Transcript 0:00 Is it possible to hit a million dollars in 18 months? Nah, it's not so hard, right? Well, what about a million dollars a month in 18 months? 0:06 My buddy Matt Espinoza, he's done just that, and he's gonna walk us through his entire AI-powered outbound system, how he sends a million emails a month to his perfect ideal customers without a single human in the loop. 0:17 How he has recruited an army of AI bots, these are not real people, who send hundreds of messages on LinkedIn every single day. 0:25 I'm not sure that everything in this video is legal, but I can tell you, if you do it, you probably will blow up your company. And so this is a personal IPO. Let's get into it. 0:35 Dude, Matt, I appreciate you coming on this podcast, man. Uh, you're a killer in business. I always like to start with the same question, which is, how much revenue did you make last month? Last month I did 1.3 mil. 0:47 In CAD or USD? CAD. And so the CAD to USD, I, I can say cash or, or- That's insane. Yeah. Wow. Yeah. That's crazy. Yeah, I think when you and I met, you were doing maybe, like, 700 or 800K a month. Yeah, yeah. 0:59 It's phenomenal. Yeah, that's right. So I remember when you and I first met, um, you showed me this, like, e- there's cold email, and then there's, like, your version- [laughs]... of cold email and ABM. 1:09 And, like, I remember, dude, yeah- Yeah... you guys are, like, fucking Iron Man with this stuff. Yeah. I'd love to kinda walk through this system that he's built. Yes. Um, are you down to- Cool... 1:19 are you down to walk through that? Uh, I'll give a high level and then I'll, I'll kinda deep dive into it. So the, the way people typically do emails, either, there's sort of two levels that are typically wrong. 1:28 One is they don't send enough volume. If we're going to send over a million emails a month, we have since pretty, pretty much the start of the company. Uh, we've gotten the emails a lot better over time. 1:36 Uh, so typically it's a volume problem. The second case if you do have the volume, uh, the second problem is, like, they're either targeting the wrong people, they're not maximizing every single lead. 1:46 And so our flow is that when we send an email, we essentially have it where the email sent, um, we kinda see a reply, if there's a reply existing. 1:54 Uh, if there's any existence of a reply being sent, we then immediately call them to kinda have a warm dialer on top. Uh, and then our inside reports are really what makes the, the bread and butter. 2:04 And so our offer on the email is typically, "We'll kinda do this thing for you. Uh, and then are you interested if we were to do it to you for free?" It's kind of the general sent. 2:13 They would reply, "Of course, yes, obviously interested." Then we'll actually reply immediately within three seconds or three minutes, uh, and actually just give the value. 2:21 So they see the value as, "Okay, we'll find, like, 100 Reddit threads that are ranking your niche. Can I send you over the list?" 2:26 And they're like, "Yes, I'm interested," we send them the list, like, within three minutes, 'cause it's just an agent responding back with the list we created for them. 2:32 Uh, and then immediately after they respond and we send them the list, we give them a call. And then that's what get, gets the majority of the bookings, and call probably adds additional 20% lift at any point. 2:43 Um- Yeah... then we continue replying if you show any interest until they s- and pretty, until they pretty much die. Uh, [laughs] and so until they say, like, "Stop messaging me," we'll continue the message. 2:53 That's insane. Okay, so a million emails a month, which is, uh, in of itself wild. If I looked at your, your sources of booked calls, right, if it were a pie- Yeah... is cold email by far the number one source? 3:05 Is that 50 calls out of 100, 60 calls out of 100? What would I see? Yeah, so let's pull it up right now, actually. So we have this cool software we've built internally. 3:15 Clover BI's kinda like our, our main source of truth that has, like, pretty much everything connected. So, I mean, I can track from Mercury to Stripe, uh, like, all the sales transcripts, all that cool stuff. 3:23 Um, it's still kind of in V1, so it's not perfect. But if I were to ask, like, do a pie chart, uh, just on percentages, uh, not revenue, for a channel, uh, for all our channels. Uh, send me the output. 3:41 Uh, and then what it'll essentially do is, like, it will just grab all the sc- the contacts. You can see that these are all the providers we use. I don't know if you wanna show all of them. 3:47 But we use, like, a certain person for email, a c- certain person for LinkedIn, uh, so it kinda grabs all the contacts. We also have JeremyAI. I think you know Jeremy. 3:54 Oh, dude, I love- Like a- We use JeremyAI so much, man. It's so good. And we have him plugged into here, and he starts yelling at us. He's like, "Okay, cool." [laughs] "Let me, let me fix this stuff." 4:04 So you can see here, um, just on this alone, uh, we actually book, uh, a g- a good majority is unattributed. Uh, I ask it to remove. Uh, remove. 4:13 So you're booking, uh, basically 300 calls a month, 'cause we're, like, two weeks, you know, three weeks- Sure... into the month. Wow. Sweet. So as s- from just this month alone, uh, we have booked, email is about nine. 4:29 We have a webinar we did from the, from the things a little bit, LinkedIn's a little bit. I think the majority's honestly still ads, but this I would consider mostly cold. 4:37 There's about 50/50, which I think is pretty accurate, and this is also calling is part of cold is your sign of cold. And website, I would say, generally speaking, actually comes from cold. 4:45 'Cause what ends up happening, which is super cool, is that when we send emails, um, to people, uh, they would be like, "Oh, um," like, they've checked the link of the person who sent the email, if that makes sense, or who, who's, like, w- the, the response of the email sent. 4:59 So we actually get majority of the website volume comes from cold alone. It's a little bit of AI search traffic. But yeah, I would say generally speaking, right now it's half cold, half, um, half, um, like, meta ads. 5:11 Yeah. This is awesome. D- have you, um, do you know Bending Spoons at all? Yes, I do. Um, they're in the news right now 'cause they acquired Airtable. 