Carta’s CMO Reveals What’s Really Happening to Startups

19:50Grant:Oh, wow. So for every new funding round, they're not necessarily giving up more of the company to do that. How does that work?

19:57Corey:They're just getting more money. Yeah, they're just getting more money. Yeah. They don't have to give up as much of the company. They're just getting more money.

20:03Grant:Well, wow. Walk me through how that makes sense. Yeah. I mean, it just means that they don't have to, you know, companies want to invest and they don't have to give up more in order to get that funding from those big, especially the big mega firms. Wow. And is that because that they're just saying we don't have to spend as much on cat, like OpEx, like hiring, that sort of stuff? So it's like, yeah, sure. Probably those things. And we know you could scale faster. So the money will be well spent. Wow. Yep. Wow. Yeah, it's fascinating. I mean, it's amazing to see these valuations go up and the founders are not giving up more.

20:44They're generally just not giving up more of the company than they need to.

20:48Grant:What do you think, Nicole, based on the data that you're looking at? I know you're downstream of the bubble talk a bit, but do you think that all of this is indicating that we're in some kind of AI bubble? Or what's your read on all of these changes? Yeah. I mean, I think, you know, if you asked, for example, Peter Walker, who is our head of insights at Carta and talks a lot about this, you know, it looks like we're in an AI bubble. I think it's about really about we're at a peak and we need to normalize as opposed to a bubble. And I believe the normalization will start happening very soon, especially because we now have these big companies.

21:31And not to say that, you know, there's lots of reporting on like, oh, Google's woken up or, you know, whatever. I don't think that that's the case. I think it was just the creation of the correct strategy that's now being applied. And so I think that that where these, you know, the larger companies really like nailing it on certain fronts with AI is going to start to force a normalization. Like a good example of that for us, for me specifically at Carta is we've become heavy, heavy users of gems. That was like the unlock. For us, we weren't really using, like we were using Gemini less than we were using other LLMs for some of the work that we do from like an answer engine perspective.

22:22But then when we started working with GEMS and started to really understand how to use them and their shareability within our organization and the fact that they would get better over time with more use, it became something we're like, wow, now we're really in it with GEMS. We were not, you know, at the outset using Gems and thought that we were going to be exclusively using, you know, Claude or ChatGPT, et cetera. Now that completely has shifted. So I think we're going to start to see more of that and that will force a normalization.

22:56Grant:So that would be, I guess, technically a consolidation of the tools that you actually need. Does he see it like happening? Yeah, I think that that's true. I think that's ground truth that that will happen over time. I mean, I think there's incredible applications for AI in all corners of the world, especially the physical world that's been sort of maybe less touched by technology over the years. And I hear as an investor, I hear about a lot of interesting ideas in that regard. So I think there's going to continue to be this long tail of highly verticalized but very potent AI solutions. And I think there's going to be, you know, there will be a tranche that is more applicable to certain use cases.

23:45You know, in marketing, for example, I think everything that we're doing around content strategy will remain something that sort of has a verticalized AI approach to it. And then you have the very big platforms that have the benefit of having, you know, the money to fund what's needed for these models, but also sort of the data to do it to.

24:11Grant:Yeah, I wonder if that means then that it'll be potentially harder to raise money as this normalization happens as somebody who's just starting. Because it's like you say, like the momentum is so powerful here where, you know, once you have something that works, you can capitalize it on it so much faster and generate revenue so much faster. and the companies are raising so much more money in order to capture the market, that it's sort of kind of, you're kind of, if you haven't already started, it almost feels like you're already behind and you're done. Yeah. But what's your take on that? I mean, we'll see.

24:46I mean, I think we don't know, right? That's what's kind of exciting about this. You know, I think like what you said, Grant, makes sense, right? Like that seems like that's sort of what has happened in the past and seems like it would apply in the future. But I really don't know. That's part of what's so exciting about the time that we're in. We're sort of watching. We're watching everything unfold real time.

