Loading summary
Transcript62 lines
- [00:01]
Human Radfar
I think the AI is the least important part of building an AI enabled service.
- [00:07]
Podcast Host
Welcome back to another episode of Builders. As always, this show is brought to you by Frontlines IO, Silicon Valley's leading B2B podcast production studio. If you're bringing technology to market and want to learn from your peers, we have a library of more than 1200 interviews with Venture backed founders and marketers. Where they talk, all things go to market. Of course, if you want to launch your own podcast, we offer podcasts as a service to more than 80 tech startups. The idea there is very simple. You show up and host and we do everything else. Now, with all that said, let's jump into today's episode. Today our guest is Human Radfar, CEO and founder of Kollective. You sold your first company 34,200 million. Have you found it harder to stay motivated the second time around? I feel like for a lot of founders, I know you're very motivated by fear and wanting money and stability and all this stuff. And then all of a sudden you hit the milestone financially. Then the next time around, you have to learn to be motivated by something different. Have you experienced that?
- [01:03]
Human Radfar
You know, it's difficult to remember. I think we're always contextualizing the past through the lens of the present. Right. So you remember things a little bit differently. But I, what I remember viscerally when I was younger was, you know, a lot of it, there's a, there's an egoic part of it, like you want to create. I'm created by nature. Like I like to play music, I like to do all these things. So there was that core, like I want to build and I want it to be big. Like my sandcastle is to be big. But there was also, you know, fear. And in particular, like 2008, when it hit, you realized like all that money, which you didn't realize psychologically, you started to count on this fake monopoly money they call startup stock could go away. Like it starts making, you know, changes. So I think the relief was great. It didn't really shift my motivation. If anything, I'm fiercer now. I often think, why? Because it's not solely motivated by money. I would love the outcome. But I was just at the conference you were discussing before. The mission matters to me a lot. Like, I care about the customer fundamentally, whatever product we're dealing with, I care about the customer because they're my family. I'm a customer, my brother's a customer,
- [02:04]
Podcast Host
my wife's a customer. I told you this.
- [02:06]
Human Radfar
Yeah, it's Coming out of the woodwork, you know, like my wife's friend's hairstylist is a customer and I know, so it's real people and like this is their life. Like I really understand that and I'm a founder so I get it. But also I care about my team so the downs are there. So I think those are my two motivations and I'm hoping that those inputs yield obviously the great output which is there. But I think that's what drives me.
- [02:28]
Podcast Host
Given the success you've had and then you obviously were doing venture, had a bunch of very successful investments. I'm sure you could have founded any company. Why'd you decide to found Collective? What was it about that opportunity?
- [02:38]
Human Radfar
I think there was a confluence of factors as there are again in anything. One, I mean the general platform I think for any founder is do you still have itch? Right? Like you have to, you have to be something unresolved at a psychological based level and that, that's where you start at that point. And I think what in particular drove me to this direction was I love working with founders. I have been doing it my whole life. I've been a founder, I've been funding them with whatever money I have at the time. And then with Expo, obviously that's what we did. But I really enjoyed the scale of AddThis right. It was with 15 million customers, reached 2 billion unique users. Billions of people would touch and use things that we created. And that level of impact for an ICP that I care about was what I was looking for in this particular problem set. Finance and administration. How do you help people focus on their core business, what they care about and kind of become that finance department, like that autonomous part of the stack that they just say, hey, you know what, they run that, that got me into that scale, but also that ICP and it's a very difficult problem. And I was intrigued. I was like, I think I'm going to keep learning this whole time, so I'll stay interested. So yeah, over the years, interestingly enough, I've become more excited about, more interested, which is painful. You know, at the beginning when everything's academic, you're just like, look, TAM is this big and if we get this percent and whatever, like it's all academic. But now it's real. It's real. It would really hurt if I couldn't work on this.
- [04:09]
Podcast Host
What's the scale that you're at today, right now?
- [04:11]
Human Radfar
We've worked with over 12,000 businesses.
- [04:14]
Podcast Host
When in 2020 did you found it in January and February 2020 or after. Because I feel like there's this pre Covid era and then the post Covid era.
- [04:21]
Human Radfar
You're testing my memory here because it's been a minute. I think it was the fall of 2020 is when we launched, because what we wanted to do is start selling and getting ready for the tax season of 2021. I think that was, at least for me, the real lights on moment. But when we were, you know, the seed closed and we announced it, all of that was, I think earlier obviously, because we needed to fund the company. It was just me and my two co founders and a dream. And it was interesting, I think within a couple weeks of our announcement or we were like leaving the office. I mean, it was an unforgettable moment. We had like an all hands, very small obviously. And I told my co founders and they're Turkish, like from Turkey, like real immigrants. And they're like, nah, it's gonna blow over. You know, like you're taking it too seriously. I mean, to be fair, they hadn't announced lockdowns really. But I was like, I think this is different. I think it's worth us sending people home for two weeks and let's just revisit every two weeks just to be safe. And then two weeks became four and four became infinity. And here we are.
