Founder/CEO Airtable Howie Liu

Auren Hoffman (00:01.294) Hello fellow data nerds my guest today is Howie Liu. Howie is the CEO and co-founder of Airtable, a collaborative spreadsheet and database platform that has raised over $1.4 billion. Wow. Airtable is at the forefront of no-code, low-code movement. Before founding Airtable in 2012, Howie sold his first startup, eTax, to Salesforce when he was just 21 years old. So now we know how old you are now, if you can do the math. Howie, welcome to World of DaaS.

Howie (00:12.597) Yeah

Howie (00:24.875) Yeah. Yeah. Thank you. Really, really good to be here, Auren. And I should do a shout out, which is we first met in the context of your prior company, RapLeaf, because that startup of mine, eTax, actually was a customer of WrapLeaf. you know, your data powered our product. So full circle.

Auren Hoffman (00:43.458) Yep. Yeah, exactly. I mean, I, you know, I should have like, wish I was smart enough to invest in air table when it started back then in 2012. Uh, yeah. So by the way, I'm, as you know, I'm a huge fan of air table. I use it every single day. Um, tons of our company use it. So I'm in it all the time. So I got, I'm like, this is like, gonna be an advanced.

Howie (00:50.176) Hahaha

Howie (01:02.235) that's awesome.

Howie (01:06.932) That's awesome.

Auren Hoffman (01:09.462) interview because I'm talking like power user here. but let's talk about. Yeah, exactly. We'll, debug live on the podcast. so let's, let's talk about first, let's start with just like this, the no code, low code movement. kind of promise to democratize software creation. Where are we at on that vision?

Howie (01:10.514) Okay, great. Yeah, we can do some live debugging too.

Howie (01:23.807) Yeah. Yep.

Howie (01:27.881) Yeah.

So I think it's really interesting because the first wave of no code, which I like to think that we helped to pioneer, was really about just taking all of these complicated but powerful app concepts. So data model, logic layer, automations, and then interface. And if you're a programmer, you know that those three layers are effectively the building blocks of any application, like model, view, controller. And...

distilling them into a form that was really intuitive for any non-technical person to interact with, right? So in effect, doing for app building.

Auren Hoffman (02:03.79) non-technical, you really mean like a non software developer, they still could be like relatively technical, you know, person or something.

Howie (02:10.526) I mean, I think that it really comes down to how well designed the product is, because in our case, we wanted to make it for everyone. I mean, literally everyone. And just as a proof point, early on, we had just such a broad range of users, cattle farmers, literally cattle farmers, building cattle tracking apps, replacing, apparently there's a whole vertical solution category for this, CattleMax and all these other kind of brittle tools. We had small legal firms, lawyers themselves going in and creating like,

Auren Hoffman (02:14.968) Yeah.

Yeah.

Auren Hoffman (02:34.933) my gosh.

Howie (02:39.52) their own rolled version of CaseMap, which is a Lexis Nexus product. So they built their own case management software in our table. And I mean, the coolest like random use cases too would emerge. Like I remember being on a screen share with a woman who I think was like an octogenarian and like going through and like cataloging like family memoirs, historical like kind of, know, kind of letters and like literally figuring out how to like drag and drop

you know, into Airtable like these, these scanned attachments. So I think it really comes down to a test of how truly easy is the product to use because there's, there's been low code platforms for much longer than we've existed, right? Like, you know, before Airtable, there were all these other products like QuickBase, even Salesforce itself is technically a low code app platform. Like you can configure it to build any app you want.

Auren Hoffman (03:11.116) Yeah, that's so cool.

Howie (03:32.671) And even before then, I mean, there were back in the day, like FileMaker, Microsoft Access, like Lotus Notes, like Ashton Tate, Dbase. And so I think what we really wanted to do was like truly lower the floor to an unprecedented level where like it really didn't require a like technical admin IT type person, let alone a programmer. Like it could be like truly anyone could go and build their own useful apps.

Auren Hoffman (03:58.658) Yeah. And then where are we at? Like, where do you think, like how much of that vision is realized? How much of it is still to come?

Howie (04:01.331) Yeah.

Howie (04:05.095) Yeah, so I think it's a perpetual journey, both in terms of lowering the floor. And we're obviously in this moment where, I like this Bill Gates quote, like he said recently, there's only been two truly astounding technologies that I've seen in my life, truly jaw-dropping moments. And one of them was the GUI. So when he probably saw Xerox PARC and the original GUI, it obviously sparked the inspiration for both.

Microsoft and Apple to go through the same. And now Chat2BT. Because we actually have, like, we've bottled up intelligence and, you know, kind of broad reasoning capability into an instantly available, fairly cheap API that can be scaled up on demand. And what that means for no code is like, I literally, if you interpret the vision of no code not as like, okay, we have to just go and create gooey metaphors for app development. But instead, now it's like, you can do it without code.

but through a genetic conversational app building. And you see some of this with the vibe coding apps as well. Like you can go to a cursor as a non-developer and type in the sidebar chat, just build me an app and it will try to do a good job, right?

Auren Hoffman (05:11.788) Yeah. By the way, I was, was, I was in, you know, I was in bold thought new today. Like just like doing that. It was great. Yeah.

Howie (05:17.308) Yeah, no, it's awesome. And it's so magical too, because I think the range of expression that you can do is just broadening and broadening, right? And the barrier to entry is dropping. Now, I think the Vibe coding products, they do have some limitations, right? If you're using agents to generate actual code, it's somewhat unreliable. There could be data issues. There could be security issues. And it's hard to, if you're not a developer, you can't actually inspect the thing that it's generating, right? You can't inspect the actual code itself.

you can only look at like the rendered output and hope that underneath the hood, it's all working as it should be. And I think that's where we have a unique take, which is blending this concept of like no code GUI components where you can see and directly interact or edit with like the data layer, the automation layer, the interface layer, and like literally drag and drop things if you don't like how it was generated, but also add on top of like this, on top of that, this agentic, know, conversational layer where you just talk to our new agent Omni to do anything.

Like literally anything that required point and click in Airtable, now you can just conversationally talk to this agent to do.

Auren Hoffman (06:21.838) Where do you think just in general, just if you think of just like a company software stack, where do you think like AI is going to change the most and where is it going to like kind of lag behind the lease, let's say over the next kind of three to five years.

Howie (06:25.627) Yeah. Yeah.

Howie (06:33.317) Yeah. So I think that every software category is basically up for grabs, right? And like, if you think about the disruption of SaaS versus on-prem software, like I actually think SaaS was really important, but like not actually as nearly as big of a deal as, as a AI and agentic product experience. Meaning like SaaS, you know, it's like, okay, great. Like we, we massively reduced the overhead and cost of like

you know, maintaining software of like maintaining this deployment. It was much easier to adopt and distribute, but like the software itself, the functionality was kind of the same, right? It's like, okay, now it's just like installed on-prem or it's like posted in the cloud.

Auren Hoffman (07:13.208) Yep. It was easier to like, or it was easier to collaborate because there were a lot of things where like, would like download to my computer and pay a hundred bucks or something before, but I couldn't collaborate with other people. So it's more of like a single player mode type of thing.

Howie (07:23.782) Totally.

Howie (07:28.121) Absolutely. I mean that's very fair and the hacks they had to do multiplayer mode before, mean in our space, like think about Microsoft Access. Yeah, mean like Access, you literally had to save a file to a shared LAN disk drive, right? And then like a network disk drive and then collaborate by multiple clients.

Auren Hoffman (07:33.912) Correct. Yeah, it was annoying. Yeah.

Auren Hoffman (07:41.155) Yeah.

Auren Hoffman (07:45.634) I mean, before, before get, would use SVN to just try to like, and I would like super hack it just to like share files with people and stuff.