25:08Corey:Yeah. Yeah. And this is very subjective, but it doesn't feel like we're feeling any signs of this slowing down. I feel like from a news volume, from a new company's new tools perspective, I feel like it's consistently moving faster than, say, the previous quarter. If I try to look at it in chunks, I mean, I mean, I don't have any data to base that on, but just as far as from a what we're seeing at the neuron perspective, it feels like there's a velocity, if that makes sense. 100 % yeah and I think some of that too is that you know some of the early you know let's I'm just going to speak for myself from a marketing perspective you know when I had you know early pilots or early exposure to some of the some of the martech or go-to-market tech companies that were sort of early in with AI some of those have gotten better with time right like being you know like using them, you know, different from sort of traditional software, like does using them make it them better?

26:17Like, yes, in theory with product feedback, et cetera. But with AI, it's completely different, you know, like the AI gets better, truly does improve with, with time. And so some of these, some of these MarTech, you know, tools are getting, are getting better, significantly better that year over year in terms of like, who's going to be the next big, you know, the next big thing on the MarTech side with AI, like, you know, like there's such interesting work, I think, happening around things like synthetic personas. You know, how we can, like, I, you know, I have a, I have a dream and of what I think we could see in the future, which would be a company that would allow a marketer to walk a synthetic persona through the world of their campaign fully.

27:12So you could, you know, your core synthetic persona, like you would be right. And actually like walk them through, you know, your out of home campaign in Midtown Manhattan. And then they're riding up the elevator and they see your ad and then they're experiencing your digital campaign. And then they're going to an event and then being able to see what the response is in that like fully realized world and test and iterate from there. So there's, I mean, there's so much more that can be done that I think that is mind-blowing. We just haven't even gotten there yet. Yeah, because I'm trying to imagine how you would do that.

27:45Grant:I mean, I guess you could do it in simulation, but you could also do it with a robot. Yeah, you could do the robot, but it's more like imagine a simulation. Like right now, for example, with synthetic personas, we're able to do things like message testing with them and message proofing with them, doing pitches and getting feedback. Why could we not do that in a holistic way? Like take someone through our billboard campaign plus our other campaign, you know, whatever it is. And so there's, I feel like there's so much more room to run. Totally. And that's why I'm optimistic that I don't think we're going to see this, you know, I think we're, I don't, at least from a marketing perspective, I don't see us having a reckoning of some sort where like, you know, 90 % of, you know, AI platforms are going to be subsumed by the bigger players coming in and sort of taking over.

28:34I still think there's amazing building that can be done.

28:39Grant:Yeah.

28:39Corey:That's fair. Yeah, I think it's easy to forget how early we still are. Very early. And the fact that we're talking about a technology that theoretically will touch every element of business and life, most likely, in one way or another. Whether that's, you know, your car or how your coffee's made or, you know, what's running lead generation at a company.

29:04Grant:uh right you know we were just talking today about the idea of like meta just having a or meta and google just having ai in their ad funnel that just constantly creates personalized ads for the for you to target that person yeah definitely gonna happen i mean yeah the whole we haven't even talked about that but like you know the whole world of advertising is about to dramatically change um like that i will like stand on business on yeah yeah yeah yeah i mean it's there's everything you know at least for me um you know the moment that these tools like like chat gpt started getting traction i'm like there's we're gonna have ads before we know it yeah because why why wouldn't why would you not sort of i mean that's i guess the that's the why would why would they not have ads um so it's just a question of how that gets rolled out i think they're starting

29:55Corey:to roll it out in yeah tranches now yeah yeah yeah and we'll see how it looks and how it turns out and how it's received. I'm curious to see what the approach is and what it's going to feel like. Yeah, 100%.

30:12Grant:What's the coolest thing you've seen from a marketing perspective where you're like, wow, like you were blown away by how they use the tools that are available today for your domain? So many things. It's really hard to choose. Well, let me start with something that I'm so proud of that my team has done. Yeah, yeah. So we have a couple of internal champions on our team around building in Claude, building in Gemini. And there's this woman, Ryan, on my team, Cesar, on the PMM team. They're both product marketers. And they've become leaders for the whole organization around how you can really tackle big problems using AI.

