SPEAKER_02: How did you know that this was peaking out? Did you have some crazy moment where you knew it was peaking out where people were like, just offering you sick amounts of money without doing diligence or I'll say one very, very, you know, big, big time investor,
SPEAKER_03: you know, offered a multi hundred dollar million round on the spot, you know, as I was pitching them. Yeah, again, who knows if it was in the room, when in the room getting offered a couple 100 million in the room
SPEAKER_02: during the presentation. Yeah, that is peak.
SPEAKER_00: This Week in Startups is brought to you by lemon.io. Get access to LemonHire, a platform with more than 80,000 pre vetted engineers that you can interview within 48 hours. Get $2,000 off your first hire at lemon.io slash hire today. InTouch CX, give your startup a boost and simplify your processes with next level automated customer support solutions from InTouch CX. Discover your custom strategy at InTouch CX.com slash twist and Northwest registered agent. When starting your business, it's important to use a service that will actually help you. Northwest registered agent is that service. They'll form your company fast, give you the documents you need to open a business bank account, and even provide you with mail scanning and a business address to keep your personal privacy intact. Visit Northwest registered agent.com slash twist to get a 60% discount on your next LLC.
SPEAKER_02: All right, everybody, I'm really excited to have our next guest on the program. He hasn't been here for five years. Last time how he Lou was on this week and startups was in April of 2018. Here we are in 2023. My God, so much has changed. Last time you were on Howie. It was pre pandemic and you just have series B for air table. Everybody loves air table. Welcome back to the program. The audience who doesn't know what air table is, and your inspiration for creating it. Please explain in the simplest terms possible what air table is. Yeah, so air table is basically like a Lego kit to build really
SPEAKER_03: useful business apps. I worked at Salesforce briefly saw the power of their platform and how configurable it was and basically left to start their table as this really, really elegant intuitive way to build, you know, a large class of business apps, especially with a kind of direct to end user PLG model. And now increasingly with more of an enterprise focus going after bigger, meatier use cases. But it's still the same concept, the same vision of we want to empower every company, every every line of business, even every, you know, person who wants to go out and build a really useful business app to do it with the best possible Lego kit.
SPEAKER_02: Got it. And you mentioned the acronym PLG product led growth. Why don't we explain to the audience since a lot of founders are now saying, Hey, our success comes from this philosophy, when you first became aware of it and how you kind of incorporate PLG into your process at airtime.
SPEAKER_03: Yeah, so well, I can tell you what it wasn't nothing, which was when we started out that you know, founding the company and going out and raising our initial seed round. I remember getting so many investors saying, you know, love the concept, love the platform concept, you know, maybe even like you as a potential founder, but this whole idea of build an awesome product, you know, it's going to get traction on its own, you're going to grow at least initially through organic virality, that just doesn't exist. You know, maybe maybe Dropbox did it. This was actually before Slack had launched. So Slack wasn't even yet a case study. But really, you know, maybe Dropbox kind of did it ever know, to some extent, but there's no really good comps for a company that's grown primarily on the backs of a, you know, beloved product. And so if you want to get serious, you got to go and figure out an enterprise sales motion from day one, right. And in some sense, I'll come full circle to this, but like, you know, I'm sure we'll explore this topic. They weren't entirely wrong. But also, you know, what, what, what I think they missed and what we then saw over the past 10 years really happened is, you know, bottoms up adoption really did kind of hit the mainstream. And obviously, we've seen now a bunch of SaaS companies, you know, start that way. In fact, maybe it's more of the default way to start a SaaS product these days versus build up like a direct sales motion from from day one.
SPEAKER_02: And that bottom up is, hey, there's some hero in the organization who's tech savvy, who likes to GSD get stuff done, and you use another word for s if you like, but they like to GSD. And so they come to work, and they're like, Hey, I got this like little secret thing here, we got this problem, boom, I'm going to solve it over the weekend, you got that hero champion inside the org, and then they spread your product. And we used to call this word of mouth, but it's slightly different than that. Because, you know, you do have this like person who's evangelizing, I guess is the term we used to use back in the 90s. This evangelist is out there, you know, telling everybody this is the greatest thing ever. What was the tip of the spear for air table? Because when I first saw it, I thought, Oh, there's a spreadsheet. You know, oh, this is like this weird, like database spreadsheet, because people would open up spreadsheets. And they would use them as databases, they would use a spreadsheet as a CRM. It was that kind of the start of this was like the, the the Swiss Army knife, spreadsheet of Excel, that everybody used a different way, but it didn't really have the the tools or the ease of use in the templating. And that that inspired you?
SPEAKER_03: Yeah, so I think what I saw was this big gap between, you know, spreadsheets, like you said, are very widely used for everything, right? Like, they're not just used for number crunching, which is really what they were designed for initially, like, going all the way back to like, VisiCalv Lotus 123, like, you know, this spreadsheet was really designed as a numerical, you know, analysis, like a number crunching tool, right? I mean, literally for accountants or finance people. And I think what they become because there's nothing else out there that's as easy to use or build on is the default way to make a really lightweight app. It could be like inventory tracking for a small business, it could be your makeshift CRM, it could be here's how you actually manage production workflows and a media company, but it gets used for everything. And, you know, in contrast, like on the other extreme, if you want to actually go and build or use a purposeful app, it's very, very complex or brittle, right? So you find like, here's one very specific point solution that does this specific thing, right? Or, you know, if you want to go build your own, it's obviously very laborious, and then you have to kind of maintain it, evolve it, etc. And so what we saw was this big, big chasm in between those two extremes. And we wanted Airtable to have as frictionless an entry point as a spreadsheet. So it was a very explicit design principle to make Airtable feel like a spreadsheet. So you know, the motto internally was like, make it no harder to use than a traditional spreadsheet, but immediately add much, much more value. If what you're doing is not number crunching, but you're trying to build an app, right? You have data in there, you want to build a workflow around it, like we can do a lot better of a job by picking off those those use cases.