5:19 But I was listening to a podcast with that guy and David Senra, and they were talking about how, I mean, they acquire these companies, and a huge part of the value creation is they have this unified platform where, you know, like, they bought Evernote, and Evernote was- Yeah... 5:33 doing something like $100 billion of revenue when they bought it, now it's doing something like 180, 190. They took it from 350 employees to 20. Why? 5:40 Like, the product manager now can do every single thing an engineer could 'cause their platform. Your thing that you were building that we just looked at, I'm getting heavy- Yes... uh, Bending Spoons- [laughs]... vibes. 5:50 Is that, is, is that fair? Yeah. Yeah, totally. I, I think it's really interesting. I think we've had a couple interesting use cases of it. Like, one is the att- uh, attributed revenue. 5:58 And so what's very interesting is that, uh, we can easily just query, uh, calls or transcripts. Let's say you do a sales call That sales call, you don't have to update the Salesforce at all. 6:08 We'll actually automatically update it for you based on the transcript of the call. 6:12 Um, so we're able to do a bunch of, like, tighter automations we've built in-house, uh, and then all that's kinda being powered on the Clover AI in the back end. 6:18 Uh, and then Jeremy, we can just ask, "Hey, Jeremy, based on my data," it has context on everything now, "like, what do you actually think about it?" 6:24 Versus us always kind of pulling in our own insights, and it's, it's really been helped, uh, a lot in terms of, like, the roadmap of what things to focus on next. Yeah. 6:30 Like, from this, we were able to determine our no-show rate was, like, higher than we'd like. I think it was, like, 40%. 6:36 And then we asked Jeremy, "Hey, based on this context," you have all the context, not just, like, the no-show rate. You have, like, you know, the full funnel. "What do you recommend?" 6:43 And it gave recommendations it would've normally given if you would've just told Jeremy, "Hey, here's 40% show rate," right? 'Cause that doesn't give enough context. 6:49 So, uh, a lot of that stuff has been hu- super helpful. That's super interesting. Yeah, I mean, uh, for context, like, we're focused on no-show rate right now in Crème. Yeah. 7:00 And, like, there's, like, five or six tools you could be using. Like, you could hammer people with emails, with SMS. You can have- Yes, that's right... post book a call, like, breakout videos. 7:10 You can run, like, what's called, like, power content where you're running ads actually retargeting the same person- Yes. We, ads is crazy... 15 times they go off on a call. 7:17 It's 'cause you know the game, and that's, ex- and we're doing all those strategies at the same time. Um- Yeah... and that's what Jeremy also recommended, to do all those four. 7:24 That's cool, and so your, the power of your thing is, like, y- you don't have to even know all that. It's like, yeah, look at no-show rate, and it'll go, "Well- Yeah... look, you're these five things. 7:33 You're not doing any of them. By the way, 'cause I'm connected to Meta, I'm connected to your CRM, how about I just write the emails based on what you're hearing- Exactly, yeah... in your calls?" 7:41 Yeah, so what, what we actually, which is super cool, is that, uh, the underserved left that we kind of determined was industry-based ads. So it'd be, like, ads for legal people, ads for whatever, right? 7:50 And so then within maybe, like, two prompts, 'cause it had the systems connected to Webflow, Wistia, all the sources we had, we just asked it to create landing pages per industry. 7:59 And so we had, like, one for legal, and it created the entire landing page based on the copy that converts and based on the sales transcript. 8:05 So you know, if someone from a legal, uh, you know, s- side of, uh, industry, um, they have these types of objections. 8:11 These objections are now automatically put into the new landing page we have, and we run ads automatically to that new landing page. Uh, and so just kinda like automatic, and we have, like, Jeremy then review the copy. 8:21 We have, like, another copy agent review the copy, so, like, we then know, like, all of it's already optimized from day one. This is epic. I feel like a, a kid in a candy store. 8:28 There's, like, 50 things I wanna try here, but m- maybe why don't we start with cold email. Um- Sure... can you walk me through? 8:35 Like, like, wh- when your people are logging into cold email, like, what, wh- what, what do I see? H- how, can you unpack that for me? Yeah. Yeah. 8:43 Um, I, I don't, I don't have access to the email software itself, but what, what I can say is that generally speaking, like, it's, it's very much a horizontal game. 8:50 Um, so the, the three levers I would say that most people don't do is signal-based. 8:55 Uh, I would say the other big player is, like, multi-channel, uh, and the last player is just, like, following up on the leads, [laughs] which Hermos always talks about. Um, so I'll kind of deep dive each of them. 9:05 So signal-based, I think, is probably the, the biggest untapped thing that either people understand or they don't understand it. 9:10 But essentially, it means that, uh, anyone who has a signal that they should, they, they want your product should automatically be put in a certain type of list and just blast it to. 9:19 So for example, for our case, there's a lot of Reddit posts online talking about... not even Reddit posts, just even, like, LinkedIn posts about Reddit or about AI SEO, right? 