The New AI Growth Metric: Months to $100M in ARR

Live from Schenectady, New York, it's SAS Talk with the Metrics Brothers, Growth and CAC.

And I'm Growth, better known as Ray Reich, founder and CEO of Benchmarkit.

And I'm CAC, better known as Dave Kellogg, EIR at Balderson Capital, independent consultant, And the author of Kellogg.

And together we are the Metrics Brothers.

And we go together like native AI companies and amazing growth rates.

What? You know what's amazing? No food pairings.

No food pairings, Ray. I was leaning towards paella and sangria since I was in Spain a few weeks back, but I didn't want to make you hungry again.

I appreciate that, Cap, because I need to drop a few pounds after doing this podcast with you. So I can fit into my suit because I'm going to your daughter's wedding next month.

You are indeed, Ray. Thank you for coming. And of course, you do know you're forbidden from talking to the guests about metrics.

Well, my wife thanks you. Your guest thanks you. But speaking of metrics, let's dive into today's topic, and that's the months to $100 million in ARR.

Yes, Ray, it's another single metric episode, my favorite, where we focus just on one metric. And today we're going to discuss, as you talked about, months to $100 million as our metric for today.

Well, one thing for sure, records of months to $100 million are falling every day. But before diving in, let's hear a word from our sponsor.

The Metrics Brothers is proudly sponsored by Benchmarkit, the B2B AI and software industry's largest, broadest, and most contextualized benchmarking index. Dave Kellogg and Ray Groth Wright have put the Metrics Brothers Podcast on hiatus for the remainder of 2026. All episodes we have ever recorded are still available on your favorite podcasting app. You can reach Dave at Kellogg and Ray on LinkedIn at Ray Reich. Thank you so much for listening to the Metrics Brothers Podcast.

Well, Ray, you see these things everywhere, these hockey stick charts where the lines go like this and they basically go vertical. Um, and of course last week, I think it was, Lovable just broke the record. So 8 months to $100 million in ARR, taking the record away from Cursor, who had done it in 12 months. And, you know, since we are the Metrics Brothers, Ray, we do need to First, do the rigorous metrics thing, which is where are we measuring this from? Is it from inception of the company? Is it from the first dollar of ARR? From $1 million in ARR? What's the reference point when we talk about months to $100 million? And then, the other thing, just as historians and trivia people, is that $100 million in ARR officially has a name in the SaaS business. Do you remember what it is, Ray?

I do. That's Centaur, and that's, you know, growing to $100 million ARR. And I think one of the reasons that came about was everyone was talking about valuations, you know, the $1 billion unicorn, but you didn't know how big they were. So, and I don't know, I saw it from Bessemer. I'm not saying Bessemer invented it, but the first time I saw was Bessemer started talking about these centaurs.

Yeah, I think Centaur was a little bit of a rebellion against unicorn because everyone got so focused on valuation metrics. That they would call a company a unicorn. They say, oh, it's a billion-dollar company. It's a unicorn, when they were referring to its valuation. And I think kind of more traditional folk came along and said, shouldn't we measure a company really by how big it is in ARR? And isn't it more important to hit $100 million in ARR than to sell some small tranche of preferred stock at a valuation that resulted in a billion-plus valuation? So, yes, this has a name. We're not going to put it in the name of the metric because in my opinion, centaur never stuck. So, it's not months to centaur status. We're going to stick with months to $100 million in ARR.

But Dave, you know, you're talking about months to $100 million ARR, but traditionally, we didn't talk in months, did we?

No. That's one of the more interesting things about this conversation is this used to take years, right? Like, I'm just going to throw some numbers at you, black line. Took 14 years to get to $100 million. Cornerstone, 12 years. HR company. BlackLine was financial reconciliation. DocuSign, e-signatures, 10 years. Coupa, expenses, 9 years. Business Objects, where I was CMO, took 7 years. At the time, that was very fast. We were IPO of the year in the year we went public. And I remember the trajectory. It was $1, $5, $15, $30, $60, $85. We hit a bump in the road and then we cleared $100 million.

Wow!

For us, Box took 7, MongoDB took 7, Atlassian took 7. And I don't know, before I'd go to the slightly newer companies, Ray, any experiences you want to talk about?

