Transcription
Hey everyone, we're a little early but good to see all some of you here already.
Others, J Karan Leonard, yes Thea, cool! Thank you for joining on a Tuesday. Um, I know, I mean the fact that you're giving us your time is fantastic, so it'll be great.
I want to make sure you get as much value out of this session. I mean, we have run lots of experiments. We'll essentially get started obviously by 8:05, but if you want to start introducing yourself, you could!
Maybe you could start with your name, maybe like which company that you're part of, you know, and like what maybe GTM strategy that you're essentially trying to crack, right? Is there a specific channel? Is there a specific, um, is it the overarching piece that you're trying to crack? I think that might help for you to cover.
So do all of you have chat access? I'm hoping you can.
Okay, awesome! Wait, did I not type this to everybody?
Okay, hang on. There we go, let's do this! Sweet, we already have 19 people. Good to see! Let me share my screen and keep it also.
Meanwhile, that would be help. Just one more thing, I don't know if somebody from Liza is here. Is it possible to see? I'd love to see if everyone's kind of names panel is there, so if they want to keep the video on, they can. I think that'll be cool.
But yes, welcome officially to the webinar! Thank you so much for giving us your time. If you're logging in from India, tomorrow is a holiday, so maybe you can get your drink and listen to this.
I definitely don't want to make this a session which is, you know, if it feels like college or a boring webinar where you're going to just sit down and think about, “Oh my C, like I'm just getting all this influx of information and then I have to receive it.”
So let's keep it super interactive. Let's definitely hear what stage you're in. Please definitely introduce yourself, company name, and what your essential go-to-market problem is that you're trying to solve, whether that's from a sales standpoint or whether that's from a marketing standpoint.
I'd love to answer as many questions along the way.
So, Duck, welcome! Joe Allan, kha cik, thank you for coming! We have Vishal. Great greetings from London. My name is Vishal, currently working as adviser for Pack Ventures.
Okay, awesome! Thank you for joining, Vishal. It must be what, 3:30 p.m. London time? But cool! So, Akilan, good to see you! Looking to start with GTM, okay, as influencer marketing and paid.
Okay, yes, let the questions keep coming in. So I want to make sure it's as contextual to you folks.
Obviously, I made a deck, and I think the deck I’ll walk you through essentially what the deck covers in a minute or so. But anything that you want me to cover specifically, I think that might be helpful.
Gopi Krishna, awesome! I don't know if this is the same Gopi Krishna from my high school. If it is, awesome! Thank you for being here.
We have Phil. Lovely, this is becoming quite international and diverse, so thank you for joining.
Great! So cool! I mean, we'll wait another minute and then we'll officially get started. Any questions?
Awesome! Good to see you, Gopi! Sweet! It might be good to hear also what time it is on your end.
It's 8:00 p.m. Indian time. I think some of you who are—if you're logging in from the US, Gopi, are you? I'm assuming you're central time, so that might be 9:30 p.m. Sorry, 9:30 a.m. for you.
Cool! So let me just walk you through. Why don't we get started? We'll have some fun with this, and please feel free to ask me questions along the way. I have a window just to look at your questions and then we'll take it from there.
Yeah, so just to give you some context, we already actually have 4 million contracted ARR, 2 million realized ARR.
I think a little bit about Lier. I think Lier comes from the word analyzer. We were essentially a data analyst product where, think of it as conversational intelligence. Right? You could upload your Excel files, your CSVs and then have conversations like data GPD.
And I think when Shiva started the company, he started selling this in April 2023, and then when he started talking to potential customers at the enterprise level, I think the number one piece kind of which started coming about was, “Listen, we TR care about data privacy!”
Right? How do we make sure the data doesn't leave our environment or our hosting? Right? So that was one.
And then I think the other thing that I think Shiva realized while trying to build ISO was it was super hard, or there was a steep learning curve with Lang chain to build agents from scratch, right?
So that's when we decided to build an agent framework. And I think a way to think about the agent framework—I have a non-technical way to talk about this, and then there's a technical way to think about it as well.
So the framework is essentially an architecture, right? So we have Django framework for backend, your Flutter for building mobile apps, which is essentially architecture, your Unreal Engine in gaming, right? These are all frameworks to build applications.
It's also a platform, right? So similarly I sometimes even compare it to WordPress. Right? In the late 90s, when people were trying to go online, everyone wanted to build a website.
And when people were trying to build a website, there were only a small set of developers that were truly able to like build custom code to get the website up and running. And then over a period of time, what's happened is today, anyone can essentially build a website from scratch, right? It's drag and drop. You have customized themes, you have templates for every industry that you name, right?
So I think that's fundamentally something to think about, which is, um, today everyone's trying to kind of build an agent and people are trying to become AI powered, and can we essentially be the platform for you to accelerate the adoption of AI?
Okay. For those of you who are not very familiar with—oops, what is going on? There we go!—for those of you who are not very familiar with Lier, this is what our website looks like: build reliable, private, and self-learning AI agents.
And I think there are two ways in which you can adopt Lier. Right? You want pre-built. Our flagship product is essentially our AI SR where it can automate sales in some sense, right?
So send out customized emails and have conversations. We have AI marketer, which is Scard. We have AI HR that we're launching along the way, which is Diane, and then we have custom agents.