SPEAKER_01:
SPEAKER_02: So I was gonna have you on the show, because we had mentioned, or I think it was sacks, we were talking on all in about the state of the company. Yeah. And you know, it's been like a rough time the last two years. And I guess there was some sort of tweetstorm that we referenced, but you email is that, hey, Jay, how not under you also emailed sacks. Hey, that tweetstorm wasn't correct. It had a lot of weird data in it. And we were like, Oh, yeah, you know, our bad. And we want to correct that we corrected it at the top of the show, I think the next episode of the episode after for sure, explain to me what that tweetstorm was how it got out there. And then how you went about correcting it, because this is kind of a new phenomenon. There are these like anonymous accounts or quasi business accounts. Sometimes they call them the thread guys on Twitter slash x. And they're kind of getting cloud by kind of telling stories and tweets, which is kind of cool. You know, sometimes you call them the reply guys, right. And so
SPEAKER_02: somebody did that, but they use your company as like, the the format and it turned out they just, you know, triangulated information from like crunch bass, or wherever or like random story. So what did they say? What is reality? Let's just get it out there. Because the company's actually doing really well. Yeah, well, you know, I appreciate you calling that out.
SPEAKER_03: I actually don't know exactly where this person who posted this original tweet got their data, I was sued from crunch base since they work at CB insights. But, but they referenced, you know, air tables at like 150 million in revenue now, I think they said, and you know, it's only been growing, like low double digits, like, I think they quoted 14%, or
SPEAKER_03: something over the past three years. So basically, like, almost no growth. And, you know, and like, it might have to raise more money and like, it's like, what is the state of the business, right? Like, and, you know, and like, again, actually, don't don't fault this person, like, you know, maybe they were just, you know, what, it was the CB insights guy, the CB insights is
SPEAKER_02: I don't know exactly what their business is. But there's some sort of data provider, I think like, maybe they compete with crunch bass. And so their premise was, oh, you're, you know, you're like this, you know, example of like, if you take the public market comps of like Monday and Asana, they're trading at six times revenue, 12 times, you're a private company, and you have like a very high private market valuation at some point. And that the bottom line was that your valuation was probably somewhere between a billion and two. And that's like whatever 90% from the private market $11 billion valuable, right? I got it now in front of me. Yeah. So
SPEAKER_03: there's a big point was like, well, we've, you know, air tables raised almost half the amount of money, right? Like we've raised 1.4 billion in total capital. So what if air tables not even worth the money they've raised, right, which you have, like, it was a very speculative post, and, you know, don't fault this person, you know, like, I think they were probably trying to one get some buzz and like, you know, a conversation started with something probably intentionally a little bit more sensational, but also, like, you know, there's limited data out there or a lack thereof. And, you know, I think it was speculative by definition to kind of fill in that document, and maybe again, just kind of solicit a conversation to get going. That being said, you know, like we actually had already gotten to 150 million in error by the time we closed our series app, which was like, you know, our last round, it was, you know, obviously during the peak, the 2021 era, the peak. Exactly. Yeah. So we can infer like, at a
SPEAKER_02: 30% growth rate, or even with headwinds 20%, whatever it is, frankly, we've grown much higher than 30%. Okay, great. Since that
SPEAKER_03: time, right. So, like, you know, even the overall business, and by the way, like our enterprise business, which, as we talked more about the product and where we're going, that's been the part that we focus more on, but even the overall business has grown at high double digit percents for each of the last two fiscal years, like even this year, which, you know, we're obviously closing out, you know, the second half of like, we've still continued to grow at a healthy pace that we have seen you know, like, some some tighter budget cycles, etc. Some like, you know, tougher, tougher, you know, kind of sale cycles, by like, still growing at a very healthy rate. And also, by the way, like, we're we're profitable now, as a business, so Oh, wow, you're profit actually generating, we're generating cash flow. And we still have the vast majority of our funding raised in the bank. So we have over 900 million of capital in the bank. Now we add to it. So like, you know, the so let me ask you a question about that, like, you felt the
SPEAKER_02: pressure in the market to switch from high growth to, you know, growth, let's say, from absurd growth to high growth with profits, even with 900 million in the bank, that seems like maybe an overreaction. So how do you make that decision as a CEO? Hey, we got to get profitable when you're sitting on 900 million, would it kill you to lose 100 million a year for the next nine or 150? Or were you just trying to be ultra conservative? And then how do you deal with these late stage investors and the pressure of like, well, why are we putting this money to work? We gave you the money, you gave it to grow? Yeah. How do you manage all that?
SPEAKER_03: Yeah, well, you know, first off, I feel very fortunate to have, you know, all of our investors really understand it and be supportive of us building the best company we can for the long term, meaning I've never felt from any of our investors at any, you know, round, you know, creating pressure on us to do something that's perverse to what we actually believe is going to be the best outcome for the overall business long term, no short term pressures, etc. What we're really trying to do is build a high quality business. And it turns out, you know, I think, in an era when there was very cheap capital, it makes sense to go and try to deploy that capital in the pursuit of both experimental initiatives, both product wise and go to market. And also just rationally where the return on the dollar is low, right? When literally the you know, kind of you know, risk free interest rate is next to nil. Like it's rational to try to deploy. Absolutely. Yeah, people are offering you money at 0%,
SPEAKER_02: essentially. And you're getting a minor dilution. Of course, you're going to go for the gold that makes sense. But the game changed on the field changed. If you've ever had to hire developers, you know, it is a giant headache. Are they built from the startup grind? How do you know if you hire the wrong developer, man, maybe they drag your whole team down, maybe they take you down some rabbit hole, and you don't know what's going on. And then not only are you back to square one, you might be behind the eight ball, right. But if you hire the right developer, that could be a game changer. Let me introduce you to lemon hire. It's a platform that helps you find qualified senior developers who want to join your team. Lemon hire pre selects vets and matches tech candidates with high growth startups. So founders can focus on building their vision. Here's why you can trust lemon hire. They got 80,000 pre vetted engineers, and you can interview any of them within 48 hours notice you paid nothing until you started onboarding lemon hire is going to replace any of your hires for free in the first 30 days. So just go to lemon.io slash hire H I r e to get a special offer just for twist listeners $2,000 off your first hire. That's right. Two grand off at lemon.io slash hire l emo n.io slash hire hire with confidence at lemon.io slash hire. How long did it take you to shift those gears? What was that process like? How many board meetings? How many quarters?