30:56And one of the things that we were very interested in doing is thinking more about marketing and sales orchestration in terms of like what happens after. So we, you know, someone is a lead for us and they talk to a salesperson, they get a demo. Like what's what really like how does that conversation go and how should we think about how we structure the next conversation that happens? And instead of that being sort of what it used, what it would have been in the past, like you might have maybe listened to some calls. It would have been a little anecdotal. You have talked to salespeople. Instead, they did this massive analysis of all of our chorus calls with this core audience.

31:36They looked at Salesforce data. We looked essentially any surface where we could glean information around how the discovery calls tend to progress into deeper sales conversations to understand what kind of assets we should be applying to those conversations. What kind of technical person do we need to have in those calls or not? And they did this beautiful Claude analysis that basically outlined for us specific with a degree of precision that was pretty insane. This basically is 80 % of what happens in your first call always. Wow. And here on the margins, these are what the margins look like. and it was phenomenally insightful.

32:27And I'm very proud of my team for doing work like this. And they're just, they're getting in there and just doing this themselves. Sometimes I just ask a provocative question or like our CEO will ask a provocative question like, well, could we do X, Y, and Z? And they go find ways to interrogate it in the data using massive analysis that before would have required weeks, consulting firms, you know.

32:57Corey:Yeah. McKenzie's Cooks. Tens of thousands of dollars. Tens, right, for like something that would be irrelevant. The thing that's so wonderful about it, and this is one of the things I love about AI, is, and I'm a former consultant, so I always, you know, believe for consultants, But you get the work and the work is automatically, you know, it's sort of instantly out of date. The world is changing around it. And that's one of the things I love is we can keep running the analysis. We keep feeding it new input. So as things are shifting on the margins, we're able to understand where those shifts are taking place real time.

33:34Right.

33:34Grant:Wow. So that's really fascinating. So that has blown my mind. Yeah, that's cool. And another thing that another technology, this is more of a, well, there are a couple of things, I think, a couple of vendors that I think are doing incredible work. I've talked about Aerox before. They're doing content engineering where you have content workflows that have sort of a writer, editor, content strategist sort of ride along in the workflow to produce really incredible content on the other side and then allow you to test it with LLM. and get dashboards and understand what your share of voice and citations and extractability look like for LLMs, which is phenomenally interesting.

34:20And then what Evidenza does with synthetic personas was truly phenomenal in looking at when we did human research and then compared it to the Evidenza research, how incredibly similar they were. and not to say that human research is it has its place too but it was phenomenal how much they lined up so these things just blow my mind and a lot of it is just work we used to do in marketing that would just take so much brain power and time consultants and spreadsheets and you know things that we can now do just really elegantly um and that's it's very exciting to me and sometimes

35:00Corey:Sometimes minutes are hours instead of months, a year. Absolutely. Yeah. It's really, it's incredible. That's why I'm really excited. I don't like to think about the AI bubble. I like to think about it as a normalization because I think what's happening is incredibly exciting for me and my world of marketing.

35:22Grant:Yeah. No, totally. I mean, in every world, in every industry, there is a use case or two like that, or if not, 20 or a million. Where what you can do now is just profoundly different and improved and mind-blowing, like you said. Yeah. Like mind-blowingly changed and improved from before. Like it's cool. Yeah. Yeah, it's great.

35:44Corey:I have a quick question I want to ask. It's one we kind of skipped over, I think, unless I'm wrong again. if if i'm a founder outside of san francisco or new york you know what are my chances of still having raising a top tier round it's still you know uh we actually just published data on this on the card of website so anyone can go and look at it the regional differentials are still very real really you gotta be there the bay area is a place to be get to the bay area yeah yeah Yeah, that's kind of what I've assumed. It's still true. It's even more true. Wow. You think at some point this is like, you know, at some point you think that sort of the Bay Area is going to lose its grip in a permanent way over the startup market.

36:33It's just not happening.

36:34Grant:Yeah. If anything, it's still that is like a very firm. Wow. Yes. On the Bay Area in our data. Yeah.

36:42Corey:Continues to entrench itself like that. Really. That's fascinating. Well, AI has given it new, has even breathed more of like a, you know, of like a hardening of this trend around the Bay Area. Right. So I wish, you know, in some ways I wish I had better news for people who are like, I'd still would love to, you know, I still love working in Kansas. I want to do my startup in Kansas or whatever. But the reality is, is it's about the Bay Area. It just doesn't have the same ring to it, though, does it? No.