- [05:19]
Podcast Host
Does it surprise you how right you were with this market? It seems now that there's like this boom of solo entrepreneurs and there's this big wave that personally I'm just very excited about that there's gonna be this whole class of previously were working really shitty jobs and now you can go out and just have be a solo entrepreneur and you can build and you can create. And I think it's just a very exciting time to be alive. Are you surprised that you got that this right? Like, did you stumble into this or was this like the plan?
- [05:46]
Human Radfar
One of the things that I discovered when we were entering this market was I think I underestimated how long and how deep the trend had been. Anyway, just to be fair, I think the misconception today is for many investors that were colleagues of mine, is that this is a new thing. And I mean even when we were starting, they were very much on the gig worker. That's what got them awake to the fact that the solar worker and they're, they were excited because they're like gig workers and creators. That's such a small part of it. Like my parents were psychiatrists, both of them were solar operators. Your wife, you know, my Brother, there's lawyers. This has been a market, it's just continuing to uptrend because I mean so many factors, it's just, it's almost like existential. But I think when I started to realize that it was going to become even like more fundamental, if you will. I don't know, I don't. It was like around the time when AI was coming, when it really started doing the math. I can't remember if we discussed this, but you know, there's this very now becoming, I think famous bet with Sam and some friends of the billion dollar business of one. And I remember a year before that we were presenting Collision. I was trying to think of a topic that was like interesting and that I believed in. And I started like seeing some exits and looking. I'm like, well if you just look at the productivity curve and you look at what AI is doing, it's kind of inevitable that there will be mathematically this business. And so like I presented on it and it wasn't super interesting, I don't think to people, honestly it was 20, 23. I mean I thought it was interesting, but I was like, here's the curve, here's what's going to happen. But what I left with when I did that research was a billion dollar business. One is going to be super interesting. But you know what's more interesting that's happening right now? You're going to have lots of businesses making five, $10 million with one person or three people. And like the Gen Z folks, this is where they want to go. They have in my opinion, low trust institutions, which, let's be fair, why would they like, we don't want to hire them. You read in the news that AI is trying to replace them. You see in tech in particular, which used to be this like very glamorous, safe, cushiony job. It's brutal. And you have Jack Dorsey getting on your podcast saying I want to cut out all my middle management after he did a half his company riff. Like I would go start my own business too. Right? It's safer. What was safe is so risky now. And so like, is it really crazy to go hang up your own shingle and consult? Probably not, honestly.
- [07:57]
Podcast Host
So it seems like early on that you knew that branding was going to be important. You have the domain collective.com.
- [08:02]
Human Radfar
sure.
- [08:03]
Podcast Host
Guessing you didn't just buy that on Namecheap for 10 bucks. What's the story behind the domain?
- [08:06]
Human Radfar
We did not. So I think there's a couple questions underneath that. The first is why collective versus any other name. And the next question is, how did you get the domain? And those are two things. So the first thing I'll tell you is we started off when we were building it. It was a project that came out of expa and I was a partner at the time. Well, full time partner, I'm still a venture partner. And Garrett, my partner who'd started Uber, credit to him, he's very brand centric. He had gone through with Uber and Uber.com was quite expensive, actually. And it's interesting, I think it was Universal Music Group that they bought it from. They got stock in Uber and they made hundreds of millions of dollars.
- [08:39]
Podcast Host
I didn't know that.
- [08:40]
Human Radfar
And I think he was like, we're going to get it earlier because it only gets more expensive. Like, you know, visceral lesson for him. So that's been a thing at Expo that we really are conscious of the brand and then consequently all of the different application parts of the brand and even handles like we have on LinkedIn, where we are Collective. On Instagram, we're at Collective. Like, we really work hard to get those things because we believe in brand. And so the first thing was we had an opportunity to get Solo.com and Collective and we were just looking for something memorable, iconic, that we could really emphasize the direction we're going and Solo could do it, right? Like, hey, we're a toolkit for you and we're focused on you. But I always felt this emptiness around that brand, right? Like, I don't want to feel more alone, right? And so we thought about Collective and said, what if you could feel that freedom and power of being your own boss and you don't have to hire the person, but you have like the power of a team behind you, like a lowercase C collective. What if as a group, there are products and services we could build that are unique, that we can make even more powerful? Like, hey, when we have thousands of people, businesses on the platform, we can do better insurance or we can give you some better data or we can get you deals and like, maybe even have cool events. Like, if you're a solopreneur, you're not getting invited to the Forbes 30 under 30, right? What if we had like an event and awards? Like, our media reach is actually getting quite big actually. And so that was what took us in the Collective direction. It just felt right. Like I wanted it to feel warmer and give us that promotion. And then I was like, I woke up and this is true. It was like three in the morning and I was like, we Just kept talking about collectives. It was hilarious because we were talking about what we wanted to name the brand before we launched. And I remembered my friend, he was an advisor for the company, Joe Apprendi, and he ran Kollective, which was an ad network, and he had sold it. So I, like, took a note, you know, three in the morning note, and I'm like, call Joe in the morning. And he was advisor, my first company. And he was like, no, we sold it and this company, Zeta, bought it. You can't make this shit up. The guy who is the CEO of Zeta is a mentor of mine, David Steinberg. And I'm a shareholder in this company. So I'm like, dude, you gotta sell me this thing. And, you know, he's a great guy, but he, like, kind of helped me over the course. Like, dude, I don't sell domains. Like, I think he was. He was either going public or he was about, you know, something. He was big. And he said, we have, like, a jillion domains. So he's like, let me double check that I still have it. He has it. He's like, all right, we gotta cut a deal. So he ended up with a good deal. Did well on it. I hope so. Got a little bit of equity, a little bit of cash. Everyone kind of won. But, I mean, it felt very karmic in some ways. It felt very appropriate because I'm like, this has to be the name. This has to be right. Cause, like, the road was paved all the way down to the end to get it. And the person that I give to, again, I have a lot of respect for him. And ultimately he's a shareholder.
