Stack Overflow Revenue (2024)

[00:00] Quick context. This was recorded March twenty eighth and twenty ninth. So a couple weeks ago at my live event, saasopen.com. We had a thousand software CEOs there. If you missed it, we hope to see at the next one, September fifth and sixth in New York City, sasopen.com. But for now, let's jump into the recording. [00:18] >> Microsoft was customer number one. Yeah. [00:21] Over 100,000,000 in revenue so Yes. [00:23] >> That is Over a 150? Can't say that. Yes. [00:26] Over a 125, can [00:27] >> you say? How [00:34] many of you have used Stack Overflow? [00:36] >> Alright. Awesome. Of those of [00:39] you who have, who even knew that they had a SaaS product? One person, literally one person. That's good, long way to go. Meanwhile, what percentage of revenue is it now? [00:50] >> It's now about 60% of our company's revenues. [00:54] And the SaaS product so for those of you who didn't raise your hand, Stack Overflow has a collection of sites that every developer you know goes to to figure out the problems that they're wrestling with. Fair? [01:05] >> Yep. [01:06] Question and answer, other developers helping each other. What is the the SaaS product then? Yeah, so [01:11] >> I think our community, which is about 100,000,000 people that show up every day to our website, come because we have about 60,000,000 questions and answers on every possible technology topic. So many of you who have used the platform over the past fifteen years contributed to that or consumed content from that, and it's become the most trusted destination for all things technology knowledge. So that is, I think, a very, very important foundation. The SaaS business, very early [01:37] >> on in 2018, late twenty eighteen, some very large companies like Microsoft came and said, We love using Stack Overflow, but we just wanna use it inside our company because the same mechanisms really we would love to have internally for proprietary information, etcetera, so we want a private version of Stack Overflow for our company to use. So that really initiated the SaaS business, oddly in the enterprise, one of the world's largest companies today, Microsoft was customer number [02:03] >> one, and then we went from enterprise down to where we are today, which we've now got 15,000 customers leveraging Stack Overflow for Teams. [02:11] So employees of Microsoft get to ask questions of each other, and they are the only ones who get to see the questions and answers. [02:17] >> That is correct. [02:18] And how much do they pay for that? [02:19] >> They pay, I wouldn't say specifically Microsoft, but generally speaking, it has multiple tiers. So the enterprise tier is anywhere, it gets close to about $18.19 dollars or so per seat per month. So it's just like any other SaaS software that you would establish. [02:32] So before we got started, I walked around. I met as many of you as possible to get a sense of what you're here for. What I understand that you're here for is two things. One is to meet people, which, you're gonna get to do throughout. And the second one is actual tips, techniques that you can use to grow your SaaS, your company's revenue. Before we do, let's talk about what revenue is now. Is this is this [02:54] accurate, what we [02:55] >> see up on on the screen? No. It's a couple years old. So, you know, this is you know, you folks, you're good researchers and folks who actually went and pulled it from any public information, but we're growing pretty quickly, think in the context of our [03:10] Forgive me. Over 100,000,000 in revenue We so over, [03:13] >> yes, that is [03:14] Over 150? [03:15] >> Can't say that. [03:16] Over 125, can you say? [03:20] >> Wow, that's backing me into a corner. It's certainly over 100,000,000. [03:23] That's fair to say over 125 though, right? Maybe, maybe. [03:26] >> Yes, yes. [03:27] Alright, so if it's over 100,000,000, what percentage comes from this new new SaaS product? [03:33] >> So over 60% we have about probably close to now 70 of our company's revenues are recurring, and they were actually non they were not recurring revenues when I started in 2018. [03:43] Meaning advertising. We would go to to these different sites. There'd be an ad at the top. That's where the revenue was coming from. Now it's subscription. Mostly Correct. It's from the SaaS product, the internal question and answer site that you developed. [03:56] >> That's [03:57] right. Partially from the AI stuff that we'll get to later. [03:59] >> Yes, that's correct. So for the first eleven years, as Andrew mentioned, we were almost entirely an ads business and a talent job listings business. Some of you may have consumed that product