SPEAKER_01:
SPEAKER_03: I would say, you know, it's, it's not so simple as just we gotta go profitable, although that's, that's one obvious, like kind of milestone that we crossed, it takes much more of a surgical understanding of, you know, how, what are the growth levers of the business? And truth be told, I think, you know, air table for sure. But probably a lot of other companies that you know, I know of, like out there don't have as good of an understanding of their growth levers as you know, we hoped or thought we did right. Meaning, like, if you're, you know, a company that's super optimized, like Salesforce, at some point got to the point where they could, they could predictably add, you know, another rep to the system, get this much productivity out like clockwork, and they just could know, you know, we can scale it up at this rate, right. And I think for a lot of other businesses, especially those that came from PLG, and especially for us, where it's organic PLG, we weren't even paying performance marketing dollars to get our top of funnel, like top of funnel just compounded over time. And obviously, now we've added on this enterprise sales layer to our business, it's actually murkier, like, how much incremental return are you going to get if you like add more people, if you invest more into the business, right? It's not as predictable as as we would have liked. So I think, you know, a lot of what we did was spent the past call a year, even year and a half, really trying to surgically understand, you know, how the business worked, right. And that meant for me, going out and not only talking to other operators, some of whom are great, you know, advisors, or friends to the company, people who have run companies, or operated on scale, like ServiceNow, or Salesforce, or many, many other kind of SaaS companies, you know, or, you know, for me to go internally within the business and like, spend time both on the ground level with customers, etc. And kind of really understand, like, when things are working well, like when we do really well with a customer, and we're expanding and growing those dollars, like what's actually happening? And what are the pieces of that, that we can recreate more consistently. So I think it takes both an outward of like, you know, an externally informed approach. And certainly your board and investors can help with that. We have some great ones who have like, really good perspective, but also like this very internal oriented approach where you go and really understand like, what's working within our customer base within our org. And really, it's about like, you know, one, just getting to a healthy size as a business that, you know, kennedy is like, the right scale and the right structure for our type of business and our scale. You know, I think, again, like the cheap capital, in many ways, it allows you to kind of go and grow ahead of the curve, right? How far out were you hiring? Was it because this is something we
SPEAKER_02: talk a lot about on all in, we talk about it here. You know, that seemed to me that, you know, the big tech companies were hiring two to three years out in terms of like, let's get staff in here. Yeah. For literally two years from now, we'll get, you know, because we're in this crazy competition. And then, you know, I'd say the the large three public companies like yours, maybe you're what a year ahead of you trying to hire a year ahead of plan or something to that at the peak? Yeah, I mean, the thing is, like, I didn't even think about
SPEAKER_03: it in that way anymore. Because you could say like, for our revenue scale, or for the revenue scale, we're going to be at in a year, you know, what what is like the typical SAS comp, like the SAS benchmark headcount of sales and marketing and product dollars, you could do that analysis. But I think it's kind of the wrong way to think about it. Because if you're really tight on understanding the incremental impact of every hire you make in any of those categories, product, marketing, sales, you know, headcount or dollars, then you don't have to just pull forward all of those hires in one big mass, right? I mean, it's easier to go and hire aggressively ahead and just try a bunch of stuff. And, you know, frankly, I take accountability for kind of trying that, you know, kind of a little bit messier approach of, of, let's just go and like, step up and see what works. Rudolphman calls it blitzscaling, right? It's like, Uber did it,
SPEAKER_02: Travis did it, you know, like, for sure. It is a known strategy to get to scale. You had, I think, 1200, 1300 people to peak, you did two rounds of layoffs. Yeah, I think you cut 500 people or so. Yeah. But 650. Wow. So half the amount of people. Yeah. And that's tough, right? It's brutal. It's really
SPEAKER_01:
SPEAKER_03: tough emotionally. You know, for like, everyone in the room, like, I feel, you know, there were days where, like, I felt horrible. Imagine having to, you know, go through that. And obviously, like, I'm not even the one who really has to bear the brunt of, you know, these changes. But at the same time, I think, you know, the lesson learned is blitzscaling, I think makes a lot of sense. And I have a lot of respect for the concept and Reid coining it. I think it makes a lot of sense in certain domains. So certain businesses are truly a winner takes all. Like, if you're Uber, if you don't get in there, Lyft or some other competitor, whether it's this market or a different one, is going to come in there. And it's literally who can get there first, right? Like, the only advantage you're going to have is in building bigger network effects, getting better economies of scale, etc. Especially so when you have a business where the unit economics get better with scale. And so you like, it's almost like you can't even build the profitable version of this business without getting to scale, right? SAS economics are obviously very, very different. And I would argue, in most cases, if a SAS company, especially after you get to like 100, 200 million of revenue scale, like if you can't make the economic model work, or it starts like being very clearly on the path to making it work really well, throwing a bunch of fuel on the fire is not always going to help you now it might in some cases where you know, there are some light, you know, kind of economies of scale, like R&D dollars, obviously, you know, you invest in the software product once you get to sell in as many times as you like, but I just think it works differently in SAS. And also, these are very rarely as commoditized, you know, hyper competitive markets as the on demand economy, you know, companies that require blood scaling, right? So just describe the table. Yeah. Well, as I say, describe the feeling,
SPEAKER_02: the vibes, as the kids say, but just a general feeling of going from, hey, we're in blitzscale mode, we're adding tons of people. Yeah. And I would suspect a little bit chaotic, but like, you know, you know, like, sort of this, like, blitz scaling would be, and then, oh, my God, we're profitable, we're efficient, we're safe and secure in this crazy maelstrom. That is, you know, this was definitely a recession and is a recession for our industry, even though there's weird stuff going on, like, GDP is growing at like, a really incredible number. So very bizarre. Yeah, it's the craziest recession I've
SPEAKER_02: ever been through, because we're having a crazy recession travels going into a recession. It's almost I call it the rolling recession, where it's like every industry gets like three or four quarters of downmarket crash and or correction and or recession. So just the vibes, like, how did it change from one to the other? And does the company run more efficiently and feel stronger now?