- [11:23]
Podcast Host
I don't want to compliment you too much. We talked about this market booming. Let's talk about something else that you were very right about, and that is services. I would imagine that in 2020, there wasn't consensus that services is where investors would want to put money. But it seems like now that's what everyone's talking about. How did you have that conviction early on?
- [11:41]
Human Radfar
So our thesis to put on my venture hat and speak in that, like, framework was you could drive the outcomes of, like, an accounting and tax firm with the efficiency of software. And that means two things. One, from a user experience, which is where we're starting, was, hey, there's a singular application. It's intuitive. It's integrated the juxtaposition with a traditional accounting experience, where, hey, there's this person in every season. It's different. You're emailing them. I mean, it's like, A very disjoint experience. And so that's what we're thinking. But then from a delivery perspective and an investor perspective, a small practice accounting firm of which there's 50,000, their gross margin is like 30%. So like it was an outlandish thing to say. I think you could build with technology and get it to the, you know, higher margins, like looking more like software. Now the threshold of that, to be candid was like unknown, like how far you could take it. But we knew we could do better than where they were and that was the seed bet. And so I think with AI, you know, I mean, I wouldn't have underwritten our investment to say, hey, AI is going to like how do you had AI at that time? That's what people are doing. I mean the AI native service, that trend is like off to the races to me. It's the same thing as what happened with tech enabled services. It's just with tech enabled services, one, I think they were building the companies the wrong way. But two, you didn't have this insane leverage that you do with AI. And so now I think everyone's like realizing, look, dollar and software, $6 in service, which one would you rather be at? And you have these large companies like intune zero. I don't know what they're going to do because their customer is the one who is an accountant who is willing to pay a dollar. Are they willing to, to burn that to go after our market? I don't think today they will, but I think at some point they're going to have an existential decision like who are they serving, their shareholder or their customer.
- [13:20]
Podcast Host
It makes sense too. If you look at your customer base, I would guess estimate probably 0% would sit there and be like, oh, I just wish there was like a better version of QuickBooks or you know, had more capabilities or anything like that. Like they just want the job to be done and they want to go back to whatever it is that they're.
- [13:35]
Human Radfar
I think so. Like I, when I was interviewing for PMs for like we have a bunch of workflows, we manage accounting, so bookkeeping, primarily, payroll tax and so forth. All these areas that we can help in and we're going to have more. I told them, I'm like, you're not here to be QuickBooks. You can refer to it for design inspiration patterns. You're here to basically do the book. So what is the optimal user experience to help them get the outcome they want without like talking to us. Right? Like what's the Perfect experience. And I'm a customer. I said, look, for me, I'm going to. I'm giving you my mx, I'm giving you my Mercury account. Okay? I want to not ever send you my statements, which you have to send to a bookkeeper, by the way, today. And I want you to do it. And then I want you to tell it's done. And if there's some questions, unfortunately, look, like Venmo, for example, or Zells, like, you just don't get metadata on why it's happening. So I might have to ask you about a few things. Every question should be as human as possible in language I can understand. I'm an intelligent person, I'm a business owner, but I don't want to be an accountant. And that's actually really hard to design for. When your reference implementations are built for accountants, like, there is a tendency to go back to, like, you know, statement of cash flow, all this and what you need. Like, they do actually need those things because. But, like, the interface shouldn't be built. And we're still. Dude, we're in the early innings of that, to be honest with you.
- [14:49]
Podcast Host
You mentioned 30% margins for an accounting firm. What margins are you after right now? Like, what do you think is possible?
- [14:54]
Human Radfar
I know what is possible. We are in software margins today.
- [14:58]
Podcast Host
Wow.
- [14:59]
Human Radfar
Yeah, we crossed over to that. And I'm confident, mathematically, verifiably confident, that we can get to enterprise margins in months.