Well, I do because I joined Netscape in '95 and it was through a joint venture with GE and then Netscape bought out that joint venture. But when I joined, they had just went public and they had went from launching in '94 to $81 million in '95. Then they went to almost $350 million in '96, and I was part of the environment and on campus then. And then it slowed down dramatically. '98, it hit $450 million. So growth really slowed. And then of course we got bought by AOL. So I think that a discontinuous innovation, things like going from client-server to cloud and the internet, or the internet and Netscape, and now AI, I think that correlates sometimes to these Rapid growth rates, Dave. So, that's my experience and my perspective.

So, 2 years to $100 million plus at Netscape. By the way, this will raise a point later. Like, I'd be honest, Ray, I don't think I knew Netscape's growth was that quick at the time. I knew it was quick. I knew they went public. I didn't know it was that quick. But I always worried, was it sustainable? Like, I'm like, is this sustainable? Because I never saw that part of the argument, which is why I never bought the stock. I probably left a lot of money on the table if I had the right timing. But whenever you have these growth rates, you have to ask that question.

I agree. And I think we should talk about that with some of these AI companies in a little bit.

Yeah. So now let's talk to slightly newer companies because those are all pretty old companies we talked about. HashiCorp, 4 years to $100 million. Zoom. And these are like 10-year-old companies or less, right? Slack, 3. Deal, 1.6. So Deal at the time made a huge, if you will, deal about how fast they got to $100 million, 20 months, I think, to be precise. And by the way, this will touch on the point that when we start getting to the 1-point-somethings, we should probably switch to months. And for Deal, I do think it was 20 months to get to $100 million. But the issue is with some of these companies, they're selling more than SaaS software. They're selling services, in which case you really need to look at gross margin, right? Because there's an old joke on the internet that the internet business model was to sell dollars for 90 cents. So, you know, negative 10% gross margins. You could grow really quickly when you're selling dollars for 90 cents.

Right.

The question is, when you raise the price to $1.10, your sales are going to cut to zero, right? I thought Deel had the phenomenon. In the end, I think I was wrong because even though while they're selling some HR services like payroll, I think they're doing it at software-like gross margins. That's all that should really matter.

Yeah, well, we won't talk about the other thing that Deel is famous for that happened in the last couple of months, but they had some unique insights into their competitive landscape also.

Oh, that was them. Okay. I forgot about that. This is the competitive analysis scandal or the mole, the TV show mole.

Another story for another episode.

All right. So let's now look at these similar numbers for some other more native AI companies, newer companies and native AI companies. The sources are contradictory here. Okay, guys. So in the research for this episode, I went to a lot of different blog posts. And first, everyone's measuring this in years. So I'm literally holding up like rulers on my screen to try and guess if it's 2.2 or 2.3. And the sources conflict. Some of them say where they're measuring from, like Inception versus $1 million. Some don't. So just bear with us that the numbers are not going to be precise and they will at times contradict each other. But right now, I'm using Akash Gupta's post on this. And he's got OpenAI at 6 years. Other sources put that at 2, obviously makers of ChatGPT. PhotoRoom, a Baldrick company, 2.8 years. So the fastest company I've ever worked with from $0 to $50 million.

Wow.

They do photo editing and background removal. Together.ai, 2.3 years. So, open-source AI platform, cloud platform. Deal, 2.2 years. Okay, well, above it was 20 months.

Here we go.

Different sources showing different things. The plot's the same. Details are different. Wiz, 2.1 years, code security. Cursor, 2 years, an AI coding assistant. So, look at that. And these are the newer companies. These are all rocket ships. And then Cursor comes along at, well, by other sources, 1.2 years. And then Lovable comes along at 8 months, i.e., 0.75 years. And Lovable does vibe coding of websites.

Yeah, you know, there's a couple more that I love. And, you know, it's kind of just, you know, adding on to years. The Anthropic story, you know, Claude is pretty interesting. They launched in '21. They hit $10 million mid-year '22. And in '23, they did over $100 million. So that $10 to $100 million happened between '22 and '23. The other thing, Dave, and, you know, maybe we don't talk about it now, but, you know, I love what's happened to some of these AI stories from $100 million to a billion. I mean, Anthropic went from $100 million in '23 to right at a billion annual run rate at the end of '24. And Reuters just came out and said they're going to do $3 billion this year. And OpenAI, once they launched ChatGPT, They went from $100 million late in '22 to $1.6 billion in '23. So I want to see what happens after $100 million too.

Yeah.