Howie (07:54.301) Otherwise, not for actual code commits, but just like you would use it for like your own like Dropbox like, you know.

Auren Hoffman (07:58.542) Yeah, exactly. Like that's what that was my drop box that we use internally in our company before, you know, it's like it was terrible break all the time. It was super at the

Howie (08:03.408) Yeah.

Howie (08:08.335) Yeah, well, I think you've officially earned your uber nerd badge from that. When you use SVN for basic file sharing and collaboration, you definitely earn a uber nerd status. But yeah, I I agree with that. I think the distribution, the sharing collaboration layer was probably the single biggest disruptive force of SAS. Even our contemporaries, think about Figma versus Sketch. It was basically the same functionality, but one Figma was

Auren Hoffman (08:17.621) Yeah.

Howie (08:36.136) much easier to sign up for adopt and then share. Anyone could open a collaborative file and same with Airtable. But I think to me though, that's still very incremental relative to the massive disruptive change of how product experiences, how software experiences can behave when it's fully AI native to the core. And so when I think about a lot of the current software landscape, much of it is just like we went out and like,

Auren Hoffman (08:58.519) Yeah.

Howie (09:05.66) you know, especially in the vertical solutions world, like we built a vertical solution for one category, right? Be it plumbers, be it, you know, dog walkers, whatever, be it like, um, you know, small businesses, like yoga studios, like mind, body type stuff. And we just like built all of the data model and the business logic rules for this particular use case into our software. And now we have like a, you know, a sales and distribution channel to sell it. like fundamentally it's just hard coded software. Right. And I think.

I think that now there's this massive disruption of with agentic app building, first of all, the barrier to building your own truly bespoke software that's quite advanced, not just simple, has radically gone down. So that obviously tilts the spectrum in favor of custom software. But then on top of that, the software itself should feel and have a different form factor, which is it should be agentically infused. So if you build a CRM today, it shouldn't just be your age-old CRM.

Auren Hoffman (09:49.079) Yeah.

Howie (10:03.076) it should actually have agents like automatically running within it that go out and do, you know, web research about your, your, your customers, right. Or go and automatically listen to call transcripts or use MCP to grab data from other sources. And so I just think like the radical improvement in the product value that you can, you can create with a truly AI native baked experience is so big of a difference that, you know, it makes all of the current incumbents, especially the legacy kind of enterprise incumbents.

Up for grabs.

Auren Hoffman (10:33.738) You mentioned MCP, like how bullish are you on being able to bring in data and kind of collaborate as opposed to like, you know, or like the alternative is like needing access to the browser to go do it for the user or something.

Howie (10:40.987) Yeah.

Howie (10:48.432) Yeah, yeah. mean, so I think both will be done agentically, right? So I think that, you know, we've seen like what browser use looks like in the context of agents, not just like, you literally scraping the HTML, like what deep research does, but like actually products like operator from Chatchabte or Manus, et cetera, that now like just have literally the ability to scroll through a website, even like, you know, hand it the user back control when it needs, know, when it needs the human to go and do a captcha.

typing your credentials, et cetera. So I think like, you first of all, browser use doesn't mean you can't do it fully agentically. But I do think, I think it's a huge deal that agents are now able to effectively go and automatically do increasingly more and more, if not all, of what a human would have had to do with a desktop, right? And not just like on the web browser, you know, or through MCP, but like, you know, anything on your desktop. And I think

Like it's very clear that we're quickly approaching that point where at least anything that's like fairly deterministic or rote, like an AI agent can definitely do. mean, we're seeing this like big, you know, kind of discussion in the e-commerce space of like what happens if like people start, you know, not going out and doing comparison shopping manually as they do today, where like they look, you know, they search for some item and clothing, Google pops up, it's sponsored results. Like you look through those, you click on them, you examine them.

What if you just told an agent everything about yourself and what you like, your style, or it analyzes you, and then it just goes out and looks for the perfect products for you, and then it even transact? Like, yeah. Maybe we're not the target yet. Yeah.

Auren Hoffman (12:22.766) Yeah, that would be awesome. Yeah. Yeah. And like, like I, I, I buy default by probably most, almost everything on Amazon and you know, probably 70 % of the time or 60 % of the time it is the cheapest price, but there's a lot of probably ways I could probably get a cheaper one. If I had an agent doing it for me and it just told me, by the way, you were going to buy an Amazon. You, you just saved, you know, um, $50. I would be happy to share a percentage of my savings with the agent.

Howie (12:35.832) Yeah.

Howie (12:40.911) Totally.

Howie (12:49.957) Totally. Totally. And by the way, like I think the common misconception I see is like, you know, AI and agents can only be used for very like, logical tests, right? Where it's like, you know, literally find me the lowest price. It's very utilitarian, but I think like agents actually can, can, you know, kind of have a certain sense of taste, right? Like, I mean, you can, you see this with like the creative applications of LLMs where

ask it to compose a poet, poem, you know, about your company in this style with this flair, and it does a very good job, right? And so I think it actually can learn, totally.

Auren Hoffman (13:18.894) Yeah, of course. Yeah. Oh, I use it to pick restaurants, pick hotels, like things that are very, very personal to me. And I try to get, okay, here's what I'm doing. Here's who I am. And, and, uh, and it doesn't, you know, maybe it's not perfect, but it does a pretty good job. It does as good of a job as like, um, you know, if I, if I had a personal assistant do or something.

Howie (13:27.268) Yeah.

Totally.

Howie (13:38.458) Probably a travel concierge. Yeah, exactly. And so I think that's really interesting where it's like, it's not just like a comparison shopper. It can actually be like a personal shopper where it's finding products that it knows you might like, you know, because it understands your tastes and preferences and style in subtle ways that are not just like a keyword match, right? So I think it gets really, really interesting when you start to look at, you know, the...

the disruption with agents as not just being strictly like rogue utilitarian tasks like what we could do with like RPA for instance where it was very very scripted and instead like much more like subjective and you know kind of soft feel tasks

Auren Hoffman (14:20.866) Now, back in the day, if you think of air table 1.0, it was Microsoft access. and Microsoft access was like an amazing product. was so good. It was, it was, it was like, it was so well done. And then at some point, I don't know when, at some point, like 20, 25 years ago, Microsoft just killed it for some reason. And they stopped investing in it.

Howie (14:25.069) Yeah, yeah, yeah, basically.

Howie (14:40.002) Yeah.

Howie (14:44.205) Yeah, yeah.

Auren Hoffman (14:46.83) Do you have any idea why? Like they could have been, it could have been amazing. Like why did they do that?

Howie (14:49.89) Yeah. Yeah, yeah. So funny enough, you know, we, we, when we were founding Airtable, we, instead of just looking to the future of like where we thought things were going, we also looked to the past to like, try to understand like, you know, the existing, the prior art, the Microsoft accesses of the world. And by the way, like Lotus Notes also conceptually similar and like a very, very successful product for its time, kind of like groundbreaking in terms of the paradigm they define there. Kind of the first like collaboration meets like structured data.

sort of type product. But long story short, like I ended up talking to a number of people who were literally involved in Microsoft Access, Excel, et cetera, like Stephen Sinovsky, Adam Bosworth, who actually created the very first GUI database as his own company and then led, yeah, yeah, exactly, led a different, one of the other leading database companies early on and then later joined Microsoft to run Access. And then Stephen, you know, was a leader at Microsoft that I think oversaw

Auren Hoffman (15:26.013) wow.

Auren Hoffman (15:34.159) I didn't know that. Okay.

Howie (15:48.087) the Office Suite at one point and beyond. But the story that I ended up piecing together was that actually at one point, Access and Excel were kind of like effectively two peer products that were both like really, really important, right? One's a database, one's spreadsheets. Spreadsheets are for number-crunching, like ad hoc analysis, databases for like structured list data, CRME, ERP type use cases that you build yourself,

Auren Hoffman (16:03.564) Yeah. Yeah.