- [15:06]
Podcast Host
This show is brought to you by Frontlines Media, a podcast production studio that helps B2B founders launch, manage and grow their own podcast. Now, if you're a founder, you may be thinking, I don't have time to host a podcast. I've got a company to build. Well, that's exactly what we built our service to do. You show up and host, and we handle literally everything else. To set up a call to discuss launching your own podcast, visit Frontlines IO Podcast. Now, back to today's episode and what's happened between having that thought and now. Like, how did you do it? I'm sure there's a lot to unpack there, so maybe just break it into almost like the pillars of how you pulled this off.
- [15:40]
Human Radfar
Look, I would probably run the playbook differently now because the tools are different. Probably could have saved myself some time. But I think the fundamentals are you need to understand the use cases. Like, if you're trying to deliver a service, you have to start from the outcome and say, what is the workflow that gets you to the outcome? And there's a Lot of non determinism in these workflows. That's why lawyers make money. Like there's curveballs. Some people, you know, have certain information, some people don't. Like there's a lot of that. So there's understanding those workflows, mastering those workflows, capturing them. And then you have to systematically kind of architect processes with software in mind. And so it's very hard because look, it depends how much capital you have, right? So all things being well, I think you would want to just like, okay, I'm just going to go check out all the processes and workflows, I'm going to build it with AI and then I'm going to launch. It's just almost impossible to do. The surface area is too high. So what you end up doing is saying, I'm going to take on some level of business and risk and I'm going to build against that use case. And then, you know, you just keep bringing these customers bill. And that's kind of more what we've done is I advise people like, look, I could rip, I could grow really fast. The demand is out of control for what we do. Like I could half my CAC basically today if I just let everybody in. But then I'll have to service them with ops people and accountants versus software. So we just have a deterministic way of doing that. I'll say something that I think is a little controversial. Maybe it isn't. I think the AI is the least important part of building an AI enabled service. I think the most important things are you need to really, really have observability and measurability on the process. The process has to be good, which means you have SMEs. So like, let's just take books. If you, you have to have a good bookkeeper, a good process, you need to understand it. But how do you know what good is? If you're 23, you're in YC, you're smart. But like, how do you know? I've asked people who build Bookkeeper, how do you verify an evaluator? They can't answer me. They're like, well, our customers say it's fine for them. Okay, blind leading the blind. Like they don't know. And so you need that SME, you need the process and you need to watch them do it. And so we've literally for, I mean, Zuckerberg just announced that he's putting tracking on every machine, right? I've been doing that for years for every operator and they know it. And so we've captured God knows how much data at this point for, you know, hundreds of people working countless hours on every single process. So I can tell you how much time they spend on a task, what tasks are being done well and then I map that back to cost and I can systematically go through and say, all right, is AI ready to take this? How will we do it? So the AI enablements, if you will, like can you make your infrastructure queryable? Is everything recorded? Is everything available? Like there the learning loops and the observation of those processes is the most critical because again the AI is getting exponentially better. And let's say that today it's good enough actually. But just because you can write code with AI, like what are you connecting? How do you know it works? That's the hard part. Like the system design is quite difficult, if that makes sense.
- [18:34]
Podcast Host
Yeah, that does make sense for a founder listening in right now who's you know, high on the drug of AI enabled services because I feel like that's everything now or product, I feel like productized services like what I had called this before but like for everyone, you know, high on this drug now, what should they definitely not do if they go and pursue this as a business model?
- [18:52]
Human Radfar
So the environment is rewarding insane growth and truth be told, you probably can grow fairly quickly I would say is if say you want to be in law or something like that, I wouldn't go. Justin Kahn is a friend, I like him a lot. He did atrium and I think obviously AI wasn't there at the time and I think there's actually room for like atrium number two, if you will and moving for startups and doing like deals and things like that. But pick a use case that's sufficiently large but it's still like very narrow number one. And number two, be very careful on your qualification criteria for how people come into your system and make sure you get good at understanding how you're going to expand, you know, because if you do that too fast there's no undo. So you look at scale factor, you can look at bench, you can look at all these guys from the tech enabled service era. The problem was not their thesis, the problem was not the demand. Candidly, I really don't think so. I've looked at the businesses, I got financials for both of them. I looked at buying both of them at points. It was, it was the execution. They were growing too fast so their churn went crazy. And the same thing's going to happen to these AI companies if they're not careful because the Only way to solve the use cases if you don't capture them is with people. Now your investors are saying, rip, you can keep doing it, but then your margins go down, your churn goes up. And I think that's what I would caution.
- [20:09]
Podcast Host
How do you have the discipline to throttle that growth and not just say, okay, all these customers that want in, come on in.
- [20:16]
Human Radfar
It's been really hard, honestly. I think now we finally, I mean, our metrics are breakout. I can genuinely, with confidence, hand to God, tell you I don't think there's a company in America that has our gross margin that's operating in our space. There's no accounting firm, there's no tech firm that is close to us. But that's years. And it took us years to culminate into this outcome. I don't think if I was to do it again from scratch, it would take the same exact amount of time. But it's still difficult in terms of the number of years. I think we were lucky. We had General Catalyst and qed and they were thesis driven. I think some of the timing with respect to capital raises, we had a little bit more time, but it was really hard. It was very difficult to do it and it remained difficult. Like, I can actually grow faster now, right now than I would have grown last year, than the year before, because I can expand my labor pool by, I don't know, depends. We'll see how efficient I get. Maybe 40% on operations and I can grow three times.