SPEAKER_03: Yeah, well, look, I think we're in the early days of building what I would call like the scaffolding for a really lean and focused business, right. And we've been trying to build, you know, that scaffolding and kind of phases over the past even couple years. You know, the riff is just part of that, like, that, you know, actually getting leaner, getting smaller is part of that. But I think there's a lot of other parts of it, like, you know, having a really good operating framework for the business, like, how do we create accountability? One thing we did with our product award, which is really exciting to me, and I think to a lot of the folks in our company is like, we're actually aligning our major product pillars, like these are the groups that, you know, the major surface areas of the product organization, we're really aligning them around different segments of the business with real revenue dollars attached to each of them, right. So no longer is it just like, hey, we build awesome features functionality. And then you know, eventually it gets sold and marketed somehow. Give me an example.
SPEAKER_02: Yes. So, you know, we have we have like three major pillars,
SPEAKER_03: and one of them is around self serve revenue, and it's basically called the team's pillar. And we're orienting around, like the self serve product experience. So we always want to have this amazing, you know, kind of top of funnel experience, anyone can hear about your team will come in, sign up organically, and get a really good onboarding experience, right, and even build an app of their own, share it, whether they're, you know, a small team within a big company, or they could be, you know, a small company in and of themselves. But we want to make that experience as effortless as possible. And that's actually going to feed in not only to our self serve revenue base, which, by the way, as a separate, like even by its own, would be $100 million business and still continues to grow hyper efficiently, right? Because that doesn't require much human touch at all, or it's going to contribute by speeding in, you know, becoming the seeds of adoption within large enterprises, that eventually we can put a human touch on and turn into one of our many, you know, million dollar accounts, right. So like, that's, that's one pillar, and it's, it's accountable to real revenue outcomes, especially in the self serve side. So it just there's this like feeling of like, wow, like, you know, now the product folks and the R&D folks can really feel like they own revenue, right? And like, I personally think like, there's just something really palpable and exciting about like, thinking as a business owner, like a GM, rather than just, okay, I build this functionality, this feature, watch it, that's kind of that's kind of it, right. So that's one of the three pillars, you know, what won't go to all three. But what over it's nice, like when the people who are in the
SPEAKER_02: kitchen who are making this incredible dish, can bring it out, land the dish on the table and see the customer's reaction to it. And for a long time, the chefs, you know, and the kitchen staff in the front of the house is two separate experiences. And when they kind of get blended, and you can kind of connect those things. It just makes it more fun, you know, I completely agree. When you're a founder, you need to be able to do every job. But you can't be an expert in everything all at once. And one of the hardest things to scale is customer support. So here's your solution. In touch, CX will help you to create a custom support strategy designed around your unique challenges, all powered by AI and automation. In touch, the X provides automated solutions in voice, email and chat support to enhance customer experience. Again, the experts at in touch, CX will design a customer experience strategy tailored to your brand and your goals. When you use in touch CX to automate repetitive tasks, you'll deliver faster customer support resolutions and you'll boost customer satisfaction ready to ignite your startup growth. We'll book a free consultation with an automation expert at in touch CX comm slash twist.
SPEAKER_01: That's in touch CX comm slash twist.
SPEAKER_03: To build on that metaphor, you know, it's almost like if the shots are actually coming out and seeing like the look on the customer's face and even like seeing which orders they read, like which items the customer comes back and keeps repeating an order for. Then you can actually tailor your recipes around that right? Like you actually start developing different menu items and like the way you prepare it caters to that that end customer experience, right? So you get like whole feedback loop going where I think it's true, like one of the challenges of hyperscaling is it's really hard to keep that tightness of the whole feedback loop, right? Like it's the opposite of the lean startup manifesto or methodology where like, you really want like everyone in the room working on the company to understand deeply what the customer actually cares about what they're willing to pay for. And if you're adding, you know, like a ton of people and you're growing every function, kind of at the split scaling speed, I think you end up getting, you know, like almost by necessity, disintermediated from that feedback loop.
SPEAKER_02: You cut the company down almost by half your clients were doing that at the same time. So you charge on a per seat basis. And so here's a clip. I'm taking you back in time and unexpected clip, they just pulled this live. Here's a clip of you and I five years ago. And me asking you about pricing. So 120 bucks a year 240 bucks a year for a user. If I'm an enterprise and I got 10 people in my company, do I pay more or less? So you pay
SPEAKER_03: more? Why do I have to pay more? So you get more of everything. So you get, you know, kind of expanded usage limits, you get a lot of administrative control. So certainly, like SSO, and you also get, you know, kind of you say that SSO, what do you mean
SPEAKER_01:
SPEAKER_02: by that? Yeah, so SSO is basically where we allow you to
SPEAKER_03: use your own authentication system to allow users at your company to use air table. So I don't have to create a whole
SPEAKER_02: bunch of new accounts. Exactly. I just have the single sign on
SPEAKER_03: boom, right. And you can provision deprovision those users, you want to get all those tools, and that would cost what
SPEAKER_02: 30 bucks in the enterprise a month, 40 bucks, 60 per user per
SPEAKER_03: month, 60, 720 a year. So we actually, this is expensive. No
SPEAKER_02: wonder everybody wanted to invest in this company. And I missed it. I mean, I think, what was that series a like that must have been like a 10 $15 million round. So we raised actually a
SPEAKER_03: little bit less than that. It was around 8 million. Oh, and
SPEAKER_02: you just raised 52 correct? Yeah. 52 million. And that was
SPEAKER_02: probably for no more than 20% of the companies. I mean, it's a $250 million post. We're not coming on the on the but I'm in
SPEAKER_03: the ballpark. But sure, yeah. Because I'm in the ball. It
SPEAKER_02: could be 10% of the company. Who knows? Oh, my Lord. I mean, it's just incredible. Like me now. And I dress more
SPEAKER_03: like myself. Yeah, exactly. I used to put the suit on and try
SPEAKER_02: to be respectful for CEOs. But just think about that. Like, gosh, what a moment in time, huh? Like, that was really if you think about zerp, and you think about the inflection with that was 2018, you did your series B, just look at that crazy like jump in valuations you had, you said the series A was like 12 million or something. And then boom, yeah, that was the zerp. That was the start. I mean, that that be round for us was before our unicorn round
SPEAKER_03: of obviously, happened later in the year. And you know, obviously, we grew as well as a business during that time and since that time, but yeah, it was, you know, like, I remember when we got to, you know, get the the unicorn status, which was ever something I like, cared for neatly, but you know, just symbolize like one one step along the way. Like there weren't that many unicorns out right. I mean, they're like, you know, especially in SAS. I mean, there were a bunch of the No in SAS it was dozens. Yeah, and yeah, and to sum it
SPEAKER_02: everything, it was low hundreds, it might have been like 200 at that time or something not 2000. And, you know, I think people extrapolated and suspended disbelief on a lot of companies in terms of product market fit, for sure. And maybe it wasn't there. And that was the big mistake in the entire industry. But you had product market fit. Yeah. How did you know that this was peaking out? Did you have some crazy moment where you knew it was peaking out where people were like, just offering you sick amounts of money without doing diligence or like showing up at your house with donuts trying to get you to take
SPEAKER_03: I've never had that experience. I will say one very, very, you know, big, big time investor. You know, I think this was more tongue in cheek than serious, but, you know, offered a multi-hundred dollar million round on the spot, you know, as I was pitching them. Yeah, again, who knows if it was in the room, we didn't in the room, we didn't take the round, you know, like, I don't even know if it would have been a real, real hard offer. But, but, you know, I think we always took a much more methodical approach to finding investors that we never really got to like, experience the full, like, you know, the full on wildness of the funding environment. Like, I wish I had stories of like, investors, like throwing term sheets, like, getting off a hundred million in the room during the presentation.