Right? So the custom agents you can use a sequence of our agents and then use the professional services team if needed to actually implement these agents and then automate your workflow.
Okay, so far so good? Let me see what the questions are.
I want to know your GTM before funding. What are the challenges you faced?
So by the way, to give you context, okay, so before we actually set some parameters, there was a family and friends round that kicked in before Shiva started.
I think it's important to set the tone because this is the second time Shiva essentially started a company, which was in the cloud space in 2015 called PowerUp Cloud. Grew to 200 people, sold it in 2019 to LTI, and then grew the cloud business, which is essentially the cloud implementation business from 2019 to 2022.
Right? So, it does help that we have some form of distribution or context, massive context. I think there's a strong founder market for it.
And also, by the way, I know Shiva because my first job after undergrad was at Tesco, and that's where we met.
But cool! So, I think let's—why don't we get into this quickly?
So what we'll talk about today is we talk about principles, product under clap, website strategy. I'll break down actually every channel in terms of what we did from a channel standpoint.
Market Maps—I don't know if you've been seeing Market Maps where they kind of talk about top AI tools in X. What's a good way to appear in that?
And I'd love for you to stick around for maybe till the end because I’ll show you how to build an agent also super quickly and then we'll have this deck that will also go on the website; then you'll always have access to this deck.
Right?
Um, so yeah, then we'll talk about our metrics of—we're tracking our tooling stack in some sense, what our GTM team looks like, and I think it's important.
Just a quick sneak peek, which I think is needed, is our GTM team has a UI/UX person who's Leonard, WordPress developer who's Kieran, which I think is not necessarily the most common thing in a GTM team.
Earlier on, a startup, when I call GTM, I'm calling is growth sales. The sales outbound team is separate. This is, let's say, call it—let's say the growth inbound team, right?
Cool! But anything—let's see! We have—how you targeted Dem monster ROI?
Just fill up—I’m just going through if there are any questions specifically that I should answer.
I think Kha had one question when you started Li AG: Did we already have customers from previous products?
So we did have, uh, from the data analyzer product. We actually grew that first, and then that led to the evolution of us building JSON sometime in April, and JSON has been a flagship product which has given us 50% of our ARR.
Okay! So I think in terms of first principles, I think it's important to talk about going deep.
Um, and I think what I mean by that is first—is one of my learnings is you want to be in a space which feels like there's a high perceived barrier to entry.
Um, so if you're doing some form of AI product to AI infra, it makes sense, first of all, to be like, “Let's not build wrappers, let's build products where—let's build platforms!”
Right? You want to be a Zerodha, you want to be an Amazon, you want to be a Coinbase where irrespective of whether you make money on a stock or not, the platform wins.
Right? So you want to be in that space.
I think to give you context, in terms of AI infra, you have trillion-dollar companies in the compute space, right? In like in the chips, uh, let's say Nvidia's here. Then you have the Google, Amazon of the world, which are essentially also trillion-dollar companies.
Um, and then you have the billion-dollar companies in the LM space, right? Anthropic—you have OpenAI, you have Mill, right? So you have that, and then now we're moving further up, which is the deployment and the observability layer.
Right? So first is go deep!
Right? So first of all, you want to go deep into understanding, just not just your users, but the space that you're in.
And I think in terms of analysis, also, I think one example that I want to bring about is when we would think of launching any product, I think we would go really deeper: analyze your comparatives, uh, what was their GTM like, what were the channels that they used.
And you can reverse engineer a lot of this using Google, right? Like you can look at Reddit, you can look at, let's say, LinkedIn— which were the posts that were most engaging.
You'll realize like a product, like let's say Crew AI, is not very active on LinkedIn. Super developer-focused, so Twitter, Reddit is what they're killing it in, right?
For us, we realized with Jason and Scott, it's a very business use case product, um, and we need to dominate LinkedIn, right?
So I think—so I think there are a few things like that we start kicking in U and then yes!
So I think one is fundamentally going deep. Second is be the node.
I think this is one of the mistakes that maybe I have made in the past, I've seen other entrepreneurs also make in the past, is like you want—you start building products in Silos, and then you talk to your customers pretty late.
I think this is standard feedback that you'll get from YC or anybody, which is like obviously like talk to your customers.
But I think one thing that we did relatively well maybe at Lier is we made sure we were talking to customers, we were pitching to customers, no matter what! Every week we're pitching to customers, we're talking to—we're pitching to VCs, talking to VCs, getting feedback.
So VCs will give you market kind of info; customers will be closest to revenue, so they'll give you the want, right?
Then looking at Twitter and getting the pulse of what your early adopters and your influencers are saying, right?
And influencers, I mean, hey, this is the latest tool that's coming out. How people are responding to it, right?
So, and then also, so you want to be the Nord; you want to make sure you're getting as much information coming into you so that you can create pattern thinking.
Right? You can form mental models that can help you decide the direction to go in.
Uh, and I think you can't—it can't be one form of data point or one stream of information coming to you.
All right?
Cool!
So then, be the node, be bold!
I think there's so much noise today. You want to be in the opposite ends of the spectrum; you want to be unconventional.
I think one one thing which is important from a GTM standpoint is, um, I think Shiva's always asked us: “Let this year be the year that we be bold!”