37:12Grant:I will say, and I wonder if you have any data around this, probably not because I feel like these folks are not raising money. But I do think that the solo founder who doesn't raise could still operate a successful business in Kansas. They just they just can't raise money out in Kansas. That's right. I was going to that. Exactly. The lot. Yeah. The latter is like that's the catch is that you're if you want to fundraise, you want to do it in the Bay Area, too. Yeah, because I feel like the same trends are still true with the exception of you can't pour gasoline levels of money on the fire. But in the same way that all of these companies who are raising right now and are scaling are able to scale and get revenue so much quicker, I feel like if you have a really good idea and you're a solo founder anywhere in the world, you can probably reach some of that same scale.

38:01Grant:I don't know if you agree with that. Yeah, and then you can move to the Bay Area when you need. Fair, fair. I love it. So I want to touch on if Carta is doing anything else with AI in your own marketing efforts that you wanted to highlight or if you're building it into the tool at all. Anything specific to Carta? Yeah. Yeah. We've been very focused on AI over the last 18 months or so. And there really are four ways, four primary ways that we're looking at how we build AI into our products. And I'm speaking mostly to the fund admin side of our business, which is where how funds operate is through a fund admin platform like Carta.

38:50And there have been there's some areas of how you operate a fund and do the accounting for a fund that is very ripe for AI. And so we've developed four sort of types of agents that we have. We have workflow agents that help with sort of the very deterministic type tasks, things like cash reconciliation, expense allocation, things that tend to have like a right way of doing them. and those tasks typically have been very human intensive because it required cross-checking things like limited partner agreements to make sure this was you know the way that an expense was being allocated was in line with what had been promised to limited partners you know essentially the way that a fund needs to be run and so we've used AI to really be able to combine all of the human knowledge of context that was through years of doing this work with people and be able to build that rich context into the agent so that it can just do those things automatically.

39:54And it's not just saving us time for the people that were doing all of that work on sort of the back end of our fund admin platform, but it also saves time for our customers. And that's where we really focus a lot of our effort. We have configuration agents that allow for CFOs to change the way that we calculate certain metrics in the platform. Like they want metrics calculated a certain way and they can actually set that calculation. It sort of stays that way until they change it. Yeah, we have agents that run in the background that essentially are always looking for, you know, I kind of call it like the janitor who works at night, like, you know, cleaning up the school and everything looks great the next day.

40:40It's sort of the same idea. These agents run in the background and find errors, things like, well, your balance sheet says this, but your income statement says this, you know, and, and we'll flag those things that sometimes are the result of, you know, human error. When you move out of a spreadsheet, which is what a lot of our customers do move out of like a spreadsheet for how they're administering their funds and move it into software, the benefit of doing that is actually accuracy. And that's the reason that so many of our customers move on to our fund admin platform and out of spreadsheets.

41:11And then we also have an insights agent that allows for customers to query our data warehouse using natural language, which is really cool. So those are the four things that are sort of in production. And really, they're just, I mean, it's, they're incredible tools for our customers. And there's lots more to come.

41:30Grant:We're continuing to build. That's awesome.

41:33Corey:That is. If you had to pick, say, one data point from your report that founders are maybe underestimating or not paying enough attention to, do you have one? Yeah. It's a good question. I mean, I think there's a few things that I would emphasize, which is the dilution. Don't feel like you have to give up more of your company. But know that investors are expecting these metrics like the billion dollars in two to three years, that some of those things are sort of hanging over your head. So speed to market and focusing on revenue is very, very important. Or focusing, I should say, on ARR is a better way of describing it.

42:26Right. Those are, you know, understanding that there's so many things at work right now in the market, some of which are to the benefit of a founder. Like you can be a solo founder, you know, don't feel like you have to give up more of your company than you feel comfortable doing, but also know the expectations are that you're going to be generating ARR at a pretty unprecedented rate.

42:51Corey:Yeah. Right. So it's a really, really interesting time to be a founder. And I'm glad there are so many insanely smart founders out there doing it.