Auren Hoffman (16:15.661) Yeah.

Howie (16:17.004) But if you look at the products superficially, they kind of look the same, right? Like they both have like this tabular view, this grid view, and Excel actually had started building more table-like functionality into Excel. So it wasn't just number crunching. You could kind of use it to make a list. And so literally, like what I heard was like there was a literal moment where like Bill Gates called a meeting because he was hearing all this consumer confusion. And the head of access and the head of Excel were in a room together with Gates and he was like,

Auren Hoffman (16:20.247) Yes.

Howie (16:45.142) This is way too confusing. We gotta like basically decide on one product to be kind of the default product that subsumes both of these use cases. And then the other one will kind of relegate into like a more like kind of edge case or kind of power user product.

Auren Hoffman (16:50.766) interesting.

Auren Hoffman (16:57.354) Okay. So it's interesting because I always thought the alternative history is like access would be amazing, but potentially the alternative history is that Excel wouldn't be as good as it is today. cause they wouldn't have been investing in it much or, or they, or, you know, maybe like, cause if they, if they kind of like spread their bets, it wouldn't have been as good.

Howie (17:09.195) Well, yeah, I mean, I think, yeah.

Howie (17:17.676) Yeah, maybe. mean, Microsoft also had like enormous resources at the time. So like, I think they could have made both products really great. And I think it was more of like a like consumer confusion issue for them. But I mean, I think there could have been an alternate world where like Excel actually got relegated into a power user tool for like finance geeks, right? And it's mostly a, it's saved mostly a number crunching tool rather than becoming like the dominant default. In fact, I remember like as a kid, when I installed

Auren Hoffman (17:35.906) Yep.

Howie (17:43.448) Microsoft Office, like whatever, like 95, right? Like the one that came with Windows 95. Like, I don't think we got access by default with the bundle. I think it was like, Excel word. Yeah, exactly. And so like, what if access had been the default bundled product and actually like more people need effectively a database than a advanced number crunching spreadsheet that can do IRR? Like, it could have been very vice versa.

Auren Hoffman (17:51.948) No, you had to pay more. It was like a professional to get access. Yeah. Yeah.

Auren Hoffman (18:00.608) Yeah, more people would have used it.

Auren Hoffman (18:10.67) Yeah. I mean, the first like web application ever built was built on Microsoft access. Um, and I put Microsoft access like in the cloud, and then I built a whole web application on top of it. And it was like, it was essentially like the database, like, I mean, I'm not a very fast database. So obviously it didn't scale that well, but the database running underneath it. And then like, then you have to like upgrade. other thing is like, yeah, they forced you eventually upgrade like SQL server back then. Right. And so you have these other kind of.

Howie (18:16.192) no way. Yeah.

Howie (18:22.54) That's awesome.

Yeah.

Howie (18:31.435) That's awesome.

Howie (18:35.863) Right. yeah, yeah, yeah, yeah.

Auren Hoffman (18:38.861) It goes over with this like a terrible product and really hard to use. it, okay. Well, okay. Now another question I have for you. So I'd say like maybe four years ago, when I was considering hiring like a marketer, my basic question to them was like, how good are you at Zapier? And if you were not good at Zapier, like if you, first of all, you'd never even heard of it. Like the, interview would end. Right.

Howie (18:50.25) Yeah.

Howie (18:58.58) Hmm, interesting. Yeah.

That's right.

Auren Hoffman (19:05.29) If you, and then if I would just like show me your zaps, right. And they would show me. then like, you were, if you were, if you had great zaps, like, okay, you're hired. And if you don't, you're not. now today, like, like outside of air table, like what, what's the stack that like all good employees should be using.

Howie (19:08.468) Yeah. that's very cool. Yeah.

Howie (19:14.473) Yeah, yeah.

Howie (19:22.903) Yeah, well, so first of all, I think that's really interesting because, you know, to me it relates to kind of this founder mode concept, which I think is a misnomer. Like, I don't think founder mode is just about like founders running a company the way they want. I think it's actually about saying like executives in general should be very hands on and in the details and be like experts at the actual craft rather than just delegators and managers. Right. And I think like, you know, that idea of like

Auren Hoffman (19:43.265) Absolutely.

Howie (19:50.399) ask a marketer, especially if they're even if they're a marketing leader, I assume like you would have asked that even if like a CMO, like if you can't get hands on with a modern stack of tools that make you 10x more effective, then you're clearly not up to speed, right? Like you're not in it enough. And I think nowhere, no, no time has that been even more salient than now with like all of the AI tools that you can use. like, you know, one, we use Airtable a lot for everything, right? So like,

You know, actually one of the top use cases in general for Airtable out there in the wild is marketing operations. So everything from like managing a content pipeline, like for this podcast, you could say, Hey, here's all the upcoming speakers. Like, you know, now I can have like AI agents in Airtable do research on each person and find recent news. Okay. That's awesome. Good guess then. You know, but like you can, you can also like have it. The cool thing about what you can do with Airtable is that think of like any one-off prompt.

Auren Hoffman (20:32.898) the way we do do that. We do that. Yeah.

Howie (20:46.644) that you can go and create in ChatchBT, including prompts that leverage web search, right? So like, let's say you're researching one speaker that's upcoming on your podcast. So here's a specific prompt that in a very, very nuanced way finds relevant details about them and their history, et cetera, using, you know, basically AI web search, right? And once you have that prompt kind of optimized, now like the current status quo with chat products is like, you'd have to go and remember to copy paste that exact prompt and give it the right...

input contacts, like if there's more data that you have about your own notes from your own notes about a speaker. And it's quite like laborious to go and like replicate that same AI use case again and again, right? And effectively what you can do with Airtable now is like bake that prompt into a workflow. Like it's just, you know, define a field that performs this agentic action, in this case, web search, and do it every single time you add a new podcast guest or upcoming episode, right?

And so think that's the really powerful idea that we have with Airtable is, you know, take any AI use case and obviously like, you know, we're all standing on the shoulders of the giants that are the awesome foundation models that we all now have plenty and cheap and scalable access to, and they just keep getting better, cheaper, faster, right? And like on top of that though, how do you build like a true like application, which is effectively just a container holding the right data model, the right workflow and the right kind of human interface.

to be in the loop with the agent and AI kind of applications. So, we do a lot with Airtable and I would expect any marketer or frankly any role of person that is coming into Airtable today, if not to be an Airtable expert, of course, like we can't only hire people who already know Airtable that well, but like I would expect them to be able to, including in the context of a interview process, to like whip up a pretty good Airtable implementation.

for a relevant use case. So you could be a lawyer or somebody in our legal team. Can you go and build a contract review workflow in Airtable really quickly? Like that's the kind of systems thinking required, I think to really leverage AI in the modern era.

Auren Hoffman (22:58.754) Nope. if you're using like a Zapier or, maybe a newer version of a Zapier or like a pipe, Jim or an end to end or something like that, like they charge on a per action basis. You know, you get X number of actions and you pay. And then if you go above that and you can get your entire team on it. So we have like our a hundred percent of our team using it and they're all using it they're all benefiting from it. Airtable is still kind of old school, right? It's her user.

Howie (23:10.526) Sure, yep, yep.

Howie (23:20.969) Yeah.

Yeah.

Yeah, sure.

Auren Hoffman (23:28.482) Right. And it's, and, so you, it's very limited to, like, you know, if you think of like air table versus Google sheets, I love air. I way prefer air table over Google sheets, but Google sheets is probably has, over a hundred times the use of our company because Google sheets is there's no marginal cost. don't have to, I don't have to pay an extra thousand dollars per user to go use it. How do you think about pricing? Cause like, I think you're, you would.