Right? And bold can be putting yourself out there, thinking big.
Right? If you're hesitant to send that outbound email out, let's go! Right?
Like you want to pitch to your first degree that you haven't spoken to in 10 years? Like pitch! Right?
Like don't hold back!
Like this is—this is one of the biggest waves that we'll see, and if we fail, we fail! Right?
But no, like we should fail—not because of like lack of trying in some sense, right?
So, so be bold and then run lots of experiments!
So we currently run—we're trying to run one experiment a month, but I think we run a lot more experiments early on, um, which is we basically create the product in two weeks or create the experiment in two weeks, and then distribution for that in two weeks.
Okay?
Uh, product positioning, comparative analysis—I think this, I don't know whether you can see too clearly, uh, but what I wanted to show you here was, I think what we start seeing in the market is—I don't know if how many if you're seeing this documentary right now?
Uh, on WWF which is on Netflix, which talks about just Vince McMahon coming out and then they were super dominant in the 80s, and then they went through this phase where WCW was their number one comparative.
Right?
And just to give you context, WWF was 80 times bigger almost than WCW in terms of just market share.
Uh, but I think WCW had deep pockets, so what they did was: we'll be the antithesis of WWF!
Right? WWF is recorded, we play live! WWF is over the top with all the characters, we will cater more to like the audience that looks more like the standard middle-aged person in the US.
Right?
Then, um, yeah!
And I think there was a lot around—and I think with Liza, I think the example is like with SAS, there's throttling, which is saying, “Okay, you start paying for usage,” we like, “Hey, can we be flat pricing?”
Right? People are saying, “Okay, data sits maybe sometimes on the side of, uh, in terms of a SAS company—say it’s on their server—hey, can it sit on our server?”
Right? So like, can we essentially be the antithesis in some sense, right?
So maybe you want to think about where you fall in the spectrum in terms of positioning. I'm bringing this up as you get the principles, product positioning right.
I think then comes the GTM strategy of like how do we actually expand on this?
Right?
Um, any questions so far before we jump into anything?
Will the recording be shared? Yes, I think recording will be shared with the folks.
Want to know your GTM before funding that, I think I already answered.
Cool! Okay, let's continue. Everything good so far?
If you can get a yes.
Okay, awesome!
Okay, lovely! Great!
So I think one of the first things for us maybe is I think that worked was potentially a thunder clap.
Um, I mean, I'm just going to show you, but essentially a thunder clap—what a thunder clap means is you want to launch it like a movie. Imagine you have like—if seven of us came together and clapped, it just sounds like a louder clap, and that's essentially a thunder clap!
Right?
Um, so one of the things that we did was, once we identified like JSON was going to be a flagship product, we said, “Can we activate all our channels simultaneously in a one-month period?”
Right?
So I think there are things that you can do which are fast and cheap.
Uh, and there are some interesting frameworks out there. Right?
I think there's one framework which the R2B guys came out of—this was in one of the webinars—like you have fast and you have fast and slow in, let's say, the Y-axis and you have cheap and expensive on the x-axis here, right?
And I think what you want to do and what we did was we activated outbound and inbound.
Like if I could maybe play this video—yeah, there's like some background noise of Coldplay—but uh, point being if you look at it, we did marketplaces, we did affiliate, we did massive content creation at scale, which is top, middle, bottom of the funnel.
We ran performance, we did influencers, we did webinars, we did product hunt. Everyone from the team simultaneously posted; we got partners, we did the newsletter—imagine you went all guns blazing at the same time!
Right?
Um, and then we also did outbound, right? You're sending emails, we used some LinkedIn outbound tools as well to send out messages—we were part of lots of groups, we sent that out, and then you have tracking for everything, right?
Uh, I think there's some questions.
Uh, yeah. So in terms of the GTM team, I think this is important—like it is exhausting because you're trying to launch a product, uh, and we are playing with a small budget.
Right?
Like I think we have at no point in time spent more than—actually maybe for Scott there was one point in time where we spent more than $3,000, but otherwise per month we have not spent more than $3,000.
And it was exhausting!
Like I think the way to think about it is you break it down—like you can see almost in my screen, right? You have week one, week two, week three, week four, which is a part of column G, and then we have a dashboard that we measured how what were the demos that we were getting on a daily basis.
And a few things to think about in terms of influencers, um, that you have some AI-focused influencers; you have some niche influencers, let's say.
You have influencers specifically for sales, right? For us, it was easier to get in touch with influencers that were AI influencers, and there are influencers in Instagram, there are on LinkedIn, there's some for Twitter—they're good for branding.
I don't know whether they are truly good for lead conversion—think of AI as like amplifying what you're doing, right?
And you need a lot of people talking about it to finally lead to demo conversions, but I think with influencers, one thing to think about is it makes sense for you to go with folks on the top right.
There, someone with 2 million followers, it makes sense for you to spend over-index on that, spend $1,500 if needed on that one influencer because there's usually a trickle down effect from that individual, right?
And if you're doing any form of like influence or conversations, like don't stop and, “Hey, just do one post for me!” Can you do something which is post plus email plus X over a three-day period?
Right?