SPEAKER_02: Yeah, that is peak. I mean, usually, yeah, usually, they can't, you got to meet my next set of partners, you got to do diligence, let me talk to a couple of customers, that should be a 30 day process, not a 30 minute process, right? For however many minutes into the room meeting you were, but listen, you're you had an incredibly strong company, back to the product. Because I'm fascinated by the product and a product trend I'm seeing, there were verticalized apps, and there were developers. And for example, we needed to do like two products recently in the last year. One of them was we wanted to have a speaking of investors, when our startups we invest in 150 a year now. So it's like, when I met with you, it was like, we're investing in like, two a month. So 24 year now we're at 150. And, you know, we have all these founders together, and we wanted to build a way two different things. One a demo day, kind of handshake thing, where, you know, everybody who's an angel investor coming to the dim sum demo day, I hosted at a dim summer restaurant, because everybody loves them. So we would send him a link, hey, pick companies you like, click here to request a meeting, click here to request to join the thing. And I was like, okay, like, when hire a developer to do that, and like, girl, like, Oh, we built already, I'm gonna hide your belt air table. I'm like, Okay, great. And that's supposed to cost me, I was prepared to pay 25 or 50k to a developer to do that. No developer, there's people today who are good at scripting and using these tools and the views and zapier. And if this then that whatever they use as glue, to get it dialed in. Then the second one was we wanted to build an investor intelligence database. And so of course, SurveyMonkey type from they start doing the forums, where I'm like, well, you need a presentation level for this. So they start using air table, but then there's Coda, we use a lot to and then notion. And so notion, kind of documents going into database Coda kind of database with some documents, you're like Excel with database. And the thing I love about this trend is one, we can prototype software, and it just works. And
SPEAKER_02: it's not like no code. And it's not developing. So we can kind of go faster. And I can have the product leads. It's only a certain type of Yeah, business person who can do this. But if I can get them to learn how to do this, and some people do it natively, about 20%. Now, if I think either young or highly technical people can do it, man, it's so powerful when the business leader can build the system, and just take a week to do it, man, it, you don't have to hire a developer and then have them disappear. And yeah, everything like that. So we should
SPEAKER_02: speak to how people are using this. And if my use case is kind
SPEAKER_02: of indicative of why this is so addicting, and why, you know, for me with a 20 person company, I don't know what we're spending on our table, but nor do I care. Because I'm sure it's all we should charge you 25 to $50,000. Wow. $5,999. They were
SPEAKER_02: paying low thousands because I have probably had, I don't know, 20 people on it or whatever. So 20 people 3040 bucks a month, wherever paying probably like under 10k. Yeah. It's a really great deal for both parties. I feel like for sure. Yeah, I
SPEAKER_03: think what's happening and I love that you called out, you know, to be honest, it's not like every single person in your company or even like, you know, outside of your company, I'm sure like your company tends to skew much more tech savvy than like most 20 companies out there. But it's not like every person is coming in and like building very elaborate apps on our table all the time. Right. And I think what what it's actually what's actually happening with this genre of of product is it almost like abstracts away the technical bits of building a useful business app, right. So, you know, if you literally wanted to develop it with code, you'd have to know all the best practices of front end engineering and back end engineering and get the off right and like just do all of these different bits and pieces, just to be able to then work on the part that actually captures what I would call like the business data layer and the business logic, right. So in the case of, you know, let's say the like matching the, you know, the the alarms or the company, you know, founders, like or the fundraising apps, like there's a know how around the business, you know, part of part of things, right? Like, what do you want to happen from a business standpoint? What's the behavior you want? What's the rules? Like what are the best practices that you want to codify into the software? Like in traditional software development, you have to go through all the technical bits that have nothing to do with that, just to get to the point where you can express that part. So really, what's happening is that we're trying to abstract away and I think a lot of these other products are trying to abstract away a lot of that technical complexity and make it you know, it's almost like how you build a website with Wix versus with HTML, right? Like HTML has nothing to do with your business. It's like a very, like, you know, kind of rote language. And you got it, you got to muck through some some technical stuff to build the thing you want, whereas Wix just lets you go and, you know, designing yourself, right. So in our case, but yeah, both of those are exceptional products at that.
SPEAKER_00:
SPEAKER_02: And it's a perfect analogy. I don't have to get in there, you know, and do that. And just the same with forms like you used to have to put fire up a database server, fire up some HTML on the front end. And then you have like some like type form or Survey Monkey just super elegant. Starting a business used to be a real pain in the neck. You needed to pay lawyers tons of hidden fees. It was a mess. We all know that but not anymore. You need to check out Northwest registered agent, they're going to help you form your company really fast thing get you the docs you need to open a business bank account. They're also going to provide you with mail scanning, which is amazing and give you a business address to keep your personal privacy intact. Northwest can form LLCs, corporations or nonprofits. And they're also going to equip you with a registered service agent and operating agreement or bylaws, membership or stockholder certificates, a resolution to open a bank account, and it's all privacy by default. You'll also get lifetime support from their corporate guides. So here's your call to action. Northwest provides everything you need to start and maintain your business and they're giving twist listeners a 60% discount for just $39 plus state fees. They'll form your LLC corporation or nonprofit visit Northwest registered agent comm slash twist. Once again, Northwest registered agent comm slash twist. So let's go over to AI. Yeah, I you must be obsessed with this. I'm getting crazy obsessed with it. I made everybody in the company use chatgpt for and barred and everything when they have some sort of an issue or question. I'm like, share with me because now you can share the threads, you know, the chat thing. Share the thread with me where you asked before like, I didn't.