Howie (23:35.538) Yeah.

Howie (23:39.55) Yeah.

Howie (23:48.735) Sure, yeah. Yeah. Basically free.

Auren Hoffman (23:56.706) Like you, would explode in usage and I think the revenue would go up. might not. wouldn't maybe go up with the X relative to the explosion, but it'd go up. Like, how do you think about pricing and you know, et cetera, as, you're, I'm sure you've thought about this a lot.

Howie (24:00.349) Yeah.

Yeah.

Howie (24:08.841) Yeah.

Yeah, I look, think pricing is misleadingly complicated or deceptively complicated, right? Meaning, you want to have a pricing model that is ultimately simple and grockable by the customer. But there's a lot of, it's like multi-dimensional optimization underneath the hood, right? Meaning, now, especially if you want to do consumption-based pricing, okay. But then what's your actual inference cost? Especially since usage now sometimes correlates with actual hard costs.

And it doesn't always correlate one to one, right? Like you're not necessarily always like saying like, you know, let's say like you price for every zap in the Zapier model, but some zaps take 10 times more costs, including inference costs in between, because they have a reasoning step with an AI agent. And so are you absorbing that kind of tolerance for like, you know, a variable margin on each of these costs? I think it gets really, really tricky. the trick

Auren Hoffman (24:43.99) Yeah, of course. Yeah, changes.

Auren Hoffman (24:50.456) Yeah.

Auren Hoffman (25:05.224) But that's the same on a per seat basis too though, right? mean, you have seats are just going to cost you way more than others. Some people are logging in all the time, right?

Howie (25:08.966) Is that,

Howie (25:13.886) For sure, for sure, exactly. And maybe even like arguably more so with the seat-based model, then you're, you know, I think like the seat-based model is effectively, you know, on the one extreme of saying, look, like we're going to absorb the variability of cost structure and abstract it away from the user or the customer so that they just see a very simple mental, they get a very simple mental model, which is like, okay, I have people and like anyone who wants access to this product, like we're going to pay this amount, right?

And of course there's some sliding scales of like, which plan do you want to be on and like what volume discounts, but like for the most part, it's like, how many people do I have that need to use this product and I pay for each one, right? And that somewhat like makes sense. That makes a lot of sense, I think to a lot of customers because like they still think about the value of software as like, you know, really an enhancement to human productivity. I think there's a whole argument around like, well, how quickly are we going to see like agents actually take over jobs?

and outright replace humans, in which case, like, you know, the value of an agentic solution is not correlated to the number of people using it. In fact, it might be inversely correlated, right? Like if you can displace like a hundred roles with one agentic solution, you know, that's actually worth more. You need to do outcome space-based change.

Auren Hoffman (26:25.74) Yeah. Or I mean, maybe there's needs to be a, so like, mean, like giving them one of the companies that we have. So they have about 60 employees. and every month they have between three to six air table seats. so they scale up and then they immediately scale down because like it's really expensive to have these. They should have 60, right? So they should have 60 people using it as an extremely powerful tool. have 60 people on Zapier. Right.

Howie (26:31.825) Yeah. Yeah. Okay.

Howie (26:40.797) Okay.

Howie (26:46.387) Okay, yeah.

Yeah. Yeah.

Howie (26:54.023) Mm...yeah.

Auren Hoffman (26:56.034) or whatever they're using pipe dream, they have 60 people on it and they're constantly using it all the time. But like they, because, because it's so expensive, like they, like, they like scaling, like it should be like, maybe you have like these, maybe you have like the five expensive seats and then you have like the super cheap seats for people just need to do a little bit here and there or something. I don't know the way to do it, but like, I think like most of the company is not, not using your product, which is such a great product. Cause it's so, cause of

Howie (27:14.12) Right.

Yeah.

Howie (27:22.961) Yeah.

Auren Hoffman (27:25.195) because of that price barrier.

Howie (27:27.451) No, that's very fair. I mean, there are like, you know, so we do have a free tier and you could add, you know, different people to a free workspace in the company, right? So like 50 of the people could always be on a free workspace and they can use like the air table free plan.

Auren Hoffman (27:41.25) But if you have a paid workspace, if you have like a paid workspace with three users and you want someone else to be interacting with it, then you have to pay for that thing. Right.

Howie (27:50.63) That's right. Like everybody in the workspace, that's not, we do have free read only users, but if you want them to be like a full editor, they do have to play at, at the same plan level. Right. So like

Auren Hoffman (27:58.678) Yeah. And then, again, like even edit like one tiny thing somewhere then they have to pay. Yeah. Yeah. One time. Yeah. Yeah.

Howie (28:02.618) Yeah. Yeah. Yeah. So that's fair. And look, I I think it is a question that predates Gen.ai and, you know, kind of like agentic or consumption based pricing around AI. Actually, I recall like a really interesting conversation I had with Zuck where, you he's giving me like some advice about thinking about the business and like he had a radical proposal for me at the time. This was like probably like at least five years ago.

And it was basically like, what if you just made like Airtable entirely free and like just go for like mass adoption. And instead of doing any kind of seat based SaaS revenue, like you have this marketplace thing and we just launched this like, you know, effectively like an app platform where like third party developers, as well as us could build like apps that sit on top of Airtable, right? And he was kind of like, what if you just like went all in on like getting scale?

Auren Hoffman (28:38.84) Yeah.

Auren Hoffman (28:54.315) right, right.

Howie (28:58.482) and your eventual revenue model could just be taking a skim on the AppS Bulletin top of table. I remember like, yeah, and obviously, mean, totally. look, obviously, Zocky is a really, really smart guy. So I was deeply thought about it. And we actually didn't end up doing that because, I felt like we could get to some hybrid alternative with a combination of freemium and other pricing factors. And also, SaaS is...

Auren Hoffman (29:04.187) right, right. That's interesting. Yeah. Yeah. I mean, it's radical for sure. Yeah.

Auren Hoffman (29:12.75) Yeah.

Howie (29:25.785) like, you know, are we like more of a SaaS business or are we more of like a consumer product business? Right. And then, you know, it's a, it's a valid question. Like we could have gone either route, right. And aim more for like a consumer as to tool. And we obviously, you know, kind of picked the road of more sassy and like, and we were also moving up market into enterprise and able to command like, you know, really large ASPs and enterprise where we have multimillion dollar, you know, paying customers who are building mission critical operations on our table. Right. Where, I mean, the value to them is like,

Auren Hoffman (29:28.706) Yep. Yep, yep.

Howie (29:53.977) higher than maybe they even get from like a Palantir solution in some cases, or like a solution built on a service now or Salesforce. So I think like, you know, there is some like, you know, kind of strategic direction and vision that is embedded into the pricing model as well. Like, are we aiming more for like the mass scale, mass everybody using it, or are we aiming to like, you know, go and serve the highest end use cases? Can we do both with some creative combination of pricing? to your point, like maybe there is an option where like we do have like a

more restricted tier of seats so that you can just add everybody in the company to that workspace. They still can collaborate, but like if they're a very limited or infrequent collaborator, you charge a lot less. So I think those are all valid questions. know, things that like we, we definitely, yeah. Yeah. Yeah.

Auren Hoffman (30:26.466) Yeah.

Auren Hoffman (30:37.87) just want to like, I just want to get off Google sheets. Like I want you to help me get off Google sheets. And I, right now I still spend, even though I, even though I love air table, so like I love air table 20 times more than Google sheets. I'm in Google sheets 20 times more time. Um, and so you, I'm just like, I'm just, and every time I'm there, I just curse you. I curse you, how I, I'm like, how he did, let me get off Google sheets.

Howie (30:50.192) Yeah.