And I think what you'll realize is to acquire an initial set of customers, we were definitely—I think outbound with some form of help, but social, like LinkedIn was—we do—we try to dominate LinkedIn.
Like there was a lot of activity around LinkedIn that we tried to do, right?
Um, so Thunderclap is definitely one strategy that I would try—like three weeks before start preparing and then you spend three weeks kind of marketing, right?
Um, sorry, it just happened.
Okay, the next thing is website strategy. I think this is something that—wait, why am I—hang on.
Sorry, one minute.
Okay, okay! So the next thing about website strategy is I pulled this out from Wayback Machine.
Uh, I don't know how you feel about—but we have actually tweaked our messaging quite a bit over the last eight months, eight, nine months.
We started with single-task agents, which were chat and knowledge search. Then we evolved into, uh, Jason and Scott, but I think our—sure there's always been a centralized theme around, you know, trying to build private. You wanted to be autonomous, you know, high accuracy.
See? I think those—like trying to car towards enterprise, and I think our messaging initially was quite technical.
And I think the way to think about it is—and it's completely fine if it's technical and evolves into a broader messaging as you move over a period of time.
Because I think there's a normal distribution which adoption takes place.
Our messaging initially was very focused towards CIOs. I think today, it's become a little more around, “Oh, someone looking to become AI powered?”
“Oh, this is how you can essentially implement Lier!” Right?
“Oh, you can do custom, or you can get these pre-built agents.”
So the language is slowly moving, and I feel as we move, we're still in that journey, right?
Still in that journey where, let's assume six months from now or 12 months from now, we start talking about, “Hey, how we are trying to in some sense accelerate the adoption of AI?” It could be that, right?
We're trying to help companies achieve organizational general intelligence, right?
So we'll get broader, maybe.
I don't necessarily recommend, uh, changing or trading your website as often more because it’s effort, right?
You got to kind of completely change your brand.
I think one of the things that we have learned is maybe like get your branding and setup early on, right?
Like truly get it out of the way and then maintain that tone.
Maybe have a vision to say, “I'll most probably build these kinds of products in the future,” and then we don't have to go back, right?
And I think that may be super helpful—like once you have these kinds of like a framework on how you want to do this.
Um, but yeah, I mean, we're hoping like this kind of like evolves over a period of time, right?
Cool!
So let's talk about channel strategy. Let me see how we would time.
Okay, we're almost halfway through, which is not bad!
Cool! [Music]
Uh, cool! I don't know how you feel about this. I'm not going to date this, but if I had to kind of talk about, uh, we get—uh, this was at one point in time a snapshot of, uh, info.
I think today we get—it’s a breakdown. I think we've been able to build some form of brand over the last few months that there's a lot of direct traffic kicking in.
Uh, and it's definitely a combination of offline and online, winning some events, VCs featuring us.
Like we got featured in the A16Z at Market Map—there was a funding announcement. I think Shiva put out a post saying, “Hey, you know what? We reached X ARR.”
Like, uh, how we hit our first 1 million ARR. How we hit our first 3 million, right?
So I think those things have helped, but, uh, today I would say the three main channels for us is definitely organic search, it's ads, and it's direct.
Right?
And then it trickles into a lot of the other things: YouTube, Twitter, there is product—very actually, by the way, for product hunt.
I'll get to product hunt as well, but, uh, my recommendation, product hunt—it's a great place to launch, get feedback, but from a lead standpoint, you don't really get as much at product hunt, right?
And I think, uh, I've seen startups over-index on product hunt and I'm not sure whether it's the right strategy from a place of zero to one.
Like it's a good place to do customer feedback loops, right?
Uh, yeah! So I would say referrals—I would say a lot of enterprise customers or good customers have come because of first-degree network.
So I would say pick a channel. I think if they were slow and, uh, if there was a channel that you could dominate, I would personally say it’s personal brand, right?
Like we got a lot of demos, I remember at one point in time we got close to 70 demos in one weekend off of Shiva's post, and then we've been amplifying that post using performance, right?
So stuff like that is important.
Uh, this is the post! As you can see, we got 3 million contracted ARR! Uh, 1500 likely, so this was huge for us, and this is how I would break down actually your LinkedIn strategy, right?
So pick LinkedIn, build your personal brand, and I think you also want to pick a lane in your personal brand, right?
Like, for example, if I wanted to build my personal brand, it would be like, “Okay, I want to be known as a GTM person,” so I might talk about—50% content should be around how I—how we are building, how we’re doing GTM.
After, 25% could be GTM of companies in general, and then 25% could be the other stuff, life, personal growth, things like that, right?
Um, so my recommendation would be, is personal brand—you can—there's a lot of automated outbound stuff that you can do with tools, uh, so definitely check them out.
LinkedIn newsletter is actually not a bad channel at all for you to build thought leadership.
We got close to 6,000 subscribers in the newsletter in like less than a couple of months, so, uh, untapped channel.
It's a channel where you may not get too many, like, “Oh, you don’t get 300 likes,” like people's attention span has dropped.
But LinkedIn, in some sense, is pushing out newsletters in a big way, right?
So that's something that gets picked up by other people, and then that is—that's something that gets shared, right?
So that's one.
LinkedIn influencers—worth trying! Definitely worth trying—but you should pick a niche.