So maybe we'll do an episode on years that we'll be back to years, years to a billion or years from, it depends what you want to look at. You could go from $1 million to a billion, or you could actually just go from $100 million to a billion. I'll let you worry about that. I do think it could be an interesting episode, right? So, but back to the, I'm going to call it Reluctantly, early-stage growth from $0 to $100 million. And I would say, based on what we've said so far, first, you need to check multiple sources if you want real data here because it's all charts. It's all charts made by marketing people without proper axes and labels. So, you're trying to kind of estimate things off charts. Second, some people are clear about when they're measuring from. Some aren't. But it does make a difference, right? One site said they would just assume 2 years from zero. to 1 and then go beyond that where they couldn't, if they didn't know when inception was, or they didn't know when 1 was, they'd add 2 years beyond inception. So people are making little assumptions. Sometimes you can find them in the footnotes, sometimes you can't. And look, the reason the from when matters so much is some of these companies are instant rocket ships like Anthropic that Ray talked about, but some aren't. Replit is an example of a company founded in 2016, Basically zero or darn near in 2022. So, 6 years to zero, basically. There were less than 10 at the end of 3Q24. And at the end of 2Q25, there are 100 plus, right? So, in 2 quarters, 6 months, they did what one might assume is $1 to $100 million. And that is what we used to call a hockey stick. And most of these charts, to me, there's a difference between a vertical line and a hockey stick.

Yeah.

Right? The vertical line just goes straight up, right? Anthropic, for example, PhotoRoom, for example, right? But the hockey stick is more like Replit. Another example of that is Bolt, which I think nearly went out of business. If I remember the story, Bolt was in it for years, almost shut down, and they went from 4 to 40 in, I think, 6 months, right? So, the reason they're not in the rest of the episode is, to my knowledge, they haven't broken 100 yet. But from 4 to 40 in 6 months. So, we're finding this hockey stick behavior, which I think can offer a lot of hope for founders because it used to be, Ray, frankly, if you didn't get your hockey stick, it wasn't coming. I hate to be old and grumpy, but it's like, hey, we're at it for 4 years and ARR is still $3 million and we're a traditional SaaS company. It's like, hmm, I think you're in trouble. Whereas now, it's like, no, wait a minute. Maybe your time has not come yet and maybe you will hit This proverbial hockey stick, as did Replit, as did Bolt.

Yeah, I know we're going to talk about this a little bit later, and I don't want to steal any thunder, but I've been in board meetings where people compared us to that T2D3. And it's like, well, if you're not there, you're not in our top, you know, decile of portfolio companies.

Okay, let's cut to that topic then, Ray, because there are other rules of growth, right? In effect, This is a company-level growth metric. Months to $100 million is a company-level growth metric. You might call it a growth trajectory metric, right? And there are, as Ray is alluding to, other metrics for measuring that. The 3 that I think are— I use the most are T2D3, right? Triple, triple, double, double, double. Shorthand, T2D3, which basically says you should grow from 1 to 100 in 7 years if you apply the rule. The Rule of 5, 6, 7, 8, 9, which I made. It's on Kellogg, which gets you from 1 to 100 in 9 years. And then, the Rule of 85% Growth Retention with some starting assumption gets you from 1 to 100 in 9 years. But all those rules are looking pretty darn out of date when we're now measuring this thing in months and the top of the leaderboard's at 8. Right? So I think these old rules are out the window. What do you think, Ray?

I think they're way out the window. I mean, I remember I saw Bessemer posted something. This was about 12 to 15 months ago, and they were talking about their AI portfolio. In the median time to $100 million for some of their AI was like 4.8 years. And the data that we're showing right now is showing a lot quicker than what they were reporting. Just a little over a year ago.