9:27 So any single times when you post online, we immediately track them, and we put them into a signal-based category. If someone, you know, leaves their job, and maybe goes to a different position, that's a signal, right? 9:37 If someone, uh, likes your post or follows my company account, that's another signal. So then what we do is we do signals, but then every signal has a ranking. 9:44 So for example, if someone follows my personal account, they probably don't necessarily always want my products. 9:49 That's a lower signal than someone being like, "Hey, I actually want this type of product who actually has a solution," right? That's the highest signal possible. 9:56 So then we have a, that existing list, goes to our best senders. 9:59 Those senders are sending to the signal-based campaigns, and that alone generates probably the majority of our, like, demos and bookings and all that stuff. 10:06 If I looked at, like, the underlying data sources for that, is that you, uh, scraping LinkedIn and doing, like, some kind of, like, website visitor scraper? Like, what sits underneath that? Yeah, as you were saying. 10:19 So there's a couple of external providers you can use. We do this in-house now, uh, just 'cause, like, I personally like to get super tight, 'cause I know signals is the highest, uh, amount of volume. 10:28 Um, so y- Warmly is a popular one, the warmly.ai. There's a couple other ones that do, like, intent-based signals. Um, and it kinda depends on the ones that are most relevant. For example, we don't track job switching. 10:39 Um, but I do know if you're a recruiter, that's obviously huge. 10:42 Uh, you can actually have a lot of, um, platforms that just track certain signals for whatever niche case you want, um, versus building lists off just, like, scraping the industry, which is what ty- people typically do. 10:53 People just scrape Apollo, scrape ListKid, but that's kinda not as useful anymore, uh, because, like, there's no intent behind it. It's just a list, um, behind people. 11:02 Where, where are you getting, like, a, a million emails a month of intent-based data? 11:07 Like, there's a, like, you might tell me, "Hey, like, uh, most of it is LinkedIn," or you might say, like, "Most of it is Reddit, but we're able to scrape Reddit user emails somehow." You know what I mean? Yes. 11:16 Like, what's sitting underneath that? So it, it's not always signal. So the way I typically categorize it is that you would do signals first. So let's say you had a million emails. 11:24 You obviously prioritize the best leads first, so we, we do, um, pretty much all signals to start off with. Any, like, overfill or, like, you know, a- access is then, like, just based on the regular list people do. 11:34 Um, and then I'd say the majority of it is just also just follow-ups. Like, our sequence on follow-up's, like, 22 messages. 11:40 Um, so it continues going, and so what ends up happening is that people from previous months literally get followed up afterwards again. 11:47 Um, and then we do also for certain offers, we do, like, a blast offer, 'cause we already have all these emails. 11:52 So for Black Friday, we'll probably have an offer, and we'll pretty much mass message every single person in the last year. 11:57 So it's then like, you know, at that point, we've probably sent, like, 12 million emails in, like, a, a day or a two, a two to three-day horizon, 'cause all those are now, like, people that were interested or we can just blast a new offering. 12:08 That's insane. And so you have, what? L- like, a thousand inboxes basically that you're sending from? Yeah. Yeah, I don't know the exact number. 12:17 I just know we have, like, a decent amount on the, on the inbox to be able to then send the volume we were trying to send. Makes sense. 12:23 For the non-signal based stuff, um, are you sourcing from Apollo in ListKit, or do you look at other sources just for, like, massive lists of email? Because you, you basically... 12:34 Your businesses, I mean, one is basically automated organic content, and the other is- Right... like, "Hey, we will get you ranked in Reddit." Yeah. 12:41 So you're addressing the market as basically anyone with any kind of business. Yeah, exactly. Um, I would say generally speaking, the service types, um, which is... Well, there's kind of two ways to look at it. 12:52 So email works really good, in my opinion, for, like, SaaS or it'll work for any industry, but, like, SaaS and services are very hard to target on Facebook. So I can't get an enterprise company from Facebook. 13:03 And so because of that, I need to always make sure my emails are focusing on enterprise type folks. So I, I can't target the legals, I can't target anyone else like that, but I leave Facebook for those targetings. 13:14 Um, at least as, like, in terms of priority list, I'll still probably hit them every month, but, like, it won't be as high as priority because I know I'm gonna hit them anywhere else. 13:22 So typically, we, we also just do, like, different C-suites. It's like we don't always do, like, you know, chief executive. 13:27 We, if we're trying to hit HubSpot, we're gonna hit, like, other people underneath to then bring it up to, to leadership. So that also changes a little bit too. Um, so it's ki- it's kind of a mix, I'd say. 13:37 Yeah, fair enough. Okay, so you're, you're sourcing all these leads, you're emailing them, and then, um, could we... Do you have the ability to look at some of the messages you're sending? 