- [21:09]
Podcast Host
And do you think there's still more efficiency? Like, do you hit a certain point?
- [21:12]
Human Radfar
Yeah.
- [21:13]
Podcast Host
What are the areas you're attacking right
- [21:14]
Human Radfar
now in the coming year? Yeah, you know, we look at all our core workflows again. I think there's some areas where we had fundamental innovation where we changed it. Like we introduced an AR concierge into the application, which, I mean, we only had humans before. And so that of course mitigated our support costs tremendously. I mean, and so taking that up, I think we can improve it 50, 60% from where we are because we understand it, we know how to do it. And I can tell you where we need to like empirically go improve it. On bookkeeping, we're almost 70% automation for categorization, bank reconciliation. So I can take that up higher. There's diminishing returns, actually, as you get close to 100, believe it or not. I think tax is a really big area for us. I fundamentally think I would be shocked in two to three years if there isn't a huge deflationary pressure on tax filing. So that's going to be. Which is an opportunity for us. I think some businesses that are existing now on like formation, for example, like, we effectively can do it for free. We don't charge you for it. I think it's free and like Zen business, Legal Zoom, all these guys, I don't have any problem with the business and I'm sure they're thinking about it, but like, I don't think they want to be in the form filling business. That's not the business you want to be in now. So I think again, it's like how like your wife can she talk to our agent? And it can be as good as a CFO that just knows everything because no accountant will do that.
- [22:35]
Podcast Host
Right.
- [22:35]
Human Radfar
Like, if your wife hired an accountant, they still wouldn't look at your payroll. She can go ask, what was my payroll last time? Where's my K1? Like, no accountant can do that today.
- [22:42]
Podcast Host
Mm.
- [22:42]
Human Radfar
But that's like, not terribly impressive technically. Right? Like, once you set the AI, what's impressive is if she says, hey, we need to, we want to give me my mortgage packet. She doesn't have to say what it is. And we can say, oh, it's these five files we'll pull together, we'll send to you. And oh, by the way, we can use MCP to go talk to three mortgage companies. Do you want us to send it over to them and we'll let you know? And then you took it in and you can talk to them like through us. Like, literally, you can chat with your mortgage company anytime through us, and we'll handle it like you would again. You're like a CFO style concierge. That's the area that I think the leap is. And then I think it's almost unstoppable. I don't know how, like, again, accounting firms will compete with that kind of business.
- [23:20]
Podcast Host
How do you think about just expansion opportunities? And I'm a reluctant customer of Legal Zoom because I was a longtime customer of Earth Class Mail. It's a great service if you move around a lot. It's lovely to have your, your mail scanned. And then I got kind of sucked into this LegalZoom platform now. And I go to, you know, check my mail every week and I have to see all these other things that they offer that I have no interest in. But Earth Class Mail, that's what I want. But I'm always impressed by just the number of services now that LegalZoom has stacked. Do you view it as everything? Basically, Every service is on the table that for directions that you could go.
- [23:52]
Human Radfar
It's all about scale of time. Ultimately, I would start with the purpose. I want to free the founder to focus on what they care about, their core business, not their paperwork. That's a big stack. Now, we were started as business financial services, right? So I have a lot of work to do. I haven't earned the right to say that we've completed that job. That's where I want to really focus. And there's a lot of work within that to do, whether it be banking, health insurance, all these other things. Like, I can go down that credit, right? Because I know all your finances. I can even tell you, like, hey, you're about to pay your taxes or you're about to do your payroll. You're probably not in a great spot. You want a cash smoothing loan. Like, I know that better than anyone else. I can help you. So I think let's do well on that. And we always just again, focus on. We go, what's our purpose? What is the next adjacent thing that helps smooth it out? Like, I'll give an example. I think, like, it took me six minutes to do my business tax return this year. I'm actually curious to see what your wife's experience was. I hope it was good, but that's not usually what it is because again, they've accessed my payroll, my books, and I just answered a few questions, but I still have to pay it. And so it was okay. I got to see my more elegant than accounting. I got in the app, I got a notification, I looked in like, okay, you have to pay these things. And then it gives me a link to the site. That's okay. That's great too. But that's it. But what if I just hit a button? It's like, I'm done. Like, you pay it. So that's how I look at each adjacent opportunity is like, what's going to remove friction for that person that's adjacent to their experience? It's not like I'm going to go to the most lucrative product. Like, I'm going to jump straight to loans because it makes me more money. I'm like, what's going to keep smoothing that experience? Start there and like, look, I could see one. If you're like, Shopify skill. Could we do consumer finance at some point? Amex does like, why not? I can go and say, hey, you already pay me a subscription. Thank you. I'm gonna give all that money back to you because I'm getting incremental Margin right. On a credit card. I could do way better. I could do what ramp does way cheaper for your wife because they already pay me. Like I don't need to take that money anymore. I just want to take a little bit. So I think there's a lot of value we can confer back to the ecosystem with a subscription because you have that base subscription and then you don't have to over monetize all these other products. Amazon does an excellent job of that. Costco does an excellent job at that. I think that's where we want to go with additional products. But I think the difference between us and illegalzoom or even Zen business, and again, I'm not on those boards or in those companies. I don't know what their strategy is going to be per se, but they're not integrated well. Like we're more like an erp. Our products talk to each other so the burden is harder. I don't add things as quickly, but everything chats with our AI. Everything is integrated. If we fire you a task, it's in context of all the rest of your things. If you didn't do your payroll, then I'm not going to do your taxes. Like they don't care. They're just like, here's an app that I have a deal for, here's another app I have a deal for. And then just get Rev Share and Rev Share. Rev Share. Right. How will you build a self driving AI system if these things are disconnected? Fundamentally you can't. And so I can't necessarily do what they do and say here's a hundred products tomorrow. If I want to fulfill my obligation to your wife and say no, it all works together and I'm going to do a good job by you. This show is brought to you by the global talent company, a marketing leader's best friend. In these times of budget cuts and efficient growth, we help marketing leaders find, hire, vet and manage amazing marketing talent for 50 to 70% less than their US and European counterparts. To book a free consultation, visit globaltalent.co.