Yeah. And to be fair, that was the median. I mean, there's 2 factors there. One, it was a median, and one, it was a year ago. But still, these numbers are a lot faster than that. And so what makes these numbers possible? Why are we able to do in months what used to take years? I think there's a number of factors here. One, AI-driven product development. A lot of these are kind of wrapped upon very powerful foundation models. So they're standing on the shoulders of giants. 2, they're able to target developers directly and to a certain extent to expand the market for developers to citizen developers. That's what vibe coding is all about, really. Vibe coding back in the 1980s was known as 4th generation languages, 4GLs, usable by anyone to build their own apps, right? That didn't really happen. And then we had the low-code/no-code movement, which kind of did happen. Companies like OutSystems and UiPath, I think, delivered low-code, no-code platforms that certainly broadened the market for who could build applications, almost moving it from like data scientists to analysts, if you will, by analogy to BI. But it still didn't unlock the regular person. And now I think this unlock, the citizen developer unlock, is huge because now anybody, like from a web development perspective, I'm anybody. And I'm building and playing around new versions of Kellogg in Replit, and I'm going to go try and do it in Lovable. So, it really is a big unlock. I think direct distribution is always good when you can sell directly to a user and just grab a subscription. Network effects. Some of these have network effects, some don't. Network effects being that the app benefits the more people use the app. AI in general benefits from network effects. Word of mouth. To the extent these all are developers within a community or security people in the case of Wiz, The word will get around. And then just teeny product-focused, rapidly iterating product teams. So, in many ways, Ray, I feel like this may be kind of the final realization of the PLG dream.

Thoughts? Well, I would add one other thing as I was just thinking about it. And that is Fortune 1000 companies. I've been talking to quite a few people in both finance and IT. They're putting a lot of experimental dollars towards a lot of different AI tests and proof of concepts. So I think they don't want to lose out. They don't want to be left behind the way they may have been in the internet in the late '90s. So I think a lot of this growth is because there's so much experimental budget going and being spent by the largest corporations.

Yeah, I agree. There's a lot of top-down pressure on companies to try AI because people are so worried about missing AI. that they are wisely putting a lot of pressure on people 2 ways. One, on cutting back on SaaS, which the classical SaaS brethren are feeling. And then 2, instead putting that money into AI. So yeah, it's not hard to get money for an AI experiment these days. And I think that's also part of what's enabling this growth. But back to PLG, Ray, do you want to talk about the PLG dream? And it does this—

Sure.

Realize it?

Yeah, let me talk a little bit because, you know, I think, you know, we've been talking about PLG and I still remember Kyle Poirier talking a lot about this 4 or 5 years ago and that this is the secret to software growth of the future. And very few companies, in my opinion, have really made it work well. I think Atlassian was a great model and our friend, our mutual friend, Carrie Lou Dietrich, was a CMO and they took it from $50 million to $500 million before they went public. So that worked really well. And then there were just a couple others that I think have worked pretty well. Canva was another good PLG that brought design to the masses. And then Figma that we just covered on an episode about what, 2 weeks ago. I think that was pretty good. And PLG was supposed to lower sales and marketing costs and put more of that customer acquisition in the cross-sell, upsell conversion to paid from freemium.

on R&D.

But actually, Dave, what I saw happening is companies hit that $30 or $40 million range, they started investing in these account management enterprise sales groups to do the conversion from bottoms-up to a top-down licensing. And I actually saw their sales and marketing investment become higher than traditional sales-led growth because they had the R&D being higher, but now they needed to do the digital marketing and the heavy lifting of enterprise sales organizations.

Yeah, that was kind of the verdict on PLG for me that you ended up with higher, both higher sales and marketing and higher R&D. The reward for that was you typically had higher growth. So PLG companies were growing faster than their non-PLG counterparts. But the kind of dream of PLG, which was, oh, the product sells itself, i.e., high R&D, low sales and marketing, didn't really happen. As you mentioned, a few people made it work like Atlassian. They really made it work.

Yeah.

And Slack. But even those companies, even Atlassian, kind of the purest pioneer, ended up with an enterprise sales team. So, to me, for most companies, PLG ended up either just a trial, really a trial program or a marketing addition. But it became a marketing funnel as like kind of a 5th horseman as they call it, right? A 5th pipeline source, much more so than a business model.

Yeah, but one thing we do, you know, we mentioned them earlier, 8 months to $100 million. That was Lovable. And by the way, Cherry Liu is also acting as their head of marketing for Lovable. So she's taken her out of that Atlassian experience to Lovable. But not only did they grow to $100 million in 8 months, they have about 2.3 million users, i.e., starting with their freemium model, and now they have 180,000 paying customers. And it's been reported that they have almost an 8% Premium to paid conversion rate, 7.8%, I think is the exact number from July. But they traditionally, PLG companies see about a 4 to 5% conversion rate from their freemium product to the paid product. So they're doing an incredible job. And one last thing I'd like to say about Lovable. A lot of people don't know this, but you mentioned Bolt. This is their 3rd product. They launched their first product called GPT Engineer in April '24. They kind of re-enhanced that in November of '24. And then this latest launch was when they made a new UI, a new kind of graphical user interface for non-developers. And that's when it started to explode.