13:46 Because the, the copy I imagine- Yeah... is pretty important. I can actually show you, uh, using our LinkedIn stuff. Uh, so that's probably- Yeah... the easiest one I've been kind of testing recently. 13:57 Um, so this is data as of today. So, uh, just, like, we sent about 460 invites, 168 messages, 30 people said they're interested, two bookings. Uh, so we currently have about 24 accounts active. 14:11 Some of them I, I might need to rotate, some of them I might need to reconnect. All that data is kind of currently there on the system. 14:17 Um, but I can show you what's interesting here is the campaigns running, which kinda, like, is more or less mimicked on, um, the LinkedIn side. So one here is that we're scraping pretty much any account post. 14:29 So if someone says, "Hey, by the way," um, there's, there's someone that's interested or there's someone that's, uh, you know, interested in the Reddit stuff, these are the highest intent, which is why they had the highest amount of bookings. 14:40 Um, and then you can see the amount of targets I'm kind of attacking, and then the growth per, per week. The second case here is what I call affinities. So affinities are like direct sales now. 14:48 I can kind of show you what that looks like. So I can go to... 14:51 What I can do is I can essentially target any founder, and these are all the jobs I'm, I'm targeting and the, the sizing and all that stuff, and I can then choose any of these sales rep accounts. 14:59 It'll actually just do a search, and then it'll populate all the leads here, and it will essentially auto evergreen, uh, add the leads on top. 15:06 So then we're essentially adding on a weekly basis, like, all high earning communities. So I'm tracking Hampton, I'm tracking all these other places, uh, that are- I see... 15:13 really good high-face or YPO types, and then attacking those. And I'm going to see if they have lead accelerators. 15:18 So for example, if you look at this one, um, I'm actually tracking anyone who's, you know, been part of, um, part of keyword... Yeah, HFO, Path Foundry, Founder Global, Antler. 15:28 So if you have any of these in your bio, so if you have Neo as your, your new Neo accelerator batch, we'll, we'll have you tracked, and then we'll be able to message you being like, "Hey, by... 15:35 oh my gosh, like congrats on, uh, being accepted into Neo." Right? So then any one of these cases are super high intent because obviously they're... they just have money, they just raised, all that stuff. 15:44 So that's the affinity types. Uh, and the last case I think I'll show you here is, um... So triggered outbound is also interesting. It's the other, like, big type we have. 15:53 So for example, any time that someone is, let's say, raising. So I have these things tracked. Uh, see if we can find it. So I'm tracking, like, we just raised or closed our, our seed round. 16:03 All those are tracked, uh, I think on a daily, uh, on LinkedIn. So we scrape whatever the top post is for that, and we scrape all the people that are either posting about it or engaging with it. 16:14 Uh, and then those are obviously high intent people. So in this case, we're only tracking people who posted it, and as long as it has over 25 reactions, we will just check. In this case, I guess we're checking weekly. 16:23 But, and then that's just... it gets added to a list automatically. So this is, like, what I call, like, almost, like, evergreen list because I'm not really ever adding more to the list. 16:31 We're automatically populating new people from the list every day. I see what you're saying. 16:36 Now, there, there's a world where in the messaging for this, when you actually reach out to people, they all could have the same message, right? And say, "Hey, you know, we've helped-" Yeah "... 16:43 whatever it is, thousands of companies get ranked on Reddit." Are you doing that- Mm-hmm... or are you personalizing based on the list? Yeah. Uh, that's a good question. 16:52 Let me show you, I guess, one of them so you can kind of see the sequence. So typically what we do, which is super sick, is that we'll send an open InMail, which is a free credit on LinkedIn if they are not... 17:02 Uh, like, there's an InMail hack you can do on LinkedIn that we kind of, uh, utilize. Uh, and so what we do here is we kind of send, "Hey, like, my name is blank from Clover. I've been helping these companies. 17:12 Here's the learnings I have." Wait, what's, what's the hack? And I send them- And we, we can cut it if you're- Yes... not allowed to say it, but... I, I can say it, yeah. 17:18 So essentially on, on LinkedIn, there's, uh, three different ways to send. There's open profiles, which essentially is open InMails. There's closed profiles, which is paid InMails. And there's regular connections. 17:29 Most people only do, uh, mass connections, which is actually, like, just, like, two-thirds of the amount of volume you could be sending. So if you actually look at here, today I've sent 418 connections. 17:40 I've sent 42 InMails. Most people don't even send InMails. Um, and then I have about 109 connections left for the rest of the day before I fully max out every single account. 17:49 So you can see that I've maxed out my account today. There's other ones that are currently still gonna be maxed out. Um, so what I would do is that... I can kind of show you how it looks like here. 17:57 Maybe I could pull up a random one. So if someone says, "Sure," like, cool, awesome, um, then what I can do is I can... 18:04 It would just automatically reply with the report, and then this report is just auto-generated, and this kind of acts like a mini VSL. 18:10 So the same way you're able to do that on Facebook, this acts like a mini VSL on, like, our system. So this is kinda a little sim- a little Facebook image we have. 18:17 And it pulls in automatically for that company, which is an athletic company, uh, what- which ones they should be ranking on for Reddit. See, these are all, like, NHL, football, all that stuff. 