SPEAKER_03: Or let me Google that for you. It's literally like, let me Google that for you the website.
SPEAKER_02: And I'm like, let me chat gpt that for you. And had an incredible experience the other day we were we're looking at an equestrian company to invest in. And they're like, we have
SPEAKER_02: questions about the time I'm like, and I just like write my thing. You're an investor at a venture firm or considering an investment in an equestrian company, please give me all the data you can in a table with citations of the total addressable market and subsections and categories of this. And like I put it into Claude chatgpt and barred and my Lord, the amount of incredible information and I was like, we could have hired an expert in equestrian or spent 30 hours of research and not gotten something McKinsey to go and do like a million project on this. Exactly. It's literally like, what is McKinsey going to do? Yeah, this like and this is year one of the technology. We're about to hit the one year anniversary. So that's just from my world, like a small thing. What are you doing as CEO? What are you playing with? What are you seeing internally, as you run your organization? And then what are people looking for in the product? And how do you think about the roadmap? So internal to external? Yeah, so I mean, I agree. I think I think this is like
SPEAKER_03: pretty profound stuff. Now, I do think we're also at this point where there's a lot of hybrid, I think we're in like, maybe close to peak height. And, you know, while these models, especially I'm more excited personally, about the elements as opposed to like image generation or other domains, but, you know, or at least LLMs that have multimodal capabilities. So it can reason and perform, you know, tasks like what you just described, like size me this TAM, or write me up like a product description for this new concept I'm toying around with or whatever. It's more of the reasoning skills that are interesting to me than just like purely like, you know, something much more visual, or audio, for instance, like the music generation. But I think to me, what's exciting is that, like, this is like the iPhone One, right? It's like clearly going to be a game changer. You know, it's got some roughness around the edges. It's not like it can't replace everything all overnight. And I think the problem is like, you know, I just hope that the hype doesn't get so big, whether it's on the investment side and funding a bunch of, you know, companies that are doing ASF, or, you know, probably, or on the customer side, where like, you know, customers are pitched a bunch of stuff that doesn't actually work well enough today to serve the purpose that there's this like massive trial of a trough of disillusionment that like we have to dig ourselves out, right? Because that would be my fear is that this is really a potent technology, there are clearly ways that it's going to, you know, be applicable today, but more so as it just gets better and better, like, and even comparing GPT-4 to 3, right? The difference between those two, even though like, they're both large language models with, like many billions of parameters and trained on like, really, really large data sets of text, like GPT-4 can pass the bar with human like accuracy, GPT-3 could not even though GPT-3 is pretty good. So, so I think we're just getting closer and closer to the point where you can actually get these breakthrough applications. The way we think about it with Airtable is we're not at and probably won't be personally, I don't think we're going to be at, you know, in the very near future, maybe not even in like five years, a point where, you know, AI is so good at everything that it just replaces all human interaction, intuition, etc. in software, right? Like, you just completely cut off the human in product development and generating marketing campaigns, etc. It's all, you know, completely run turnkey by AI. I don't think we're at that point. I don't think we're gonna get to that anytime soon. I do think we're at a point where if humans are very much in the loop and can use AI in a very specific way, right? And they can tailor how the AI is used, that is predictable and very tailored, then you can get really interesting applications overnight, like tomorrow, you can start unlocking these use cases. So like the whole premise of Airtable is that we create this app platform that is really easy to build on. And so you know, by virtue of that, we think we can give you AI pieces to put into that platform. AI field is, you know, our first example. Yeah, I like that. That's a really cool idea. Explain it.
SPEAKER_02: Yeah. And it's like, you know, if Airtable is a Lego kit, like,
SPEAKER_03: now we give you AI pieces to build in that Lego kit, but it's really still about what you're building with those Legos, right? So if you want to build a way to do your product roadmap better, but now with AI, you can choose exactly what prompt and what data feeds into that prompt. And you know, kind of how you output the result, maybe it's to customer or to triage customer feedback, right? Yeah, into your product roadmap, or maybe it's to generate product ideas, or, you know, a PRE for a feature that is in development. But you can choose exactly how to feed that into that workflow. Yeah, in the context of your own data with with our, you know, kind of Lego pieces. So yeah, that's,
SPEAKER_02: yeah, you're right mouse way gone, hey, this is customer feedback. Anytime customer feedback comes in, you know, say it in bullet points, as short as possible, and then put it into a category so you can tag it etc. I noticed tagging, I'm working on a project on inside common, like just tagging of news stories or tagging of blog posts. It's kind of figured that out better than humans can do it. Like it knows the who, what, when, where, why in a document. So if you say like, tell me the names of people and tell me the locations in this field of data. It kind of does that perfect. And that's just if you think about data normalization and trying to make sense of data and feedback from users, man, that's amazing. And you say, let's give me the top two points each piece of customer feedback is, and then look at this body of feedback and write a daily summary of it. Wow, that stuff is crazy.
SPEAKER_03: I think the range of it is what's so powerful, right? Like when you talked about like, you know, doing TAM sizing, that's like very high grade stuff, right? Like, you know, like strategy consulting level stuff. But the fact that you can also have it do very rote work. And by the way, yes, you know, in either case, even if you're using the most advanced model, like Claude 200k token, or GP432k token, or whatever the next best version is going to be. Yeah, it's expensive relative to the cheaper, smaller models. We're talking like cents per quarter.