Yeah, yeah. Well, that pains me. That pains me to hear. Yeah. Oh, OK. Oh, all right. Well, I'm going to I'm going to record a loom of you cursing me and just autoplay it until I finally figure out the fix to this. But I think it's a fairly push. And like, you know, I do think now, especially in the agentic era, like we have even more reason to rethink pricing, right? Like, you know, we just did this big like AI native relaunch. And one of the things we did was we actually bundled

Like before we had AI as an add-on to Airtable, right? So you have the seat price of Airtable proper, and then you basically had to be on a paid plan of Airtable to then also pay for an add-on of AI, which also then came with some, like basically inference credits to run AI use cases, right? And we basically now said, no, like everything needs to be agentic by default. Like even the free user should come in and get like a vibe coding like app building experience.

that we're going to eat the costs on, right? mean, inference is like, you know, cost money and like, we're just going to eat that cost even on our free plan. And so we've already kind of like made some changes to our pricing thinking and gross margin thinking to be able to support the type of like new product paradigm. Like it wouldn't be that radical or it would be very natural for us to then continue to like think about like, okay, what are new disruptive ways that we can just get more people to get value out of their table, right?

Auren Hoffman (32:14.53) No, how do you route the essentially the, the, the AI requests to the, to the best model. And it's both the model that like, well, both perform the best on that specific topic, but also doesn't like crazy, you know, there's a lot of models, which are super low cost. You can even run them locally. so how do you, how do you, how do you do that yourself? Like, is there like a sort of layer that sits on top of it and is a routing system?

Howie (32:22.99) Yeah.

Howie (32:26.372) Yeah. Yeah. Yep.

Howie (32:33.776) Sure. Yeah.

Howie (32:39.598) Yeah, yeah, yeah.

Yeah, for sure. So we have basically two very different but complimentary types of AI in, in Airtable, right? One I call like AI building and AI at runtime. AI building is basically where you come in and this is the vibe coding experience. We have Omni, this, this agent that you talk to and can build or analyze any app in Airtable, right? So like, you can tell it build me this dashboard in my CRM.

go create me this new app that is a podcast tracker, go add this new object or field to my existing CRM. So it's kind of the app building agent, right? And that one, we basically do have an underlying system where we can choose different models for different types of requests, right? And some we optimize for latency so we can very, very quickly do a simple request. Some we optimize for quality where...

Like actually it turns out the reasoning models always don't always do the best at text processing type tasks, right? So if you give me a bunch of notes and you say, I just want you to like extract out the key people referenced in these free form nodes and then go and update those notes into each of those people records in my CRM. Like actually we found that like maybe a non reasoning model like a GP 4.1 actually can outperform the reasoning models in a lot of these use cases. So like we're doing a lot of

Auren Hoffman (34:03.191) Yes.

Howie (34:06.543) our own selection of what is the best model to use for each of these different types of requests or layers of requests.

Auren Hoffman (34:13.054) How does it like, how like, just kind of know based on what the request, like, how do you even know what the request is and how to route it? Like.

Howie (34:20.943) So, I mean, that's an LLM step itself, right? So there's like the LLM step of like, and I mean, so one, like evals are super important. I think this is something that like, you know, is probably underappreciated, is like, with these, you know, building an agentic experience is so unlike traditional deterministic, like coding, right? Like when you think about like your algorithms class from CS, it was like very mathematical. There's like a closed form answer to like the optimal way to do

Auren Hoffman (34:24.023) Yeah, OK.

Howie (34:50.159) like a sort of a list, right? Whereas with these agents, it's like you kind of, do have to like do some trial and error. And by definition, like having a really good eval setup is how you can very efficiently like kind of navigate the somewhat stochastic nature of LLMs, right? Like you don't know predictably like how well they're going to do for any given prompt until you try it.

And so you just have to try it and evals are the way to try them in a consistent, you know, kind of quantitatively measured way. So we have a really good eval stack where we effectively can go and like just try lots of these different combinations. can A-B test like, okay, what if we use this more efficient, smaller reasoning model to, you know, kind of decide whether to call on this, this bigger, slower reasoning model for this type of request. I mean, there's literally like, I have a internal kind of like,

power user config mode where I can actually go in and like hand override the different models we use at different layers or for different types of requests. And like there's like literally like a dozen.

Auren Hoffman (35:55.214) Oh, wow. Okay. And I'm letting you like, you to, have to continually doing that because like all of a sudden you can be like, wow, we could save like a hundred grand a day if we just do this thing. And it's like, it's as good. Sometimes it may even be better for the user. Yeah.

Howie (36:05.272) Yeah.

Howie (36:08.559) Sure. Well, that's the thing is like, would never, I mean, I think we're still at the point where like, we're not really cost optimizing. Like I would rather eat as much cost as we can in this moment to deliver the best quality experience. Right. And I have high confidence that like cost will come down naturally for any given quality level. Like, I mean, look at 03, like Sam Altman literally tweeted like a couple months ago, like 03, we just 5X or decreased the price right now by 5X. Like that's insane.

Auren Hoffman (36:35.67) Yeah. Yeah. It's insane. Yeah.

Howie (36:36.598) Like literally as of this tweet, we dropped the pricing by 5x, right? And so there's a massive race to the bottom and there's like software optimizations to inference. There's obviously the hardware costs going down for like the same amount of like, you know, effectively flops, you know, of compute. But...

Auren Hoffman (36:51.118) By the way, he's taking the, he's taking the Zuck pricing strategy kind of like he's, he's trying to get mass adoption, not trying to optimize. Yeah.

Howie (36:55.17) Yeah

Totally, totally. And I think that's probably the right thing to do in their race, which is like, I think there is a huge advantage to the provider that has kind of the like cognitive referent or default set of models, right? Like that not only people use through the consumer product, Chatchmutee, but then even like programmatically through API, like if everybody just like learns the ins and outs and the nuances of like the opening I models and the texture.

Auren Hoffman (37:17.164) Yep, yep, absolutely. Yep.

Howie (37:27.15) of 03 over let's say 4.0 Sonnet from Anthropic, then there's huge long-term strategic advantages to that, right? It's becoming like... Exactly. The memory part for sure on a per customer level makes it much more sticky, but I also just think it becomes a standard, right? It's kind of like Excel, going back to our earlier discussion, like Excel becoming a standard. And it means that, I remember in the Windows 95 era,

Auren Hoffman (37:34.848) Especially if there's memory, especially when they have memory, right?

Auren Hoffman (37:46.573) Yeah.

Howie (37:55.821) I think my family was poor and we actually didn't want to spend the $300 on Microsoft Office at the time. And so we bought some knockoff version of it. I forget what it was called. It was Word Perfect, actually, I think, and the associated products. And it was just not as good. And I remember going into school. in our school, we had a computer lab. And we learned Excel and Word and all the Microsoft products there.

Auren Hoffman (38:03.288) Yeah.

Auren Hoffman (38:09.619) we're perfect. Yeah. Yeah. Yeah. Of course. Yeah. Yeah. Yeah.

Auren Hoffman (38:23.18) Yep.

Howie (38:24.621) and I learned all the shortcuts and the advanced features and I went, oh, it's like, well, the thing that we have doesn't even have all that, right? So having the standard where like people just like learn developers and consumers alike, learn, yeah.

Auren Hoffman (38:30.05) Right.

Right. I mean, you are, you're like, thinking by your, this is why I got to get, I want air table. Everyone, you, everyone using air table as well. Yeah.

Howie (38:39.413) Okay, all right, all right, you convinced me. we're gonna mark my words. We will do some radical improvements to our pricing model to expand the funnel by far of people using the product. I'm very aligned on that.

Auren Hoffman (38:48.206) I want every kid in school using it.