- [27:05]
Podcast Host
when I reflect on all of the conversations I've had with founders, I'm at I think 1400 now. I would say there's probably only been one person so far who started out serving small businesses and has had zero temptation to move upmarket. Or the plan wasn't to move upmarket and that was Tomer London from Gusto.
- [27:21]
Human Radfar
Oh yeah.
- [27:21]
Podcast Host
For you. Great. Like who were your inspirations? Because I feel like that was also not a very traditional thing to do to say okay, let's go out and serve very, very small businesses. It's not even SMBs, it's solo entrepreneurs.
- [27:33]
Human Radfar
Yeah, Tomer's a friend, by the way. We work with gusto. I'm actually a really large fan of him. It's interesting that they do actually. I mean, depends on what you think small is because they did move up market a bit. But I think to their credit, they'd really do a great job on like solo, you know, I mean, there's some overlap, I think with us as well. I don't really look outside of us. I don't for like this. Like, I think at some point if collective becomes sufficiently automated, like we're already. If for example, my business grows or someone's business grows, like we need to serve them. I don't want them to feel pain and have to graduate if they don't have to graduate. And so if I can do that in a way that's organic and natural and say, like for example, they're adding people in payroll, they can do that today. Their books get more complicated. For the most part, we can do today. So I think as the technology advances and AI advances, the only way I would move up market is if I can do it without sacrificing the experience for my core. In other words, the AI is so good, yeah, sure, you can just stay on it. I can handle you adding payroll, doing books for tax up to 100 because it's just so good. It's irrelevant to me. That's how I would go up market. But I wouldn't have this conscious. Like, I need this market. I need this TAM. It's 27 million people, right? And it's growing. Like, how selfish do you have to be? I could build a hundred $200 billion company at scale if I was really able to crack it. Like, why would I leave? And I really, I like this audience. So what keeps me here is my dedication to the icp. What would make me go up is that, hey, if my product that I built for these folks works really well for you, great, I'm happy to help you. But if it doesn't, I'm. We. I have built my business on turning away people and it hurts like we had at a point we were turning away 99% of applicants because we're so focused. We only did California, we only did cash based kind of only the certain services. And we did that purposefully because we're like, we can't do a good job out in these other states yet. And even when we do Expand. It's still painful because we have to learn it. I think we're thoughtful, we don't always get credit for that, but I think like we're not radical expansionists in that sense. But now with technology, our expansion rate is going to go faster. I think you're going to, you know, we're getting some like inbounds because of. I was, I sent Sequoia a love letter. Alfred Lynn I actually personally went and said thank you to him because I'm like, you've got me more of investor inbound than anyone. I don't know if you read their paper.
- [29:44]
Podcast Host
Oh yeah, impossible to not read that.
- [29:46]
Human Radfar
And Emergence as well had a playbook. And so I think, you know, if we continue on that it's going to be non intuitive that we're going to move faster now. But that's because we look like a software business. So software businesses go fast and I'm just, I wasn't one before and I was growing pretty fast, but I think I can just keep accelerating.
- [30:04]
Podcast Host
I don't know about you, but kind of my favorite time of the year is like in between, you know, Christmas time to the new year And I think 2023, maybe 2024, it was not so pleasant of a time. When I went to log into Bench, I was a customer of them and saw that they had shut down and we were all fucked basically. Or I think it was like 27,000 customers. Whatever it was, it's a very, very large number. And they basically said, yeah, you're fucked up. Like what was that period like for you? Were you, you said you looked at the business like were you trying to swoop in and buy them? And then was there any impact for you just in terms of customer trust that you had to go and try to put extra attention? Because I feel like that kind of fed into this idea that tech companies can disappear overnight. But my local CPA, who's been there for 20 years, like most likely he's not going to go away.