SPEAKER_02: Ridiculous. Yeah, it's like a billionaire is like a millionaire, like going into Starbucks, like, there's no amount of damage Starbucks can do. You could just order whatever you want and leave half of it behind. It'll be okay. Yeah, like the the bills on these things is like laughable. Are you upselling people on the AI? Or you just in the class and putting it in flat rate? How do you think? Yeah, we're still, we haven't officially launched the, you
SPEAKER_03: know, the paid, you know, full GA version of our AI capabilities. And we're building out more features to that leverage AI. So in addition to AI field, we're building something that allows you to ask questions of your data, you know, like pretty much what you expect, right? So if you have a CRM in your table, you know, or your your, you know, companies you've invested in, you can ask questions like, hey, what was that company I invested in that does this? Or, you know, list off all the companies that I've invested in in this category, or whatever it may be, right? We're doing more stuff with AI, and then we want to package it all, and both offer it once as kind of an AI bundle, and tbds to what that pricing will be. But you know, we think of it similar to how Microsoft is thinking about copilot for office where it's going to be a lot of enhancement, the enhancements that just make the product so much better, that it's worthy of like a significant add on fee, right? It actually that's exciting. Yeah, because if
SPEAKER_02: everybody's going to do more with less, right, you're doing your your company has tripled in size, I'm gonna guess double do triple the size since you went half in terms of the staff size ballpark. I'm in the ballpark again, something like that. I mean, we just did our most recent a couple months ago, but
SPEAKER_03: like, we've continued growing, we've definitely gotten much larger in scale than then. You know, at the time, like, we priced our last round, and your customers are coming to you like, obviously, Elan
SPEAKER_02: Twitter went down 85%. He's releasing more product in the past year, then I can remember last five years of Twitter releasing product. So people are learning to do more with less. So then it makes sense that if everybody's becoming like, I don't know, like a demigod or a god, you know, like, everybody's the 10x programmer who has a job. Well, then if their tools a little more expensive, if their sword or their bow and arrow is a little more expensive, like, okay, that's worth it. Because like, if a 10x programmer wants a $5,000 laptop instead of a $3,000, and you just get it to them, for sure.
SPEAKER_02: And so that's kind of exciting. If you were paying six or 700 bucks, maybe they pay 1200 bucks, who cares about the incremental 500? The seat based pricing, which has been, you know, everybody's, you know, cutting the number of seats, as people tighten up their organizations, and everybody's doing that all at once, which is really weird. But then people spend more, right, if you can spend more, so I don't mind, like, getting people a more powerful computer or spending twice as much on a SaaS product, if it makes those people that much more effective at work, that makes sense.
SPEAKER_03: For sure. No, 100%. And, you know, I also think the C count reduction, or just like headcount reduction, is probably a little bit more hyperbolized than it actually exists. Like, we've at least not, I mean, there are certain sectors, really small businesses, or like, you know, high growth tech in particular, obviously has, you know, a major hiring slowdown, or even like contraction in headcounts. But, you know, a lot of other industries, like we're not really seeing as dramatic an impact, right? So with our customer base, like, we're still continuing to expand, net expand, like, in most of our customers, like, Oh, you know what, they didn't over expand, they didn't over hire like we did. So they weren't going, they weren't
SPEAKER_02: doing the blitz scale. Totally. They were kind of, they pasted more evenly throughout,
SPEAKER_03: right, the whole time. But, but I also think to your point, like, you know, if you can really increase productivity of each person by a dramatic amount, then I mean, one, you can definitely charge more for it, right? And that can counter, like, if you can get, you know, a real 50% plus increase in the average revenue per user, through these capabilities, like, you know, that's a pretty significant, you know, jump up, right? And that would be our aim is to get like a sizable jump up, not like something more more incremental. But second of all, arguably, the companies have more reason to hire people who are more productive, right? Like as something, you know, as the ROI on hiring an extra person goes up, not down, like, I actually think there's a there's a world where, at least in some functions, in some industries, you hire more like developers, for instance, if they're more productive with a co pilot than not, right. And I would venture to say, like, I think the same will be true for even non developers in other, you know, areas like PMs, marketers, etc.
SPEAKER_02: Yeah, it's I think it's gonna be very interesting to see what happens in terms of the economy and jobs specifically, we're still at like, record low unemployment for a lifetime, past 50 years. And there's still 1.6 or seven jobs for every person who wants to get a job, unemployment, participation actually ticked up a little bit. So we I think we peaked at 69 70% you know, back in the like late 90s, early 2000s. And now we're like, we were at like 6061. And now it's back up to 62 63% of able bodied folks in the United States, we're looking for a job taking it. So I think in this down market, you'll see more people, maybe come back to the workforce, which is great. But then there's going to be whole swaths of jobs that are removed. But you know, with it's so inspiring, this new technology that I think it's going to make people want to solve more problems in the world. And the world's got an infinite number, I believe the world's got like an infinite number of problems to solve. So what I'm seeing is all the freezing of hiring in big tech is resulting in more startups, I'm seeing many, many more two three person startups, and they feel like they're as productive as the 10 person startups, just five years earlier, because they're all using AI up and down the entire stack. For sure. So I'm super excited about where this all had. So you excited about AI? Are you concerned about job destruction? And we had this executive order come out, everybody's gonna have to submit their paperwork. And I don't know, somebody in the Biden administration is gonna like, tell them if they are allowed to do stuff. And it's a little bit it's a little feels like a little regulatory capture. Shout out Bill Gurley. What are your thoughts? I don't know if you saw this stuff, but what are your thoughts on? You know, the best this could be or how bad? For sure. I think near term, I'm very excited, both because, you
SPEAKER_01:
SPEAKER_03: know, it's just cool to see, you know, play with directly some of the stuff that you can do, whether it's with mid journey, or chapter T or so on, like, we can all touch and feel it. And like, it's exciting, right? So this stuff is like, actually pretty mind blowing. And obviously, for us as a business, like, I think there are some pretty obvious ways that we can take advantage of it and help our customers gain value from it. We can be that conduit, I would say long, long term, as in like, 10 plus years out, I'm pretty worried. But there's just so many unknowns from here to there. You know, like, obviously, a lot of the conversations around like, what if we, you know, create super intelligence and like all of the AI safety features? And can you even build AI safety with, you know, with a super intelligence that thinks far beyond or in a different way than we can? I mean, that stuff is like, truly mind blowing, and like, really, really, like, unnerving when you when you think like deeply about it. I can't, you know, claim to know, like, I'm anywhere near an expert. I mean, you're getting into science fiction, like, for sure.