Howie (38:54.741) Well, do think, mean, like on this note of drawing the parallel back to like Chatch-PT, I do think in the AI era, like there's clearly some products that have been successful doing an enterprise sold motion. Like, you know, some of that come to mind are like, you know, the Harveys of the world, like, you know, very much enterprise sold, even Palantir's like custom AI deployments, right? Yeah. But like, but I think by and large, the vast majority of real AI value

Auren Hoffman (39:15.436) Yeah, of course. Yeah. Yeah.

Howie (39:24.054) creation and adoption and even revenue has come through consumer led or like end user led adoption, right? Like PLG, like, mean, look, chat to be T alone. Yeah. I mean, touch me T alone. Like I think opening eye is now like a 10 billion revenue run rate business. Right. And I assume the vast majority of that is chat to be T. And so like, I mean, chat to be T alone probably is like the lion's share of total AI revenue out there today. Right. True. Gen AI revenue, at least.

Auren Hoffman (39:30.53) know, bottom up referrals, talking about it. Yeah.

Auren Hoffman (39:38.936) same.

Howie (39:52.012) And so like not to mention like you have the cursors of the world, you have the mannaces, et cetera, that are all like PLG led. And so I think there is a strong favoring of like, you for AI experiences, better experience, it's better shown and like, you better yet, like just self-served rather than sold or told, right? And so I think even more.

Auren Hoffman (40:09.878) Yeah. Yeah. And no long-term contract easy to just, you know, try it out. So some of these do have much higher turn than others. but, but it's, but it's good.

Howie (40:16.17) Exactly.

Howie (40:20.0) Yeah, but you know, like I'll take a 10x bigger funnel with a 2x higher drop off, right? You like that's still, you still net out way better.

Auren Hoffman (40:27.148) Yeah. Yeah. I know salespeople to pay and you know, all this other stuff, right? Yeah. So your tax are much lower. Like if your tax go down by, by 20 X, like who cares if your churn goes up by three acts.

Howie (40:30.6) Exactly. Exactly. And I think that...

Howie (40:39.767) Totally. Yeah, yeah. I mean, your cat could be zero if you're trying to be teet, right? It's just like massively viral. Like people just talk about it. And I actually, I mean, I've seen the retention curves for...

Auren Hoffman (40:43.906) Right. Right. Right. That's right. Yeah. Yeah. Well, actually, I don't know that I bet you their cacks are like insanely high because most of their cacks are like Sam tweeting and Sam's time is probably worth, I don't know, a million dollars an hour or something. Right. So it's probably like the cacks are probably like insanely high. If you put his hourly rate on it. Yeah.

Howie (40:55.818) Alright, okay. That's fair. That's fair. That's fair. Okay, well... Okay, yeah, although you could say he'd probably derive some amount of like personal entertainment value from the tweeting. So is that actually just like a moonlight job or his actual like paid job? Yes, exactly. And at one point at least he had like, he said he had like zero equity in opening up. So actually maybe his time is free at that time. So...

Auren Hoffman (41:10.382) Good point. Yeah, yeah. Okay. Yeah. So he wouldn't maybe he would have done it anyway. Or, you know, a good boy. Yeah, yeah.

Auren Hoffman (41:24.598) Right, right, right, right. It's a point. Yeah. From OpenAI's perspective. Yeah, exactly. Yeah.

Howie (41:25.292) Yeah, exactly. yeah, mean, think like punchline is, you know, I do think like within the era of AI, the importance of like getting just more people into the top of funnel to just like actually use and try your product is more critical than ever. you know, to your credit, like on the pricing model stuff, it's like, you know, we definitely want like, I think Palantir short term bullish on long term uncertain on

I would never bet against them, but short term, they've clearly found this repeatable motion of selling a turnkey way to just deploy AI into the enterprise. Longer term, I think there's a question of, well, does all of the AI deployment just become self-serve and so easy to spin up that companies don't need the current Palantir very heavyweight sold model? They send an FDE, you're paying millions of dollars for a custom build. Do they just go and...

Auren Hoffman (41:55.79) Yeah.

Auren Hoffman (42:21.228) do think when you're dealing with, like when you're dealing with comp, like, security type of information, information that is national security related or something like that, you probably still gonna need to be sold to for quite a long time. Yeah. Yeah.

Howie (42:34.6) for sure, for sure. Definitely for that. Palantir today is actually now already trying to sell down market as well, right? I mean, even like smaller companies that are not dealing with like high security issues doing like much what would be considered like much more banal deployments. Like here's like how we're deriving insights from customer feedback. Here's how, yeah, yeah, yeah.

Auren Hoffman (42:55.338) really? Okay. I don't, if I ever saw a random company doing that, I was like, okay, that's not a good deal. you, Palantir is like, you be paying Palantir at least $5 million for your implementation. If they're trying to sell like a hundred grand thing, that's probably not the right thing for that.

Howie (43:02.214) Heh.

Howie (43:05.738) Well, think what's, yeah. I think there's this like near term windfall that Palantir is getting around like CEO led, you know, kind of purchases of Palantir as an easy button to show that they have some AI implementations, right? It's like the CYA, cover your butt to the board and to the public market maybe. Even if you're a smaller company and not in security, not in like high data sensitivity to just be able to say like, I have an AI deployment, right? And I think.

Auren Hoffman (43:21.964) Yeah, okay. Yeah.

Auren Hoffman (43:34.476) Bye.

Howie (43:34.889) I think that like as the self-serve ways to do that and do that in a meaningful way, not just like a gimmicky or like, you know, like toy way, like early on people were, you know, people were all using touch me T for like superfluous, like fun gimmicky things, right? Like people were like trying like, you know, the, the like talk like a pirate thing or whatever. Whereas now I think they're actually figuring out how to use it for, more meaningful use cases and better yet, like using the underlying models in a tool, like an air table to then like power a more recurring workflow that is actually

high value, right? And maybe it like, you shouldn't be paying like hundreds of thousands or millions of dollars a pound here to like do a bespoke build for.

Auren Hoffman (44:12.62) And what there's a ton of Acre hires that are going on in our industry back in the day. And Acre was like a pejorative word. today Acre hire could mean like you're getting like a billion dollars or more than a billion dollars. Like what's your thought about where we are in this like weird moment and how does that, does it continue? Does it stop? Like, is it just like this odd, like weird like moment in history?

Howie (44:15.582) Yeah. Yeah. Yeah. Yeah.

Howie (44:24.892) I know, it's,

Howie (44:30.409) Yeah.

Howie (44:35.283) Yeah.

Yeah, I mean, it's totally changed. I mean, like I was acquired in my first company by Salesforce. Right. And at the time, like that meant like maybe a low seven figure payout and like some some retention plan or whatever. But like, I you look at like the recent acquires of been very, very recently, Windsor for scale, et cetera. These are like, you monetary sums that look like full business buyouts. I mean, like the scale, 14 billion dollar price tag. I mean, that's like

Auren Hoffman (45:03.693) Yeah.

Howie (45:05.844) bigger than a lot of public companies, Like, mean, mature cash flow generating.

Auren Hoffman (45:08.41) It's an actual hire where the founder gets into the Forbes 400, right?

Howie (45:12.916) Totally, yeah, exactly. And the crazy thing is, it is truly an acqui-hire because partly maybe to avoid antitrust scrutiny, it's like they're not actually buying the business. They're not buying the company, the entity, the business, the IP. They're intentionally leaving that behind. And in fact, like scale the business is still an independently operating company, right? And so I think to me, it's like a little bit of the key person theory, right?

you know, kind of human civilization, like how much of technological progress or just like, you know, progress is driven by like, you know, a couple very, very important key actors who have the unique insight or wherewithal to just really move things forward. And it seems like, and I would generally agree with this, like that right now in this moment, given the ingredients of like all of the potential things you can do with AI, and especially if you have massive resources like a meta,

that you can gain leverage with. Like if you win at AI, that is worth trillions of dollars to you. But if you lose at it, like you've basically become irrelevant. Like I think we are at this moment where it's not just about hiring up an army of like a hundred thousand people to work on a problem. It's about finding like the critical people who like really have the combination of like insight, ingenuity, but also just extreme like motivation and intensity to make big, things happen, right? So like,

I think it's actually somewhat rational for these acquires to happen, not just because they have to be done this way to avoid the antitrust issues, but because Meta is basically saying, look, I value Alex and the team that's coming along so much to really accelerate my efforts on AI that it's literally worth, mean, 14 billion is not even that.

big of percent of their total market cap, right? Or cash on hand.