- [30:49]
Human Radfar
I don't know if that's true empirically because I think some of these people disappear more often than you think. But I'm trying to remember, you know, to give you an accurate read. I think there's a couple different audiences and a couple different experiences we had. I think the first experience was the venture community was looking at this category at that time very differently. So it was a pretty systemic shock. I can't tell you the nature of the financials. I don't even know if I'm still bound by the NDA. Because then they got bought by employer.com, i think. And I don't want Jesse to be upset with me there, but I think the amount of debt that they had was astronomical. Okay. So, number one. So I don't think they ran the business in the way that you would recommend. So the venture community, I think definitely there was some fright around this, and I think that's where you saw, like, it wasn't the only thing. There were some catalysts around this in terms of our customer set. We didn't really get a lot of people. I don't know, it just. It was not intuitive to me. But we didn't get a lot of people worried about our solvency, which, you know, I could understand. I mean, we're a startup. Like, we're not profitable yet. Like, it's a bet. It really motivates me. You ask about why I keep doing this. Now it's very different. I have people that are depending on us, and I don't want to disrupt their lives, my employees and these people. Like, economically, for me, it has to be a pretty big outcome to really, like, change my life to a point where, like, that's the sole motivation. But, yeah, I mean, it was more scary for me, you know, candidly, I think, because I'm like, that's going to be tough. But our customers didn't matter. We got a bunch of new customers, which is great.
- [32:15]
Podcast Host
Did you guys do, like, a portal to, say, transitioning from bench move here? I know a few others did that and they had a.
- [32:20]
Human Radfar
We try to maintain some level of respect for other founders. And I did want to be helpful. So I posted something. I don't remember exactly what I said, but I think I posted some along those lines of, hey, if you were impacted by this, like, we've created a code. We had our sales team and our consultants and everyone, like, on the phone on the weekend. Because I imagine people are flipping out more as a PSA and a service. I mean, I think it was an opportunity, obviously, but I think there's some level of, like, classlessness and predatoriness that go after. And it was a weekend because it, like, they got bought like that. Yeah, because I remember I started looking at the deal. I mean, we conclude internally, we back to the whole, like, responsible growth thing. I'm like, I can't. I don't want to. They had so many, like, SMBs that were like, like, SMB SMBs. And I think their margin profiles are all suffered. I mean, God, I wish I could go into financials. We learned a lot but I was like, this isn't my core customer. They have a lot of my core customer. But honestly we're bigger than them in that segment is what I learned. And now we're way bigger than them in that segment. And so yeah, I would've loved to buy that part of the business but I don't think they were excited about that.
- [33:22]
Podcast Host
And you don't have to share anything that you can't share. But like what do you think Bench got wrong? Obviously the debt part, you mentioned that crushing debt. What did they get wrong just in terms of trying to operate a productized service?
- [33:34]
Human Radfar
I think they did what everyone did, which was we're going to grow, we're going to have a capacity model that's operational. We pick an ICP and we'll use tech and engineering and automate as we go. The challenge is if you let in too much variant icp, what do you automate in what order? Like it's just too much surface area. Scale factor is to me was the example of that. And again I think the guy who founded it, very bright guy, I talked to him when they were having their issues. I don't think he was wrong on the thesis, I don't think he was wrong. I mean Kotu did it, a lot of smart people did it, they're not dumb. It's just I don't think anyone really realized first off, I don't think the technology was there like AI. Now you could probably try to be a little bit more cavalier, which I think is going to happen and say I'm going to let in a bunch of people and I'll figure it out later. You could not at that time. I still think you have to be more disciplined than you think because you will burn the customer fast. But I think just primarily not having this playbook, we're like, okay, we're going to focus on this. Here's why we're going to focus it, here's what we're going to do. Having a thesis, a hypothesis driven build is what I noticed. Also to be candid, I think they got caught in the throes of like Venture is very trend driven. Tech enabled services were hot and then there was a period where they were not like post 2021 capital efficiency everyone shifted away from margin like companies to software. Now look at that software. Nope, I don't want it. And guess what? AI companies don't have great margin. And so they got caught in this weird kind of vice of they built the model kind of on Par capital was very expensive, became very expensive. So they were trying really hard to increase their margin. I mean, you should see their financials. I think they did it too fast. They probably did their own because you could see the churn really increased. Like they didn't, weren't able to keep the customer experience on par. I don't know, I wasn't in the company, but I could see in the financials what they tried to do. I understand it. I think that's it. And my caution to everyone who's in this space is like, these are not stupid people that were running these businesses and investing in these businesses. Just be thoughtful again. I think you can go faster with AI stuff. I think you can be more cavalier, you can take more risk and maybe no less. We were just very thoughtful, maybe too much, but that's why we're here.
- [35:47]
Podcast Host
So you didn't buy a bench, but you bought someone else. Announced last week. Talk to us about the decision to buy that. You know, if you're so focused on your business now that you're saying no to customers, why separate that focus and buy another business?