43:04Grant:That's awesome. We have this question. You're probably not going to be able to answer it, but I want to get your take, if at all. What technically counts as ARR? Like, for instance, when a company like OpenAI or Anthropic reports ARR, are they also including the API? revenue in there or is that just subscription revenue it depends on the company of that yeah it does depend on the company how they're reporting and there's actually been like a lot of reporting on reporting for like you know how these companies are choosing to report is you know that's they're private companies and they they can they oftentimes will footnote the way that they're

43:44Corey:calculating these things yeah and release or not release whatever they choose as private companies I guess, short of what investors ask to see. That's right.

43:55Grant:But it's technically the 12 months of revenue you make per month, right? Let's just get a formal definition of ARR. Broadly speaking. Broadly speaking. Okay. There's always that asterisk on there that everybody adds. Yes. Okay. Got it. Yeah. If you're really interested in a company's financials, you've got to read how they're how they're describing the way they calculate certain things. Nicole, is there anything that you wanted to touch on that we haven't touched on? We can kind of spin it as a question. You know, I think one thing, and I get this question a lot from, like when I go do speaking, from young professionals, like what kind of job should I pursue because that's going to be AI proof?

44:39And it's a hard thing to answer. But what I tell them about marketing jobs from my point of you is that AI allows marketers the freedom to do higher order work than we've had, than we've been able to do in the past. I think that does create challenges for early like entry marketers, because a lot of that work would have been the kind of work that you would, we now have AI do for us. So, but coming in with those AI skills is what kind of gets you past, you know, that initial barrier when you're very early on that you're coming in with an AI skill set. So I emphasize to marketers, AI is not the death of marketing by any stretch of the imagination.

45:24It frees us up to do the kind of work that marketers want to do, which is creative. And by creative, I don't mean capital C creative necessarily, although that's also true. I mean the kind of work that we do at are core in marketing, whether we're an analyst, you know, or we're truly a creative, like our job is to create and marketing. And I don't think AI changes that at all. In fact, I think it empowers that work, but that means that all marketers need to have an AI skillset. And so, you know, it's something we interview for at Carta. Like I expect every candidate coming in to be able to tell me how they use AI and how they've done so meaningfully in their roles.

46:10And I think that's sort of the right way for uh for for folks that are concerned about i really wanted to go into marketing and now how am i going to do this because you can use gen ai to do all the things it's like

46:20Grant:well not really right you still need a person to prompt it right at the end of the day yeah you still need a person uh for sure and you still need like the texture and the texture and taste level yeah um of of a person who's a professional that that it's very difficult for gen ai to nail yeah yeah the slop the slop issue is a huge issue because there was like there was a great article i read recently that was like talking about how to scale without slop and you know that's a tricky one because you're we live in an infinite confident content world where we can create anything so it's like how do you scale without making it sloppy it's a tricky problem and slop the slop is brand destructive so you know as we started the way one of the things we started with talking about the importance of brand.

47:07If you're a founder, don't skip the part of creating your brand because it's still happening real time. Your brand is still happening, whether you establish what your brand is, your attributes, your vision, your brand pillars, your North Star, all of those core aspects of brand. If you don't define it, it gets defined along the way. Things like AI slop, it might seem like it's at the moment, you know, gets attention or it kind of serves a purpose, but it's actually quite brand destructive. So, you know, steering away from that and having people, I think, especially having people who are skilled in brand and storytelling, like early on in an AI startup, I think are really critical hires to make for marketing.

47:55Totally.

47:56Grant:That's interesting.

47:57Corey:Well, Nicole, thank you so much for joining us today. It's been really interesting learning about learning about all things startups pretty much yeah that's right that's what that's what we're here for at carta that's right where can people go to learn more about carta and what you guys are working on over there carta.com so there's excellent resources everything's free you know we democratize access to our private markets data so have at it i love it awesome yeah that's amazing well if you're watching and haven't yet please take just a moment out to subscribe to the channel and give us a like.

48:31Corey:We really appreciate it. Helps us keep bringing great guests like Nicole here today. Also, make sure you pop by the neuron.ai and sign up for the newsletter that helped start all of this. And on that note, that's everything for today. Farewell for now, humans.

48:59Thank you.