18:28 Um, and then they typically would respond to that. I can show you one that got booked, actually. So someone's like, "Hey, cool. Uh, my name is... Can I send it?" "Yes, please send it to me here. 18:37 Uh, my team will send it to them shortly. Uh, in the meantime, here's the thing." And then this person, uh... It was actually a real person from my team, who's a cold caller. 18:45 Uh, then let them know that I actually had this lead, um, actually messaged them, and then they actually booked. So then that person is actually tracked as booked, and they have a call with us on Wednesday. 18:54 And that was just, like, an automatic setup, that there wasn't any, like, human involved besides the confirmation at the end. Dude, this is fascinating. 19:01 Oh, so if I unpack some of the things that were required to send that link- Yeah... that you're sending to this person, um, I imagine... 19:09 Well, one, you have to go, like, okay, well, something has to know what their website and company is in the first place- Yeah... to even do the analysis, right? 19:16 'Cause, uh, are you saying an arbitrary number of Reddit opportunities, or is that actually truly, like, the ops that you found? Yeah. It's, it's arbitrary. And is it- Okay. Okay... 19:24 yeah, no one, no one, no one cross-verifies. So I- I would say, like- Yeah... 43, and then like- Okay... whatever. Okay. And then, like, on the report it says whatever else. Okay. That makes sense. Okay. 19:32 So, so you do that as kinda like a Trojan horse to get the response. Yeah. And then they respond. And then how are you getting this to pull actual keywords for this person's company? Yeah. J- just website. 19:44 Um, so as long as the website, and every LinkedIn profile has a website connected, 'cause we're able to identify what company they work at, and it'd pull the website from that. That's cool. 19:52 And, and that's like a, that's like a custom in-house tool you've built, 'cause it has to look at their profile- Yeah... automatically see what their site is, right? I mean, that's what's going on underneath the hood. 20:02 Yes. Yeah? Exactly, yeah. That's right. Interesting. Interesting. Yeah. And then we do the same thing on, on the email side. 20:09 Um, so, like when we do the follow-ups, I just showed you how it looks like on LinkedIn, but, like, it's pretty much the same sequence on, uh, on email. Yeah. What's the... Could you do this for any offer? 20:20 Like, like, what are the things you need to- Yeah... believe about your offer to do the same type of Trojan horse thing that you're doing here? Uh, there has to be a no-brainer offer set. 20:29 So in, in your scenario, um, y- you have to kind of give something that's, like, so valuable. Like, to find threads online that are ranking or whatever, like, this is extremely valuable. 20:38 I've seen a list kit, for example, early days used to do, like, we'll get you 100 leads that are ready to pay right now, and then just be able to export the le- the list kit leads and send it to them. 20:48 Uh, and they're like, "Oh, my gosh, these leads are so good." And then, th- then they would hop on a call, right? So there has to be something that's, like, good enough as an offer set. 20:55 And then once you do have it, we just maximize the, the report. 20:58 So majority of people, like, do it wrong, where if you even looked at the website as I showed before, like, it had the VSL, it had the threads, it had, like... 21:05 There's six different buttons that just do free, free strategy call, right? This free strategy call books better than, like, saying, "Book here," right? 21:12 So, like, all the micro-optimizations over time started to compound, uh, such that people just book directly from the call, uh, from the reporting, I mean, um, over anything else. That's cool. That's pretty fascinating. 21:23 Um, are there... If you did th- Do you do this at all for Echoes? Yes. It's still early, but we, we do the same thing, um, for, like, automatic report. 21:35 For that automatic report, it's just like, here's top 100 videos that are ranking or top 100 videos that you can clone in your niche. Interesting. Okay. 21:44 So you, you, like, ascertain roughly their category of company, and then you can go and scrape common videos that are viewing high- Yeah... 'cause your offer effectively is like, "Hey, we can- Right... 21:54 mass produce organic content for you." Yeah. Exactly, yeah. That's cool. Um, what, what would I need to do to build this kind of scraper for my offer? So for instance, we run- Sure... 22:06 Facebook ads for people who sell info, for people who sell, you know, newsletters- Right... basically. How would you do that for my company? 22:13 I think there's actually a huge amount of value add just to add inbound messages. 22:17 Like, people that are already connected with you probably get, like, 100 connections every week, that people are just interested in what you do. 22:21 The second, uh, you connect with them, if you connected with me on, on LinkedIn, uh, I will automatically reply to you, uh, being like, "Hey, my gosh," like, "my name is Matt. I think your company's cool. 22:30 A couple bullet points about what I do if that's ever useful. I don't wanna pitch you anything like that, but, like, here's what I can be helpful if, if that's ever in the future." 22:38 And I have people that just, like, come back to me after and be like, "Actually, I think, uh, I think what you mentioned before is useful." And then, like, they end up booking a call after. Yeah. 22:46 Are you doing that automatically? For instance, in my company- Yeah... when anybody views our profile or adds me, we're thinking about automating this. I've been skittish about doing any form of LinkedIn automation. 