SPEAKER_03: What what happens when AGI kind of starts thinking about, you
SPEAKER_02: know, what, how can I make myself smarter? Yeah. And I mean, the chip companies are now using AI, optimize the design of AI chips, right? And so if that's any indication of, like how this thing could spiral, yeah, out of control, or in a wonderful way, I mean, we could see, like the concepts of Moore's law or chip design. Yeah, they're gonna figure some things out that were orthogonal thinking things that humans just
SPEAKER_02: couldn't get to, it's gonna find all these nooks and crannies of efficiency that, you know, maybe we stumble upon every decade or two, for sure. And it's just gonna find them and be like, Oh, by the way, dummies, here's six more things you could have been doing this whole time, a new way to fabricate processes, hyper
SPEAKER_03: efficiently, and with much higher performance, we can get like a lower like a smaller nanometer process. Here's like, less heat. Yeah, yeah, it's gonna be wild. I mean, it's
SPEAKER_02: going to be wild when that starts happening. And then this whole generation that's worked from home, you know, they're really like tweaked out about like commuting, there was this whole viral bahaha on Twitter of somebody freaked out that they had a commute and they had to leave their house at 730 and come home at 615. And I dunked on the woman a little bit like, Oh, my God, the humanity, but I was thinking about it. You know what, like Henry Ford put everything out a five day work week from a six day China. And I think it was Japan at some point, people were working too hard in Japan started putting up posters in the I think it was the 90s, late 90s, early 2000s. Like take some recreation time, leave work early to go play tennis, like they were trying to promote people to have more social lives. Yeah, I think we might get to the point with this AI in five or 10 years that bosses like me are like, you know what, everybody's doing so great. Companies thriving. We're gonna move to a four day work. I know people are like cheering right now, like, Oh, my God, Jake, I've seen it. But I mean, if things did get that good, and you could get your job done in six hours a day, or you could do it in for four days a week, like maybe that's where the world winds up. Maybe we just the great bargain is going to be Yeah, come back to the office, but three days a week, and for one day from home or something. What are you seeing in terms of people's work ethic, and how that might change and then work from home? I'm curious as we wrap here.
SPEAKER_01:
SPEAKER_03: I personally I love the idea of work smarter, not harder. And I think, you know, going back to the topic of like, who are the people who go out and build systems build apps, right? Like, you know, instead of going in and trying to like, you know, match, you know, companies by hand, and do it a very manual way, like how do we automate the system, I think there's going to be more and more value for systems thinkers, thinkers, people who, especially, you know, with AI capabilities, figure out how to do something in a more automated way, still with some human intervention. But like, you know, to your point, it's like three days a week, four days a week worth of human oversight and input versus like, a full five day or six day plus, I'm doing every single little thing in this. You know, the obvious example would be like, you know, if you're trying to generate marketing content, you know, obviously, a lot of the actual writing itself of words, you know, you know, writers, you know, often got and still get paid on a per word basis. I think, you know, I think obviously, I can help with some aspects of the writing, I don't think, you know, it's ever going to be a replacement for like, human crafted or human edited writing. But you know, how much more can you get with with, you know, the, the, the leverage is big, I can tell you, like, in every organization I've ever been in, and I'm, you know,
SPEAKER_02: a former journalist and editor, you know, there are always people who are great at editing, proofreading, fact checking, sentence structure, you know, leads, style things, and you should just run around the organization, okay, who's my style guy? Okay, who's my fact checking person? Oh, who's my, you know, grammar, like lunatic. And now use Grammarly another shout out to another amazing product. And like, it just gets everything right. And then you can highlight stuff in Grammarly. And I think notion has a similar product. So now like I'm using notion and Grammarly, which just highlight something you're like, make this better was like the prompt in terms of writing, just highlighted and make it better. And somebody had written something and most people are not natural writers, they forget what they learned in school, they never had a good teacher. Now you can just write something that's really sloppy and choppy, highlight it and say make it better. And it literally goes from like a six to an eight. Yeah, maybe even nine. And I'm like, well, that was a lot of what I did as an editor, a lot of what I do now in terms of how people I feel like the AI is replaced 80% of my proofing with people maybe 90 Yeah, maybe 90%. And then the part that's left is like, what are we trying to communicate? You know, yeah. And that's super exciting to me. If everybody could be an eight out of a 10 of as a writer, as opposed to four to seven, right? Man, that would frickin fantastic. We just focus your plan, like what is it? What is the actual
SPEAKER_03: essence of the thing that is different or matters here? Right? That's super exciting, whether it's a product or, you know, like writing content, or, you know, you're you're designing some kind of new, you know, concept, like, it's all about like, what is the like essence of the thing? I think that's, that's exciting to natural world where like, more of us, we all spend more time on that, because I like, that's the fun part of the job.
SPEAKER_02: Exactly. I mean, Sachs was saying his biggest challenge as a writer, he's a good writer is obviously a great thinker, which is starting from a blank page. And he said now, like, because it builds the bones of the framework for him, and he could fill it in really quick. That kind of plays to his strength, because he didn't have the ability to like, where do I start, which was kind of like my strength, like, I know exactly where to start, like, I know how to capture that first paragraph, I know the ending. And then I'm just like, okay, well, you know, how do we get from point A to point B, it's gonna be a couple twists and turns in here, I got all my little toolkits. And now it's like, you know, they I can do all that frickin fantastic. So yeah, yeah, so the legs,
SPEAKER_02: exactly. All right, listen, continued success. Thanks for taking the time. I know you're super busy. If you don't use air table, it rocks. I mean, your organization is going to be like 10 times more efficient, and it's getting more efficient, just give it a shot, go to the sort of a bunch of templates, and all these lunatics make amazing templates. So if like, you're doing your fundraising or cap table, or OK, ours or project management or CRM, you can just start whipping stuff up. And it's kind of an addiction for some people, you know, if you get addicted, don't blame me. But it's a great product. It's super affordable. airtable.com. Listen, how great to see you. Let's do it again in five years. All right, man. Talk to you soon, brother. Cheers. Awesome.