Okay. So to start to wrap this up, because we're starting to get close on time, there's one related metric to all this, which is team size, because not only are these native AI companies growing at spectacular growth rates, right? On the order of 12 to 24 months to $100 million, unheard of by historical standards. Not only is that happening, but they're doing it with very, very small teams. Like Lovable, rumored to have 45 people roughly, right? And a firm called New Fund Capital did a blog post where their new metric is employees to $100 million, which I thought was interesting. And they're basically saying, hey, Native AI companies are getting to $100 million on 100 employees or less. PLG companies got there on 500 or less. Human-powered got there on 1,000 or less, right? Interesting metric. The other way to look at it more traditionally is ARR per FTE. And I'm just going to pick a size because the value did scale a lot in traditional SaaS. But for a $50 to $100 million company, because we're talking about the $100 million range today, a traditional SaaS company would have ARR per FTE of $150,000 to $300,000, whereas a native AI company is almost 10 times that, 8 to 10 times that, at $1 to $2 million plus per employee.

Dave, it's incredible. You know the metric I'm really looking forward to seeing from these AI companies?

No.

Enterprise value to revenue multiples. That'll be a fun one.

All right. All right. Well, speaking of that, let's start to wrap this up here, Ray. They are— the thing I find most interesting about this is, why are they raising all this money then? If you can get to $100 million on, you know, 50 people, why do you need all this capital? Because Wiz has raised $1.9 billion. Cursor has raised $1.1 billion. Together has raised $550 million. Lovable has raised a modest, by comparison, $230 million, right? I really honestly, Ray, don't know. We're not going to speculate about it here. But I'm not sure what they need all this money for.

Yeah, well, I would love to speculate, but let's do that outside of this episode.

All right, so let's start to wrap this up then because the clock is getting there. Look, this insanity isn't stopping anytime soon. I think we're going to see the records keep falling. I think one of the funniest things I saw, so we'll wrap this up with something humorous, is a guy on LinkedIn did a post where he goes, lovable quote, We vibe coded DocuSign in 10 minutes, and then some guy named Kihan Zhang has a video which is, "Hey, I vibe coded Lovable in 75 minutes," right? Which begs the question kind of of the day, which is how deep is your moat? And in some ways, how long will these companies last? As we talked about, is this going to be easy come, easy go? Hey, I got to 100 million in two years, but I'm back to zero as it was in the bubble. I had friends at a bubble company. who took it from 0 to 10,000 employees, back to 0, all throughout the bubble in bubble services. So are we going to have that phenomenon? Don't think so. But we do have to think about how sticky this revenue is. And if I worked at one of these companies, I'd be trying to find a way to make it real sticky.

Hey, Dave, do we have time for one story or should we wrap it up?

Tell the story. We can always edit it out, right?

Okay. So there's a company that created AI, launched in '21 specifically for marketing. In the first year, they hit $45 million. The next year, they hit $75 million. The next year, they hit $120 million. So, $0 to $120 million in 3 years. And last year, they did $55 million, almost a 50% reduction. And that talks about how deep is that moat and how long does it take competitors to try to cross that moat with their product.

Yeah, yeah, yeah, yeah. If you're— yeah, I hear you. Great example, Ray, and a good moral for this. If you're happy enough or lucky enough to be in one of these companies, God bless you and good job. But let's go try and build that moat. Thanks a lot, and, uh, we'll see you all next week.

Thanks, Dave. That was a fun one.

Take care.

The Metrics Brothers AI, SaaS, and More is a production of The Metrics Brothers Growth and CAC and a member of the Benchmarkit Podcast Network. By accessing this podcast, you acknowledge that The Metrics Brothers make no warranty, guarantee, or representation as to the accuracy or sufficiency of the information provided or the humor content of the jokes told. Ray? Information, opinions, and recommendations presented are, according to our spouses, probably wrong and provided for information purposes only. This podcast should not be considered professional or, for that matter, unprofessional advice. We disclaim any and all liability for any direct, indirect, indirect, misdirect, incidental, special, ordinary, consequential, inconsequential, or other damages arising out of any use of, or God help you, reliance upon the information presented here. Ray Grothreich is based in New York City and available on X at Ray Reich. Dave Kellogg is based in Silicon Valley and available on X at Kelblog. Schenectady, which is French for unspellable, is not our actual production location. Thanks for listening.