I think what we realized was we got a ton of impressions, we got like attribution. I don't think we necessarily got the right audience, but so far, the reach across all channels has been the best with LinkedIn.
Uh, for Jason, actually, when we were with some influencers, Instagram really picked up for us.
Um, we worked with a few, and then—and I told you, right? There’s a trickle-down effect that usually happens with influencers, and so far, like with influencers, we haven't spent much money, right?
So these are not like expensive influencers, so it may make sense for you to work with micro-influencers rather than macro, right?
Or you work with one macro, and then you double down on that specific channel, right?
Uh, downloadable templates, I think that’s an interesting—in terms of running performance!
Today, uh, we still get a lot of traction for our downloadable templates! So to give you context, uh, what we have at the bottom—a hundred plus AI use cases, automations for product managers, cold emails for sales—
We all get so, like AI use cases to show you, right?
Um, like Lier shows up on the first page, right? So we’ve gotten a lot of, uh—and 4,400 searches a month—no joke, right? It’s quite a—so for you, it might be about “Let's just pick 10 keywords that we truly want to rank for,” uh, and then can that become a downloadable template?
And then if people obviously give their business email ID, and then from there you kind of move to the next step, right?
Any questions?
I'm just going to make sure—are there any questions that we need to take?
It's a good time to get some questions going.
Yes, we have tried Medium.
Honestly, Medium has not really—like you know, the one big learning for us is maybe over a period of time, we tried—we went really broad.
We said we’ll do 14 different things, and then we came down to, you know what? Maybe we should do four or five things, right?
We did community, performance, affiliate, partner network, SEO, ads, right? Cold outreach, newsletters, and I think in the Medium context, we did Medium, LinkedIn newsletters, and a blog.
I don’t know, like I feel like Medium is maybe an interesting developer audience to go after, but pick one, and then you want to dominate that, right?
So the other one is regarding Jason—how did you acquire the initial set of clients? What customer segment are you going after as you scale?
Uh, I mean, we're letting inbound decide, right?
So initially, there’s a lot of—we’ve got inbound from across the board. I think our product is definitely suited for mid-market and enterprises.
So like, if you are in that segment which has a sales team, uh, you have at least a 5% sales team to 25% sales team or bigger and you’re looking to scale, it might truly make sense for you to try JSON, right?
I think Carlos, you have a question: how do you connect people who come interested in your 3 million in ARR in less than eight months to your product?
Uh, I don't know what that question specifically means, but like today they come in, they book a demo, and once they book a demo, I think we've been very founded at sales so far.
Uh, Shiva has done a fantastic job in closing a lot of folks early on.
Uh, I think one is assurity and confidence, one is like giving them the big picture, and I think that's interesting.
Um, I think there's another question by AP, if you don't mind guys, sorry, I'm just going to take some of these questions, um, which ICP are you going after?
Isn't it tricky with LinkedIn where you get broad-based interest?
So LinkedIn, again, like I told you, right? LinkedIn for us—for us to get as much inbound going with LinkedIn, right?
So, uh, LinkedIn ads are actually quite expensive, so we minimize spending ads on LinkedIn.
We do boosting of ads—sorry, boosting of posts on LinkedIn!
Right?
Cool! So that's with LinkedIn. Have we tried cold calling?
We actually haven't! I think we’re truly starting our outbound process now.
I think there’s been some level, but I wouldn't say—like it's, I think a lot of the outbound has been via emails, and we've started dogfooding as well, but like, yeah!
Like cold calling, I think it's more a follow-up rather than a call first, but I actually—that's something I’m—I in the current scenario, I may not be too clear.
Okay, Shashir, uh, I think you have a question on what kind of paid marketing cost should you invest and what was the ROI on it?
So let me get—I have some SEO content; let's do one sh. If you can see back SEO, have performance ads, have influencers, so let's cover all of that.
Yes, and we'll talk about the two-member team also! Yes!
So with SEO and content, I told you, SEO accounts for 25% of our traffic today, um, and I think this—this is important.
I have this post as well that you can click on and then check it out, but at one point in time we were creating close to 20 articles a month.
My recommendation is, as much as it worked for us initially, like focus on high-quality amazing content, right?
Instead of doing 20, you could do one, because we end up getting like 10 articles essentially, let’s say, lead to most of our traffic, right?
So you want to pick topics that you really want to go after; let’s say, like, you are in finance—like, what are some of the two or three topics you could do keyword analysis on?
And be like, “Okay, I’m going to go down that route and create definitive guides, create videos, create images.”
Like, AI SDR for example—we rank—if I had to show you—AI SDR, where is that?
Yeah, AI SDR which is the space that we're going after—right? AI SDR is the third section here, which is 1,600 searches a month, which is good for us, right?
So we're trying to make sure we stay there.
Uh, there are a lot of other players as well, but I remember at one point in time we were also—if there are videos that show up for YouTube, we were also showing for the AI Azure video.
Okay!
Um, and I think if you're a two-three-member team, I would almost just be like, “We will do only LinkedIn!”
We will do only LinkedIn, but we will create such fantastic content on LinkedIn that we will try to hit the primary, secondary, tertiary audience only; that’s how I would start.
Right?
You can get your first 10 customers easily from your first degree; you'll get your next 50 maybe from your second and then the tertiary will kick in.