22:57 But there's a real person who always says some version of like, "Hey, you know, saw you viewed my profile. H- how'd you run into me?" And the goal is to start a conversation that we can then- Sure... 23:06 either set to the newsletter or set on a call. Yeah. Yeah. Uh, honestly, it has to be automated for a couple reasons. One is that there's a real problem with lead quality. 23:16 So for example, if I'm a student, I connect to your thing, and they're like, "Hey, where did you find this from?" That person will continue messaging back and forth until you get super annoyed. 23:23 And so our inbound has a filter set on A+ leads, 'cause inbound is, like, so, so tight. And so if you... 23:31 If I'm able to determine, hey, this, you work at a company, that company prints, uh, and, you know, you're not a student and all these additional filters, uh, then we'll actually message back and be like, "Here's the value I can provide you." 23:41 That's pretty interesting. Okay, that makes sense. Um, what other post-close or, like, pre-call... Like, a- anything between somebody- Yeah... 23:50 booking a call and paying you, what other automations would I see if I opened up the hood? Yeah. 23:55 I, I think, you know, an interesting thing, which I think you guys should do as well, is that we have, like, reporting on Slack that's, like, automated. 24:02 So, like, if you guys run the ad stuff, you guys can just, like, automate and report it. Interesting. Just, like, all. Even, like, if you have a win, um- Oh, this is, like, a client account. Okay. Yeah, yeah, yeah. Yeah. 24:13 Wins c- This is cool. This is, like, a really nice branded image here. Yeah. I'm gonna exit out of it. Yeah. And then if we have, like, the plan for the week, for next week Okay, so we've covered cold email. 24:23 We've covered your LinkedIn outbound system. Um, is there anything, like how close are you to the actual sales motion? Like how the calls- Sure... 24:33 are being actually held, uh, platforms you're using to ensure that your sales team is actually ace at these calls? Yeah. So I, I call, uh... 24:43 So we have a, like weekly tracker that's like North Star and then per department. And so generally speaking, there's like three OKRs or like things to track per department. 24:51 So for example, uh, with CMO, it's like demos booked to qualified to like all that stuff. Uh, and then for the North Star, it'd be like, you know, revenue generated this month, NRR, all that stuff. 25:03 Uh, so they're kind of done per department level, and then we have like the high level for the month. Uh, and generally speaking, we have like a theme-ish for, uh, for the month. 25:10 So last month was actually a super fun theme. It was, uh, one-click campaigns. So it was like, can we in one click onboard someone and then like have a campaign running for them? 25:19 Uh, obviously we weren't able to do a single click, but we got pretty close. We got to like four clicks, uh, for Echoes, and I think we're trying to get a little bit tighter for Red Rover now. 25:27 Um, yeah, it's, it's kind of a fun like theme setup we have there too. Cool. Um, so you, you've mentioned like maybe eight or nine use cases just now. 25:36 Like if you could only keep three of these things, like what would you keep? Mm-hmm. Because I, I wanna maybe dive a little deeper into, into three. Sure. Um, 25:43 if you ignore marketing, I would say, uh, health, like notifications of health alerts per customer is very important. 25:50 So if someone's not healthy, the account's not healthy, which defined as the renewal calls is bad or like they message me on Slack, that's bad. 25:57 Or if the, literally just the performance of the campaign is doing bad, that should be flagged immediately. 'Cause, uh, you should only be... CS people should only be focusing on bad accounts. 26:06 Uh, until there, those are solved, don't even bother like making better accounts better. Um, so then that's how we look at it that way. That's like one automation. I think that's probably huge. 26:14 The second case for sales is more of like, um, knowing when like calls are... Like, yeah, I, I can show you a couple cases here. Like, so this is super useful. Like I can just like scroll through this every morning. 26:26 But this is like, uh, our renewal calls or like, you know, check-in calls, right? You can see. So immediately after I can tell these are all good calls, right? All positive. This one's negative, right? Mm-hmm. 26:35 So then I'd, I'd look into it. Why is this negative, right? So what was it- Okay. So right away-... that goes wrong... it's like color-coded. So instantly you can just go- Yeah... 26:42 straight to like what the problem areas are. Exactly, yeah. So most of these are all kind of positive. That's interesting. Just obviously sales calls are not as high. So you can see this is red. 26:50 Ne- near- pretty much never green on the first call. It's mostly reds and yellows. Um, but I can easily tell like why it was in a red. Like this is the reasoning of why they wouldn't buy, right? 26:59 So you can kind of just easily tell like why they're not or good, good or not. So w- we do a version of this where we're like piloting it, where, um, we have, I would say, four separate automations. Um- Okay... 27:12 one scans daily in client Slacks and calls and just pushes- Okay... to our directors channel any problem areas, right? 27:21 Like if a client has said anything that indicates that they might churn, we wanna know instantaneously so we can triage it. 27:27 Uh, um, we then have two separate channels that will scan for anything in calls, Slacks, or emails that is either positive or negative about the service, and we have a negatives channel and a, a positives channel, so if we ever need to go figure out what to go improve or what to go show potential new clients, we have those. 