Auren Hoffman (47:09.358) But, but, but Meta values Alex, who is, know, going to be an executive at Meta important person, but not the CEO, not, not anywhere close. they value him way more than Apple values Tim Cook in terms of compensation. Like, like, like, like, like, order of magnitude more. They value Alex than, than Apple values Tim Cook. Right.

Howie (47:17.789) Yeah, yeah. Correct, yeah, yeah.

Howie (47:26.76) yes. mean, monetary, financially. Yeah.

Howie (47:36.678) Yeah, yeah.

Auren Hoffman (47:37.134) Um, and so like the CEOs of public companies are less valuable than often like the, the 100th AI engineer at Metta.

Howie (47:49.594) Yeah, I mean, I think it's a very valid point. like, I actually think it's a good system if you can value talent for like rationally based on how much leverage they provide the business, rather than having some artificial, like internal reasons or hierarchical reasons for why you can or cannot pay talent, you know, a certain amount, right? Like, would kind of, I mean,

it becomes like a whole like, you know, capitalism. like, you know, at least in this like specific regard, like Ben and Jerry's tried the experiment at one point of like the best paid employee, which happened to be the CEO for them, like could not be paid more than 10x more than the lowest paid employee, right? Who might be somebody like literally manufacturing, you know, the ice cream in the factory. And of course, like, you know, I believe in like fair wage, like, you know, people should be paid well and like so on. But like, I think like

you also need to just hire the best person that creates the most leverage for the most important jobs, Or the roles. And so I think like it's, you know, maybe an interesting comparable here would be like in professional sports, like you have some players that probably make far more than the like manager of the teams, right? Like, and that's rational because like, if the player is that good, like the team actually wins and then the whole franchise becomes more valuable.

Auren Hoffman (49:13.838) Yep. And the analogy may hold all the way through because it's very possible that this employee is only important in a very brief moment of time for three years. They're going to add value and then they're not going to add as much. Just like a, you know, a great professional athlete may only be good in a very short moment of time. Um, but like, you know, 20 years from now, they're almost certainly not going to be playing on the team anymore.

Howie (49:24.059) Yeah. Yeah.

Howie (49:39.687) Totally, yeah, yeah, exactly. Yeah, and I think like, I mean, this is one of those things that like, feel like I've really, it's like, you know, a known like best practice or, you know, piece of, you know, kind of managerial theory, which is like, you know, it's a team, not a family. You gotta like think of like, you know, each team member is like being a part of the team in this moment. And you can appreciate them for that and always think highly of them in retrospect.

But not every team member needs to be part of the team forever, right? And in fact, like that includes the CEO and sometimes the founder as well. But like it is always about like solving for what we need at this moment in time, because if we don't put one foot in front of the next and like, you know, really execute well in this moment, like we're not going to have the chance to play a good hand in the next moment.

Auren Hoffman (50:26.222) Yeah, that makes a lot of sense. Now you've been doing air table for 12, 13 years now. Like, how do you think about like, you know, okay, you, how do you think about the next 12 to 13 years? How do you think about your future and what you're doing and what does one, you know, obviously build a great company, but like, you know, um, how do you think about this? Like going forward?

Howie (50:35.815) Yeah.

Yeah.

Yeah. Yeah. So, you know, I remember reading this like HBS article a while back called like your strategy needs a strategy, right? And it effectively said like, look, like, you know, the right way to think about strategy is actually market dynamic dependent, meaning if you're in a market or a moment with a very low rate of change, like it's a stable environment and you can maybe call like a certain era of mature sass.

as being kind of an example of this. It's kind of deterministic. You can kind of just execute in a pretty known way. And it's really about predictability and how do you build a really durable business? And a lot of the PE buyouts are examples of this where it's like, okay, we know what we're getting. We're not gonna expect some crazy rabbit out of the hat innovation from this business that we bought, but we're just gonna chug in day in, day out. can probably continue to scale this. Whereas if you're in an environment where like,

there's an extremely high rate of change or like a lot of uncertainty, which I would say we're definitely in across all of software now because of AI. Like I think you have to just be completely adaptable. Like agility is much more important than like having like a clear known long-term pathing of we're going to do this and this and this and this. And this is exactly what's going to happen. Right. And so like to me, it's about like playing in the right space. First of all, like, we picking like the right vision and the right category?

Howie (52:09.338) to disrupt and maybe the category doesn't fully like look baked yet, right? But no code to me is absolutely like a category that holds even more relevance and is even more exciting in the agentic era, right? Like by def, like literally if you interpret the words no code, like what can people who don't know how to code do with software or app building that only gets more magical and like the market only radically expands in the AI era. So are we picking in the right space? Are we playing the right space?

I feel very confident the answer is yes. And then, you know, it's less about like, what's our, like, let's predict 10 chess moves out what we're gonna play. And instead say like, do we have a good hand right now? Like, do we have the right pieces?

Auren Hoffman (52:50.146) Like, but like just asking like more of a personal question, like at some point, like Zuck still there, but you know, he's, hasn't been there as long as like Jensen Wong has been there forever. Right. He's been kind of doing this crazy, but then, you know, Larry Page is no longer CEO. Jeff Bezos is no longer CEO Bill Gates, like left a long, long, long time ago. and like at some point, like people want to go do other things.

Howie (52:53.369) Yeah. Yeah.

Yeah. Yeah. Yeah.

Howie (53:06.607) Yeah. Yeah.

Howie (53:11.523) Yup.

Auren Hoffman (53:15.65) How do you think about it personally? Like they, I'm sure they still have those challenges and those other types of things. Gates left like right when the internet was happening. Like it could have been, maybe he should have stayed, but he decided he wanted to do other things.

Howie (53:16.111) Sure.

Yeah!

Howie (53:23.812) Yeah.

Howie (53:28.173) Yeah, I mean, that's totally fair. So to me, it's like, there's two questions. One is like, do I have the energy and the motivation to keep doing this, Period, right? Or doing anything, right? Versus like, you know, some people go into like semi-retirement, maybe they do like an easier job or whatever. There's nothing wrong with that. like, yeah, it's like the hobby projects or like, you know, and no offense to anyone like, you who goes into like permanent like board role, you know, kind of state, but like,

Auren Hoffman (53:38.008) Yes.

Auren Hoffman (53:45.422) Yeah, yeah, they're on the wine or winery putting around and stuff.

Auren Hoffman (53:56.078) Or they become, they become a venture capitalist. You know, that's the typical retirement, right? Yeah. Yeah.

Howie (53:56.262) There is a path where you no longer want to be... VC can be hard as well in very different ways. But do you just not have the energy, and it's fine to admit it, to not be an operator anymore, full-time operator in intense mode. And then two, even if you are motivated to be an operator, is this no longer a space that interests you? And I can definitively say on the number one front, I am more fired up than ever before.

Auren Hoffman (54:18.498) Yeah.

Howie (54:22.797) Like, I mean, look, I mean, maybe partly it's function of age. Like I'm thankful that like I got my start very early. Like, I mean, you, decided the math early, but like I'm only 36 right now. Right. And so I feel like I'm just beginning. and I've just kind of learned the basics of like how to be a somewhat effective operator and founder CEO and whatever. So like I'm nowhere near ready to retire. And then on the two front, like, is this the right space? I think, I will say like, I think.