So I don't believe that you need to dominate SEO—do SEO.
I think we picked that path early on.
Uh, I know Shiva and I have felt slightly differently about SEO, but I think today it can feel like, “Oh my God! With Chat GPD, with perplexity, with, you know, with obviously Gemini also with its results: like is SEO dead?”
I think SEO is still showing to be relevant today.
I think now you got to be smart about creating high-quality content, right?
Uh, so yes! So SEO is definitive—guys, downloadable templates, landing pages, comparison pages, top of the funnel, middle, bottom of the funnel content.
We use tools like SCM Rush, which is expensive. And by the way, just to give you context, like this is applicable maybe for a company that has raised a little bit of money.
If you're a one-person, two-person company, you know, like tooling can get expensive, right?
Like for example, Rush is $300 a month—like if I was a one-person company, like why would I spend $300 a month?
So I would maybe play that game a little differently, which is can I get an accelerator which gives me these credits or discounts, right?
Can I work with friends who can essentially give me some form of tooling access if needed, right?
So I think that's something to think about as well.
So I think that’s SEO.
Uh, next is performance. I think with performance we spend $6 to acquire a lead, we've been spending—you can see, I mean these are essentially last week; I haven't put the click-through rates and costs, but we—5 to 6% is our click-through rates—we get—we spend $50 to $60 a lead.
Um, yeah!
And I think that's decent, but we need to get better, right?
And I want to—and I just want to set expectations, which is this GTM distribution motion does not work if the timing is not good and there's a definitive need of the market, right?
Like SDR started becoming hot; like people were like, “Oh, what's the first thing that I can anchor against revenue which might have maximum impact, which can have repetitiveness that I can replace the repetitiveness of humans that was essentially your SDR?”
So sales development representatives, right?
Right?
Uh, cool! So that’s performance.
With influencers, I would say do it as part of a thunder clap journey.
It might make sense instead of you having to interact with a bunch of influencers—might make sense for you to interact with one influencer that can coordinate with everybody. Surprisingly, they're all well-connected, right?
So I think that's one thing which I would recommend.
Instagram and LinkedIn are interesting.
Um, yeah!
But maybe you should try both and I think in terms of frequency, like—let's say you end up working with 10.
Like we tried both. It took two days or three days and we put like 12 influencers behind it and then we were like, “Oh, for a week we got 12 influencers to push it out,” and then we gave a gap and then we did 12 influencers again, right?
So maybe that's something to think about as well!
Uh, yeah!
Then I think, yeah, okay some questions here, let me just take some questions around influencers.
So what is the CPL from influencers? How does the economics around influencers work?
Objectively, for us, the number one influencer that's worked for us is Generator A with 4 million followers.
They have a newsletter, um, they have a strong LinkedIn presence, they also have a podcast.
Expensive, podcast is expensive, um, but they are magnanimous!
Actually, one of the few influencers by magnanimous, which is like, “Hey, you want X? Okay, you know what we’ll do this and we’ll add this other piece, right?”
Um, now, CPL from influencers—I think you got to figure out what your CPL benchmark is, right?
Like they say your CAC needs to be less than three times LTV, or your three times CAC should be less than LTV.
Like if our LTV is, let's say, $40,000—I mean, it's not—but let's say it's $40,000, you don't want to spend more than $40,000 divided by three, which would be $13,000 to acquire a customer, right?
Which is a lot!
But I think the way to think about it is with influencers, think of them as a branding play.
It makes sense for you to use—to use the content to run performance.
You'll get better ROI, right? Because if you can get that content, then you can use that with performance to target your specific audience and get better click-through rates because they are better at creating content!
Right?
By the way, YouTube influencer is really expensive.
Jo Allen, I think something to think about is like there are some influencers we have spoken to, which is out of control.
Like I see the value that they bring, but like to spend—somebody told us $115,000 to make a video or $20,000 to make a video for if you have 230,000 followers.
So, yeah!
So I mean, I wouldn't say it is like that's something like you should figure a way out where you have some trickle-down effect where you have created virality.
Then an influencer, like, you know what? I want to cover this person, right?
So maybe a good unconventional way to do this also is can you run performance to influencers, right?
Pick their names, pick their email addresses, add them to a custom audience, run ads on them, right? Let them try it out, and then you take it from them.
The other piece that I want to add is once you actually use a few influencers, you get a lot of inbound. Influencers are pretty good at coming back to you and saying, “Hey, listen, I want to promote your post; can we have a conversation?”
Uh, yeah!
I think we are very pro-sales and marketing as a company.
Uh, our teams are also set up like that.
We have a strong, obviously, product engineering team, but I believe that we are in an era today that you should sell first, you should market first, even if you haven't achieved product-market fit.
Because product-market fit could potentially take time—you show them what the aspirational view could look like, and then from there you kind of trickle into—or make them as design partners along the way, right?
Like if you have a fantastic product, people are willing to wait.
Uh, cool! Sorry!
Yeah, affiliates, at one point in time, 7.5% of our traffic was affiliate and, uh, I think the sweet spot for affiliates was for us newsletters—not influencers.
Influencers want instant gratification. They want to get paid straight off the bat.