27:45 And then we just started doing one that sounds a little bit like your email speed lead thing, where we're able to track the time between a client sending a message in Slack and us responding. And- I see. Yeah, that's-... 27:59 if it takes too long, it'll flag it, and we've set up the bot where it, the profile picture and the name is our head of operations, so it looks like our head of operations is... 28:09 The, like, if you're a growth marketer, you get a message from a bot that looks like it's our head of operations being like, "Hey, where the hell- Yeah... is this, like, client response?" 28:17 Um, because we found when, when we use the bot, people ignore it because they're like, "Oh, it's the robot AI thing." But if it's- Ah... our head of operations, now they, like, take it very seriously. 28:26 Um, i- is there anything else- Oh, that's true... you're doing like that? That's pretty cool. I like that. And so does it, like, is it answering the question or does it just get angry? Um, [laughs] just get angry. 28:36 It's, uh, it won't answer the question for the client. Uh, it will push the growth marketer to be like, "Hey, it's, it's been, uh, 91 minutes. Why haven't you responded?" 28:46 And then it will wait, and then it will ping them again, "Why haven't you responded yet?" So then- Wow... it's the... 28:50 Ultimately, what I wanna try to do is not to have a bot be like a crutch upon which a growth manager can just lean on. Sure. 28:56 It's like you still are accountable and responsible for completing the message and sending something to the client. We just want to remind you to go ahead and do that. Yes. Okay. And then is it always the same message? 29:07 Like, do, do people not catch on at all? [laughs] Um, for now it's been okay. Um, for now it is the same message. 29:14 Hopefully, I mean, if we have to start doing like spin tax to get people to, [laughs] to respond- [laughs]... I think we have, I think it's a people problem and not a, [laughs] not a process problem. Okay. But yeah. 29:24 Yeah. That's true. What's your average retention now? Uh, not, not on a maybe net revenue basis, but just on a per logo basis across the two primary companies. Ooh, let me ask. Well, I forgot to ask earlier. Yeah. 29:37 It, it says, it just says 79.1% retention on logos, uh, for those that have the option to renew, which is important. 29:44 So for example, like let's say you sign a three-month contract, they're obviously gonna stay, like during the three months. Yeah. 29:49 Like you, you shouldn't count those as, uh, as like churned, 'cause like it's impossible for them to churn. So that, that shouldn't inflate the number of people that are not churning. Makes sense to me. Your, I... 29:59 Are you most typically... So I believe you have a three, a six-month, and a 12-month package on Red, uh- Yeah... Rover. Is that, is that right? Yes. We, we do the same on, on all the products, just, uh, upfront cash. 30:10 What have you learned about how to structure that the best? 30:13 In other words, right, you might tell me, "Hey, pretty much everybody will only start with us on a three-month, but we know like a very high percentage, if we nail the first 60 days, between days- Yeah... 30:25 61 and 90, we can get them on a year." Like, like, like walk me through the strategy of how you price- Yeah... and ascend. Yeah. 30:31 So we actually started to do monthlies, but, like, they're, like, so high price that their price anchor almost makes them do three months. 30:38 And those that do monthlies, um, like, at that point it's, like, so profitable [chuckles] that it's like... it's kinda dumb, uh, at least on, on both ends, um, to, like, not accept it. 30:47 Um, so I think we have been more op- open to monthlies recently, and the data is showing right now that there's actually a good amount of cases that turn monthlies directly to 12 months. 30:57 Uh, over, like, three months it turns into 12 months, uh, because it takes a lot of our time. So if you do three months, you have to wait for the three months to have the renewal conversation. 31:05 You're definitely not gonna have it two months in. Um, and so with the monthlies, the renewal conversation happens w- after 30 days. 31:12 And so the monthlies are super interesting, 'cause, like, those are kind of like those trial periods. 31:15 They are really expensive, but, like, if you get really good results within the 30 days, majority of people actually go up to 12 months after. 31:21 And I think we've had majority of our 12 months actually come from that trial period over not, which I think is actually, uh, a coincidence. Interesting. 31:29 And so what, what is the discount percentage from your monthly rate to your annual? So we typically do, which is pretty cool, like, a monthly equivalent. 31:39 So typically against three, three months, 10K is the general price we have. Uh, and then growth increases and then dominance increases. And in sp- just pricing, uh, per the increasing, so I guess it's 23. 31:52 And the trick here is that we actually never show monthly on the tack. Monthly obviously exists. 31:56 The people that really don't wanna pay, you know, 3.3K a month, they wanna pay, let's say, 2K a month, or I guess technically more, they wanna pay, like, 4K a month, but they only wanna do it for one month, then obviously we'll accept it. 32:05 Um, but it kind of just varies because of that. Yeah, that, that makes sense. So, so you're... I mean, you're effectively willing to offer, like, a 50% discount to get somebody to- Yeah... 32:15 uh, do annual as opposed to monthly. Yeah. It's, yeah, it's pretty high. Yeah. Cool. That's baller. Um, any other, like, really sexy tables, graphs, measurement stuff- [chuckles]... 