jordan Crawford: Yeah. Well, I mean, it, it's a couple of things. It's always a couple of things, right? Um, so one of the things is that, um, it, it used to not be as competitive, right? You could [00:43:00] start sash like, they're like, oh, like what if I sold contact data, like nobody's selling contact data. Amazing. You know, and it's like, it's like, what if we automated emails and it's like, no one's out of any emails amazing.

You know? And so. So you, you existed in a less noisy environment where there was a lot less competition. And, and I think that the other thing is that generally when you start a startup, uh, and you say you don't know who you're targeting, like you just don't know, right? You think it's this and, and you adapt.

And so what happens is you start to collect customers. It's like. Well, what did you buy me for? I bought you for a toothbrush. And it's like, what do you buy me for? I'm gonna use this as a desk. Desk, whatever, man. I'm gonna use this as a camera. I'm gonna use this as a roof. And it's like, and so suddenly you're like, well, do I sell to people that want toothbrushes or roofs or cameras and, and you're like, oh, I'll sell to all of them.

I will treat them all the same. And so the way in which we built those, those revenue engines, like they didn't start with a aggressively opinionated understanding of who the customer was. We built these horizontal [00:44:00] tools like, I mean, hey, this is a tool that anyone can use to send email. It's like, who needs that?

Everyone who does, who doesn't send email? Everyone sends email. Um, and, and those organizations are in the worst place now, the best founder that I've ever advised. They, they got so much money from this. I had helped them in the early days use AI to invent ideas with ai.

About how AI could be used at that company. It's like really, really meta. So we were using, we like took the company's website. We said, what do they do? We said, yo ai, if I was building an AI product for them, what would it do for them? What are the problems that it would solve? And so we basically inverted this model and it said We are Stanford yc.

Young, hungry, and like, these are the exact things we do. So it invented the value prop of the company, and they started getting like floods of calls, right? Like, oh, I, we need an AI of course, to go evaluate our financial risk with blah, blah, blah. It's like, you do that. It's like, absolutely we do that. Yeah.

Um, and so what happened is they started [00:45:00] getting all these random customers. Now the VCs loved it. They're like, you're getting money. Get that money. Get that money. Um, and so what, what ended up happening was they, uh. They, they did the one thing that I advise everyone to do, and they did, and they're the only one that I know that has done it.

They shed a RR, they got rid of 170 k in a RR because it wasn't the niche that they were going after.

And so they said, we're not, we're not doubling down on this sector. We will not be amazing at this sector. Like, and so they said, sorry, Siara, like thanks for the memories. Like we are gonna end the contract.

And by doing so, they. Understood the opportunity, cost of poor targeting, poor product, poor understanding. And now they are grown like gangbusters because they're like, we just serve in this one niche. Like they're Tam right now. 2000 companies. Like, and it, and they're just, they're literally, people are replacing [00:46:00] people with their solution because they went so narrow and they made the value so obvious for that segment, and that segment was like so underserved.

They picked their pool. We don't do that now, we don't think, you know, it's like the joke is right. First time founders focus on products. Second time founders focus on distribution. The best way to solve for distribution is to pick a very, very, very niche audience that exists in the 1980s in terms of technology, um, and has like plenty of money to spend.

Matt Amundson: Yeah. Whew.

Craig Rosenberg: I'm exhausted.

Matt Amundson: Are you already?

Craig Rosenberg: I couldn't hear anything he said. These headphones aren't working. No, I'm just kidding. No, I heard the whole thing. What the hell?

Jordan Crawford: Don't you have someone to FaceTime by accident?

Sam Guertin: Yeah.

Craig Rosenberg: oh man. I, I, I, yeah, I'm gonna be spending the rest of the night, uh, pocket face on people. Okay. So, um. So that was, that was amazing. And I feel like it's the tip of the iceberg. So, [00:47:00] I learned a lot. I think gonna have to have you back.

Jordan Crawford: sure, sure. I mean, if we do this again, we, we can just talk with it. Yeah, we can just talk with AI next time. I mean, I do this live for people where it's like we just take a brand, we dissect it, like say great, like let's go take a brand. Like what, what do you want? How do we get to this? How do we use AI to like figure out this segment?

Like do deep research. Um, it's like not crazy hard. Once you have this framework. Um, I have a course who to target and what to say That's. That's upcoming. It's a thousand bucks. Um, it'll be released I think, next week. Um, but it really tries to walk you through this methodology, like walk you from the past.

And then there's a bunch of like really cool prompts now, and with the, the very latest round of models, Gemini 2.5 experimental like this. A lot of the stuff that I'm talking used to take like 50, 60 prompts like in October, November, December. Like, I can do my contract, I can do my, my scopes in two, two prompts now.

Um, uh, and so [00:48:00] they're getting much, much, much better.

Craig Rosenberg: Man. I know. See, you can see where my midlife crisis comes from. The game's changing over the course of months, you know,

Matt Amundson: Yeah. Yeah, I mean, we were having conversations with people about this six months ago, and it was a totally, totally different game. And it sounds like, uh, Jordan, you were having conversations with customers six months ago and it was a different game. So this stuff is evolving so fast. Like, how are you, like, I, I know we're, we're almost up on time here, but like, how are you staying up to date?

How are you staying in front of the rest of the market as people start to come clamoring into the space?

Jordan Crawford: Um. I'm not, uh, what I'm doing is I'm just talking with the models that chat. NN never in my life has there been a more intimidating box than the chat TBT, the Claude, the Gemini entry box because it, for the first time in, in, um, in our lifetimes, uh, and maybe human [00:49:00] history, we are the limiters.

Like the models will of course, improve, but.

I have like four paragraph long prompts to ai. And, uh, and I still am the limiter. There are more creative ways that I could be using these things to get more creative outputs with less work. And of course there is a translation problem. But, um, the thing, the single thing that we need to learn is just how to talk with these things.

And the best way to do that is just to get your reps in, like just go talk with them and, and move beyond questions of like. will this bike fit in my van? Like, or like, or how much is a Popsicle, right? It's like, no, you're like, I have a Toyota Tercel 1996. Um, I have a huffy 1 25 CEC bike.

I've got three kids. Um, we're trying to go to Tahoe and I wanna make sure, can all these things fit in the car? Um, and will we have a good time? Where should we stop on the way? [00:50:00] And what are the things that I'm not thinking about that could make my life miserable on the trip? Like, that should be your prompt, right?

Like, and it can do that. Um, and it's only gonna get better at doing that. So I don't really, I, I try to consume by creation, um, and, and let the models teach me.

Matt Amundson: Sweet golly.

Craig Rosenberg: it. By the way, should I pocket dial one of the AI platforms and just see what happens?

Jordan Crawford: That would be a great way. I mean, I think if you called sesame, it's this sesame is really, really Just put it right. Just

Matt Amundson: put it, Yeah.

Craig Rosenberg: Just keep it open. All right, man. Well, that, that was excellent. I appreciate

Jordan Crawford: Thanks for having me on.

Craig Rosenberg: that's a great show. We're gonna have you on, you're gonna do some, some, I think with my portfolio as well. So I'm just excited to start

Jordan Crawford: Okay. Thanks for having me.

Craig Rosenberg: watching your videos and it's just great stuff.

Jordan Crawford: Yeah.

Craig Rosenberg: on like, being a revolutionary

Jordan Crawford: Oh, thank you. Yeah. very grateful. Thanks

Matt Amundson: And thanks for joining us man. This was fan. Fantastic. Thanks so much,

Jordan Crawford: [00:51:00] amazing.

Craig Rosenberg: All right. That's the transaction.