Uh, affiliates are willing to pay me per lead or—and marketplaces. So those are two things I would recommend.
Product Hunt, um, Scott actually, which is an AI autonomous marketer, did really well on product hunt; it was the fourth product of the day.
We didn't spend a dime on product hunt!
You can see some of the products that we've launched—Kathy, Scott, Jason, Lier itself—and we have used a hunter like today at a point where you don't have to use a hunter to launch.
But it does make sense to use a hunter; I just—I don't know; it feels like initially there's some distribution that is in a hunter.
And like, uh, you should never pay anyone to hunt; you should never accept money from a product hunt standpoint.
There's a lot of conversations and groups that you can be part of, which is like, “Hey, help me get an upvote?”
You know, like, “Hey, can you, uh, support us?”
I truly believe you hit a point right now; product Hunt is good for validation like leads.
I remember when we did Scott, we got three leads from product hunt as potential demos, right?
Where is three compared to all the—and there are companies out there which are spending money to get a lot of this going.
So I know this is maybe like polarizing thoughts, but I don't know—I’m not a big fan of like spending money and like try to gamify or do anything related to product hunt.
Uh, get their early customers and then kind of—like get their early feedback and then move on, right?
Marketplaces, I think for us, we have, uh, we tried Future Pedia, we tried—there’s an AI for that—everything AI was something that I used to run.
We’ve gotten very little leads or demo requests from marketplaces.
With these marketplaces, uh, but we are excited about obviously moving more towards enterprise, which is G2, capital software device.
I think that’s something we're exploring, but that's expensive, so we want to be strategic about how we use that.
But good place to list, good place to get a backlink and move forward from there, right?
Uh, outbound, I think this has generally been a stack we've used: a pool, clay, instantly.
Uh, we've started dogfooding Jason; we use sales navigator, uh, used nurturing and downloadable templates.
I think downloading emails from Apollo, um, enriching using clay, sending it on instantly.
Uh, we also now have integration with HubSpot, which I think makes sense to start, but otherwise we are essentially Google Sheets, right?
Okay!
Uh, podcast! It is—it’s not a lead-gen machine, like, you know? I used to be pretty excited about podcasts, “Hey, you know what? It would be a good foot in the door, talk to CIO, and like, you know, they could potentially convert to customers.”
I think with podcasts, there are two ways to look at it: one is it is in the slow and expensive bucket.
Uh, it’s about access, right? So like your first 25-50 episodes may not make any difference to anybody's life, right?
But sure, you can pre-purpose that content.
Like Shiva and Rishab actually run a really good podcast in terms of just stuff that's happening around the world in terms of the technical side of things when it comes to AI, right?
It's good, and if you talk to both Rishab and Shiva, they will tell you, “Hey, it’s good!” Like I mean, it's a good foot in the door with someone on the enterprise side.
It's helped them stay up-to-date with things, right?
Uh, because they kind—and it's kind of helping them formulate—like the best way for you to actually learn is you look at—look through the content, you try to teach, right?
Do you share and then you try to build, right?
Um, yeah!
Okay, cool!
So foot in the door!
And then obviously you can repurpose channels and stuff.
Uh, community! Community is a slow burn for us!
Uh, Slack or Discord, I would say use Discord for any developer community—it’s better for gamification, better for levels, uh, better for automations.
But I think we went with Slack because we want to go mid-market enterprise and most of them will be using Slack, right?
So we technically have two communities; we're trying to build the Slack community K AI community, which is a bigger play for us.
Imagine we have agent studio coming out—anybody should be able to build an agent with that studio!
Actually, that's something we're supremely excited about, right?
So, uh, we'll talk about that soon enough, but with that, what will happen is we'll have hopefully a flourishing community that will sit in Discord.
But I think with Slack, it will become about how do we teach people to essentially build agents, right?
Because I think one of Shiva's missions in—from his personal story—is as much as he's a computer science graduate from undergrad, he felt like he learned a lot of the stuff himself—very DIY—in January of 2023, right?
So it's a way of kind of giving back! But it's also like innately, hey, use Lier to build at the core of it, right?
Cool!
Um, cool!
So in terms of size of a marketing team, Carlos, we’re almost today 10 people.
Um, UI/UX, WordPress, SEO, brand head, and content performance—we have someone for affiliate and HubSpot!
Um, I’m forgetting social—right? We have design, we have chief of staff, we have content, we have a content writer, right?
So I think actually almost 11 including me.
And the team is expanding!
The team is expanding!
Um, and I think it makes sense for you to start small; pick what channels are working and then from there double down.
So I think we're doubling down on what essentially will be the verticals we’ll go after, right?
Uh, actually, by the way, we have 35-40 people, 35 people at Lier right now!
New products, I think that’s something that you should always try to launch.
Even if it is, um, even if the capabilities are minimal, like I think you should try to figure—because it's a good lead-gen for us!
Like magic prompts—we figured a way how to embed with an agent studio.
YCB is something the team did with Shiva which was, “Okay, like if you had any question—this was essentially built on Lier where all the information about, uh, which I think through Paul Graham's articles and Creator on YC was there—was kind of running this, which was added as context to this bot, and then you could ask questions around like how do I get to YC? What are the reasons? What will increase the probability of me getting in?”
Right?