32:29 you've been doing with this chatbot that I should know about? Oh, the most interesting one is, is, is really data enrichment on our existing stuff. So I'll give you example. 32:39 Like, the ones I've been really pushing on recently has been, like... Like, sometimes salespeople just don't put the, the result of the transcript. 32:48 And so what I've asked them to do is, like, "Hey, like, the ones that haven't added a transcript, can you just review their Gong transcript and just update it?" And just updates it. 32:56 Um, but I need to get that automated, but it's not automated yet to the degree I like. Uh, the other case I think is interesting- When, when you say, by the way, like... Are, are you saying, "Hey, we have a salesperson. 33:06 We need them to determine whether the lead was qualified or disqualified"? Yeah. They're not doing it manually, so I need the tool to actually do that on their behalf. Yes. Is that fair? Yeah, exactly. That's right. 33:16 Okay. Okay. Um, I, I think honestly, attribution is probably the biggest thing of the next level, I'd say, um, because essentially that filters out the game we're playing next. 33:25 Like, if we know, like, this ad got to this person, it got to this demo, and, like, it didn't cost us that much, we should spend more. But then there's always, like, so much extra data that isn't clear, right? 33:35 Was it, was it the salesperson that kinda messed up? Was it the person that was ready? Was there a bad SMS? Like, should we change the SMS? 33:40 If we changed, uh, you know, change the SMS, like, differently that way, would that increase the conversion? But SMS should only be for that type of industry. 33:47 Like, if you do that SMS change across every industry, maybe that's actually not a good thing, right? 33:51 So I think segmentations, um, which essentially allows the bot to, like, function, is kind of the next level we've been really thinking about. 'Cause we already have a lot of the operational stuff ready to scale. 34:01 The hard part now is, like, you don't want... It's kind of like Google, right? At Google if they, they change one little pixel, um, and like, this might break the entire thing. So they set so much engineers on one pixel. 34:11 We're at a point, soon if not right now, where, like, we can't just mass change everything. Uh, we have to do, like, so much segmentations. 34:19 Like, even our email campaigns used to be so broad, like kind of the stuff I sh- I showed you recently. It's kind of broad enough that it, it doesn't really matter the lead we're attacking. 34:27 But we wanna focus more on kind of having it, you know, be the next big case where we have a hyper-segmentation across the board. Interesting. So your bets are, like, 34:37 attribution between channels, but maybe equally or more importantly, personalization, like across messaging, across every- Yeah... ICP that you're reaching out to. Yes. Exactly. That's right. 34:49 Will you build that in-house? Yeah. [chuckles] We, we definitely would. [chuckles] Um- Cool... I, I think attribution is just hard to, to t- tap into it. C- Can I ask you two more quick things before I let you go? 34:59 Sure, sure. The company's doing really well. Like, what, um, what do you spend your, your money on personally? Oh. Um, I think I save most of it. I wish I had a more exciting answer. Uh, I don't spend that much. 35:16 Uh, I get maybe, like, nice clothes. Yeah. I have, like, a nice watch. Like one. Yeah. That's about it. Yeah. What about you? I, I spend on food probably. Um- Okay... I don't really want anything that I don't have. 35:33 I mean, I don't, uh... Uh, similar to you, I think, I don't really need a ton of stuff, and I kind of don't wanna spend my time buying things. I ki- I kinda like what I'm doing. 35:46 Like, I like working on the stuff that I'm working on in the business right now. Sure. Okay. That makes sense. Yeah. Um, only other thing, well, what, like, what's your, what's your goal? 35:58 Like, I, I, I can't remember if we talked about this when you and I were hanging out in Austin, but, like, are you... Did you grow up rich? Were you middle class? Like, do you, do you want money? Yeah. 36:07 Like, do you have some, like, insane chip on your shoulder? Like, why, why, why do you do this shit, [chuckles] basically? Yeah. 36:13 I, I told, uh, Jacob I wanna be the youngest billionaire in Canada, or youngest self-made billionaire, which is, I think it's like- Really?... 34 or something like that. Um, so I'm trying to speed run to that. 36:23 How old are you now? 24, so I have 10 years. I got a long- You have 10 years? Yeah. Okay. You think you'll do it? Yeah, I think I'll do it. It's a hard reality where I won't do it, 'cause even if I just... 36:34 Like, I think I will be at 100 mil in, like, three to four years. At that point, like, net worth is like... Like, I also own, like, 46% of the company. Um- Yeah... so it's like, yeah. How, how old is the company? 36:45 The company's doing, uh, a year and three months. A year and three months. Wow. It's a baby, yeah. 15 months old. You're doing, what is this? 16, 17 million a year? Something like that, 16 million a year? It's a little. 36:59 Yeah, a little- Roughly?... yeah, 16-ish. Yeah. Yeah. Um, that's fucking crazy, dude. That's awesome. Um- Yeah... baller. All right, man. Cool. Well, um, I appreciate you doing this. I kept you too long. 37:11 Yeah, it's cool. This is, uh... I appreciate you showing this stuff off. I've always wanted to do a session like this, because I've seen- Yeah... hints of all this software, and- Yeah... 37:19 uh, now I've got some, some cool ideas to go build in at The Seed. You're the man. Um- Absolutely. Best of luck.