Auren Hoffman (54:28.182) Yeah, Yep.

Auren Hoffman (54:34.094) Totally.

Howie (54:50.571) If we were reaching this kind of mature, like, you know, this starts to look more like a PE type of business where like you just kind of do the same thing, you're in, you're out. And like the margin of difference between me and some other. Yeah. I mean, it's like, okay, like I would no longer be that fired up about it, nor would I probably be like Six Sigma better than, you know, the, the, median person at it. And so like, there's probably somebody else who could do that equally well or better. Whereas now it's like, this is really a refounding moment and like,

Auren Hoffman (54:58.766) Yeah, yeah, yeah. And then you go get, then you go hire a Tim Cook or something.

Howie (55:18.112) I get to do something that I'm really excited about, which is like, start with technology first innovation, which I think is really important. Like, you know, in the mature market environment, like it's no longer about like finding disruptive tech innovation. It's more about saying like, okay, like SaaS is kind of mature. Anybody can build a database like, or a hard coded database with like application logic on top and create a CRM for X, right? Like, and then there were so many different permutations of X, like here's the CRM for this kind of doctor or this kind of dentist or this kind of.

you know, a plumber, et cetera, you can build a good business that way. But what fires me up is like, I think of a technologist at heart and like, I've always found that like really understanding the cutting edge of what's possible technologically. And maybe in the beginning of Airtable that was like building our own robust single page app framework before there was good prior art, like React didn't exist, Backbone even didn't exist. Like we have to roll our own like rich front end experience. You know, we also had to build our own like backend and create a real time database engine.

and a real-time collaboration layer, et cetera. All these things were actually hard technical problems that we also combined with a creative sense of taste of how do we not just build tech for tech's sake, but make this an awesome and delightful product experience. And we're going to learn a lot while doing it, right? And so to me, it's like the new Airtable has every bit of that and more that we had in the beginning. And so it's like I get to go and play with all the cutting-edge tech.

What, how do you build agents? How do you build like, you know, new things with models? How do you go and use existing products off the shelf? And then how do we like creatively think about like, re-productizing that into a new form?

Auren Hoffman (56:57.88) Now a couple of personal questions. What is a conspiracy theory that you believe?

Howie (56:59.652) Yeah. Oh. Um, I don't really have that many, like, conspiracy theories. Like, I guess I would put it this way. Like, I generally, like, I think I actually hold, like, the world as a, you know, I hold a very stochastic view of the world, which is like, I don't think, like, I know how everything works perfectly. I don't think anyone knows how everything works perfectly. And I think there's almost this just like...

this certain level of randomness and just known unknowns. so therefore, it's like, I think to me, it's never about was the lunar landing definitely real or was it fake? Well, probably given my priors, it's like P99.9 % likely real, but 0.1 % maybe fake. I just think, yeah. I just watched that comedy movie, Fly Me to the Moon with Channing Tatum and Scar Jo.

Auren Hoffman (57:44.536) Yep. Yep. Yeah, I'm in the same place as you are, I'm in that one. Yeah.

Howie (57:55.139) And it's literally about this. And obviously, the whole thing was meant to be comedy. I found it to be a really... I just love to think about the what-ifs or the possibilities. Not because I then come away thinking, oh, it's 100 % faked. But it tickles my fancy to imagine alternate realities and then hold both those things. I don't know if there's a big conspiracy theory I hold.

Auren Hoffman (58:23.854) All right. Well, last question. ask all of our guests what conventional wisdom or advice do you think is generally bad advice?

Howie (58:31.063) Well, I definitely think the stuff that Brian Chesky talked about in that now famous Foundermote talk resonated very deeply with me, which is like, there was an era of Citlain Valley, and this was true for Airtable as well, where the conventional wisdom told to us was like, hey, you gotta go and apply all these traditional ways of running this company, hire traditional leaders who have a traditional skill set, and so on.

there's probably some truth to that. I mean, look, you can't just run fast and loose in every way like a two-person company, even as you reach maturity and need to be public ready and so on. But I also think if you lean too much in that direction, which a lot of companies, including us, did, then you lose the ability to have that agility, which is especially important now. And also, you lose the unique alpha that I think these highly innovative companies that, by definition,

Auren Hoffman (59:20.034) Yes.

Howie (59:28.98) If you innovate it well enough to get amazing product market fit to begin with, like that's your alpha. That was your alpha at that time. And if you can keep doing that instead of just like maturing and milking the original innovation for a really long time, like if you can go create new innovation and like reinvent yourself, like I think that's both profoundly valuable, but also requires a radically different approach to running your business, running your company. Right. And, you know, Brian would call it founder mode. I call it just like, you know, kind of high agency.

like being in the details, being radical, being high agility and disrupting yourself every, you know, at least a couple of years.

Auren Hoffman (01:00:04.334) What, what's the biggest, know, what's the biggest kind of conventional one that you just like, you took out or you stopped doing or whatever, and it just had the most impact in the company. Was it like, okay, we don't need all these weird HR rules or something or.

Howie (01:00:19.36) Yeah. Yeah. I mean, so some HR roles you still definitely need. but no, think it's like one, like how I hire. So, you know, we touched on this a little bit earlier, but I think, like when I think about exec hiring, like execs are not just about like managing people in process. They really need to be experts of their own craft. Right? So if you're a marketer, if you're a product person, you actually have to be good at product. You have to be good at marketing. You, and especially now, like, you know, you have to be like highly fluent in.

like getting hands on with like real AI use cases and building things and like, you know, exactly. Cause like, first of all, it takes fluency with tech and products to be able to talk about those products and to be able to make decisions about them. And now more so than ever, it's all so nuanced and so fast moving. You have to be in it. So detail oriented and in it experts of their craft as executives, I think is one big shift for me. Two is like,

Auren Hoffman (01:00:52.526) You can't just be a manager.

Howie (01:01:17.825) I think that I just kind of accepted as common Silicon Valley wisdom that you have to have all of this process and structure in general in how you do things. First of all, we used to use a lot of the RACI model, which is how you define different multi-party responsibility and accountability. I mean, literally, that's what it is.

Like I think anytime you have like a really complicated matrix of, these five people need to be informed about this thing and this person needs to be accountable to this, but that person's responsible for this. Like you're probably over complicating it. And what you probably should be doing is like picking one person to be the DRI with a small lean team to just go spike on this rapidly move and like, do as much in a zero dependency way as possible. and just like try stuff, ship it faster. Yeah.

Auren Hoffman (01:02:12.366) This is the, you know, like the HR innovation that Amazon had was the two pizza rule, right? It's just like we're going to do pizza and then somehow you just API it to everybody else, whether you're the HR team or whether you're a product team.

Howie (01:02:17.492) Exactly, exactly.

Totally.

Totally. I think to like the implicate, sounds simple, but like the implications are quite profound because it means that sometimes structurally it actually changes the nature of work that you take on, right? Like, I mean, there's certain types of projects that you can't do this way. And in fact, if you prioritize this style of execution, like it actually shifts how you execute and therefore even like how your product or business model starts to look over time, right? Like you become an accumulation of like lots of rapid.

learnings and quick hits rather than trying to make these like huge, slow, single swings that either have to be right or you fail.

Auren Hoffman (01:03:02.222) Oh, this has been amazing. Thank you. Howie Lou for joining us. The World of DaaS. I follow you at howieTL. What does the T stand for by the way? Okay. Great. I follow you at howieTL on X. I definitely encourage our listeners to engage you there. This has been really awesome.

Howie (01:03:09.098) Yep, just my middle name, Thomas.

Howie (01:03:19.506) Awesome. This was really, really fun. Thanks for having me, Auren. And yeah, happy to chat.