- [36:01]
Human Radfar
Yeah, so we bought Open Ledger and they were, you know, they're very much trying to be embedded accounting providers. So can you help companies like ours or people who want to provide and offer accounting as a service in a broader suite? And I think there's obviously a market for that. There's a number of companies in accounting, payroll, I mean, this was a whole banking, you know, space for us it was more about accelerating and moving along the path. So when I start with like, I want to help people focus on their passion of their paperwork, I want to be able to make it smoother for them. There's a certain degree to which I can solve that problem, renting and using other people's infrastructure. And if the more infrastructure that I bring in house, I can do a better job in particular on AI. And so we said, okay, this is an investment that if we make it again based on economics too, like, of like purchase price and all these other things that we feel like it's worth that investment because it's mission aligned, goal aligned and we are going in a direction where we want to own more of the infrastructure, own more of the functionality so we can deliver a better experience. It was a really, you know, it was a good first move for us. I think there's a lot more that's going to happen that we're going to do. And in particular, you mean, look at the fintech ecosystem. I think there's A lot of companies that will benefit from having a home that has distribution. Like, think about all these, like, insuretech companies that do small insurance right now that they're, like, doing okay. But, like, we have distribution, so we could. And our customers need it and it's on our roadmap. So I think there's going to be more. But we would do it cautiously because I've done M and A before in this case. You have to be able to absorb it. You have to be really aligned. You have to think about the downside if it doesn't work and price it accordingly. I think if you price for all success, you are inevitably very unhappy because most of the stuff doesn't work.
- [37:47]
Podcast Host
If you could just wave a wand and everything that you want to have be true about the world, this vision that you have, like, what does this all look like in five years, 10 years?
- [37:57]
Human Radfar
Yeah. I look at it again from the lens of our members and what I would want them to experience. I want it to be, like, a boring choice. Like, when you're building, I mean, AWS obviously has competition from gcp, but there's this moment where, like, you're building a startup, you go on aws. Like, you don't think about, like, when I built my first company, we had to think about what data center do we want? We had to think about what box types we want, how many boxes do we want, what's our lease agreement, what's our OS like? Literally everything. And if we were wrong, fuck, you're over. Right? And AWS changed all of that, right? It was just like, it took out the complexity. You're a developer. You're like, all I need to worry about is software. If I'm wrong, I will just spin down some instances. That's it. If I'm right, I'll spin them out faster. It's a variable cost. Amazon will handle it and I trust them. That isn't a tremendous accomplishment. Like, just think about that. How much of the Internet runs it, how much entrepreneurship occurred because of that tool. If we can be that for these solopreneurs where they're just like, yeah, man, like, I'm talking to my friend. I have a cool idea. I wanna hang up a shingle and, like, the friction's zero, right? That feeling. I'm just gonna click a few buttons, collect it, spins up an entity for me. They'll, like, connect to my bank account. If I don't have one, they'll give me one. Everything just like, click, click, click, click, click, and you click your Way through that back office. And it's super easy. And, like, if you get something wrong, okay, don't worry, it's not gonna kill you. And this part of the business is easy. If you need to wind it down, you click a button. I think that feeling is important. But to get there, I want us to really be like, again. Imagine you have a CFO24.7 in your pocket who will take care of all these services. Like, I don't think about my books. I don't think of my taxes. My CFO Jeff, we joke around all the time. I'm like, I'm trying to put him in your pocket at a really, really effective price and make him available 24. 7. And he can spin up or spin down services, whether we own those services because we're best in class, like Apple or like Apple. Also, we have, and I don't know what this looks like in MCP Agent land, but the equivalent of, like, Agentix Store, because I think with financial services, you're going to need some level of security provisioning. And we go to Amex and we go to Ramp or whatever, and we're like, hey, we have mcp. You have mcp. Let's have a gateway. And like, our companies, tens of thousands, hundreds of thousands, God willing, can just, hey, I want to go look for a mortgage. And our two agents chat, and if people decide to hire you, they bring you on for outcomes, and our CFO can chat with you, and you can manage everything. And if you decide to fire that person, it's like, click, fired. Like, the easiest hiring firing process you've ever had. But I always think it from an experience perspective. But, like, that would be pretty dope, right? Like, if you could do that as a solopreneur, and, like, you just don't care about any of that stuff. Imagine how many more people start companies because the friction's nothing.
- [40:37]
Podcast Host
Love it. All right, man, we're gonna end here. Collective.com.
- [40:40]
Human Radfar
thank you.
- [40:43]
Podcast Host
Well, that's all for today. Today's episode of Builders, brought to you by the Frontlines. If you want more amazing content like this, visit Frontlines IO, where you'll find the library of more than 1500 interviews with founders, marketers, and other GTM leaders, where we unpack the tactical lessons from their journey. And of course, as always, if you do want to launch your own podcast, we'd love to have a conversation with you. Visit Frontlines IO Podcasts as a service. Mention that you listen, mention you love the show, then we'll give you a 10% discount. Thanks for listening. We'll catch you in the next episode.