All good email lead gen, Lier lead, right?
And then we’re trying to do some of these new products—there's some products that's coming out actually over the next few months which we're quite excited about, which is again will continue being free products.
Uh, we also have a course generative AI stack for enterprise leaders, which kind of gives you the AI infra rundown of, “Okay, if I want to understand what AI infra startups are up to and where they can all fit in?”
You can go check out the course!
Cool!
Any questions?
I'm going to take a quick water break!
Cool! We have ten minutes, so I’ll just quickly go over this.
Market maps—a lot of market maps primarily come from VCs, truly, and I think one of the things that we have—maybe you have to create the hype so you start getting inbound.
Like VCs today are essentially doing a sales job also, and they are reaching out to you.
Like you’d be surprised—you’d be surprised how the word spreads if you're building some momentum because they all speak to each other, they all are part of the same WhatsApp group.
All the analyst associates are usually in one group, right?
At least I know the ecosystem; it’s that way, so if you speak to somebody, they'll pass their information on and like, “Hey, you should speak to them.”
And you—you gotta be smart about how many conversations you want to have.
Like it starts getting, uh, exhausting as well.
Exhausting, sorry!
Uh, so yeah!
Market maps—we've been featured on, um, A16Z, we’ve been featured—CB insights, we got featured recently in a few others—HFS research is another one, um, and I think Capgemini came out of the list, right?
So I think that's important—so mostly it's very—it's a good brand play.
I think from a—I don't know, like from a lead—it’s hard to kind of attribute say I got X from this, right?
Uh, tracking and metrics!
We have a weekly deck; we do everything on HubSpot.
Today we have a GTM sheet, uh, we definitely measure leads, demos, opportunities, customers, and ACV.
Um, what you're seeing on the left is like all the pieces that we look at on a weekly basis.
And this is just growth, right? The sales team does its own thing, product team does its own thing, other product engineering team does, right?
Tooling breakdown!
We actually use a lot of tools, and actually—surprisingly, tools are getting expensive—it’s crazy—tools are getting expensive!
Uh, yeah! We use Canvas, Gamma—we’re big fans of Gamma.
We definitely use the deck for deck creation. There’s a lot of repurposing that we got to start doing.
Uh, we use TA for screen share, sometimes we use Loom as well.
But, uh, any form of like videos you're creating for courses—we do use that.
Um, some of our agents are powered by perplexity and 11 Labs, maybe affiliates.
We were actually—we were using Rewardful and then we switched to Referral Hero because you get better integration with CRM.
Cool!
Team breakdown!
Actually, that's mostly our engineering team, uh, but that's Faraz, Shas, Sugo, Prine, and Jimmy.
But yes, I think somebody asked me what the team looks like.
In this world looks like, founder at sales, brand and content, design, UI/UX, website, CEO, social, performance.
Cool!
Finally, additional fundraising is, I would say, like just create inbound.
Like don't do the outbound reach; I think what you could do is maybe—obviously there are a lot of email lists, so you could kind of build in public and then send them updates every month.
Uh, but if that is building in parl, I think that's something will work really.
I remember Shiva did that once we hit a certain milestone, and then that led to a trickle-down effect, right?
Of like people say, “Oh, let's chat. This is cool.”
Uh, and I think VCs are also becoming—like I think 1 million is a sweet spot, right?
Cool!
Tool slide!
I think Shiva is asking for the tooling stack, yeah!
Social media—actually, we had to use Hootsuite; we switched to Buffer.
Uh, LinkedIn sales navigator, we use Jason.
We have, uh, Surfer SEO, which is cool on the SEO side.
Yeah!
Okay, cool!
I said, create a bar.
Okay, awesome! So if you have—
I guess just a last sneak peek, which I think may be potentially important: We have four minutes!
Is obviously if anybody wanted to try JSON, Scott, or AI-powered the company, please go check out Lier DoAI!
Let me also maybe potentially put the link here!
Okay, so thank you for coming on today.
If you want to book a demo, please book a demo!
Our sales team, I'm hoping, will get back to you in like 24 hours, right?
But what I want to give you a quick sneak peek of is also the Lier agent studio.
So we’re moving to a platform where anybody should be able to build an agent, and I think that’s super exciting, right?
So definitely go to agent—we haven't done this massive launch yet, but I’d love for you to kind of look at it.
Uh, check it out and, uh, yeah!
And then you can basically go to the studio, create an agent here, set up the environment, right?
Pick an LM—you can use the API key, which I think is still—which has some limits of Pyon Pyon 24, uh, but yeah, you could do that!
Go to modules, go to tools.
There is—it’s actually pretty straightforward. You have any questions?
You can definitely join the Discord channel.
Uh, if Viber, if you're here, could you share the Discord link, please?
I don't know where I was supposed to be here; I don't know where I placed it.
There we go!
Okay, cool!
So we have a Discord channel if anybody wants to join!
And, uh, yeah!
You have our email addresses as well: shiva.un yes CS contracted AR!
Uh, we sign onee contracts!
Cool! Thank you!
Go!
Cool! Thank you, Gopi! Thank you, everyone!
I'll stop sharing my screen, maybe!
Uh, yes! And for the deck, I think we'll send it over email.
And if you need the deck, you can definitely have a look!