in the past. And then in 2018, 2019, the whole goal was to pivot the company to focus on just on SaaS and for various reasons. [04:17] Alright. Let's talk about why. Yeah. What was the company going through that made you say, hey, we've got a successful business, we got a shift? Was a challenge. What was it? Yeah, think we've got an iconic set [04:28] >> of founders, Joel Spolsky and Jeff Atwood, who built Stack Overflow, an amazing platform. But if you think about the audience, the audience actually doesn't care about ads. I mean, nobody loves ads, period. But especially for a technical audience, it's actually extremely annoying to have ads floating around. [04:43] Plus we all have ad blocker. [04:44] >> Of course. So in that context, I don't think it was ever thought of as this kind of perpetual revenue model in the future, and it was extremely episodic depending on the economy. People obviously cut down on ad spend when the economy is not doing that well and so on. So I joined in 2019, and immediately the pandemic in early twenty twenty hit, and then we saw that very directly. We had talent job listings, people stopped hiring. [05:10] >> So why would you actually put job listings? And then why would you actually advertise on the platform when your ad spend and marketing budgets are the first things to go? So that gave us this moment of realization that maybe don't let this crisis pass, and let's actually double down on what we came what I came here to do basically, which is to convert and transform the company into a SaaS company. [05:29] And my fear was, any time you have a content based site that makes money from advertising, it's so easy for it to go to trash, to up the page views, to create ads that get more and more pushy, and the combination of all that would have ruined things. And so you said it's gotta be SaaS. How'd you figure out what the SaaS product would be? [05:48] >> Well, think it was primarily based on the user pool. I think a lot of the customers like the Microsofts of the world, when they came and said, hey, this is what we care about, the early version of the product was literally a copy of the public Stack Overflow, but obviously completely a private version of it and just sort of tested in the context of these big companies. And then we said, okay, now actually we need to [06:08] >> go build a proper SaaS product and so let's actually go do all the things that is necessary to accomplish that, which all of you I'm sure know of. So that was the impetus. It was straight from our users, straight from companies saying we love the platform, thousands of people within companies use Stack Overflow, all your companies, if you came from big companies use Stack Overflow every day. So it sort of already present in people's workflow. So [06:33] >> the idea was, why don't we actually exist in the context of the company's workflow? [06:37] Yeah. Alright. So we're gonna talk about how you got your first customers and then we'll grow on from there. But what I'm showing up on the screen is what you're charging is in the middle graph over there. That's the price per seat. Yeah. And you still do have a very healthy advertising business. Yep. I'm podcaster. Been one for years. I actually wrote a book on interviewing with Nathan. Nathan kept up here. But you have podcast ads. [06:57] You have banner ads. You have all kinds of ads throughout the network. Yeah. Okay. The very first customer you said was Microsoft. Where did you figure out? How did you get the first customer? [07:07] >> Well, I think the the the notion of them actually coming to us, so I think it was completely a user pull, a customer pull. You can't underestimate the power [07:16] of Wait, they weren't coming to you and saying, how do we spend money with you? What were they saying? [07:20] >> The users basically said, look, we love to do this, but the thing was that when people were posting content on public Stack Overflow, that started getting people's attention on, you know what, we actually wanna make sure that this content is not posted publicly with [07:31] You mean on a regular basis, they would say, we wanna ask questions, but we can't have people outside of our organization give us the answer. Okay. [07:38] >> Or or post it externally by accident as an example. Because, you if you're posting a piece of code from within Microsoft on public stack overflow because you want feedback, etcetera, that would be, I think, obviously a problem. [07:48] Here here's my understanding from the conversation we had before we got started. You looked at the people who are using the free version of the site. Mhmm. You said, what organizations are we seeing come up a lot? Mhmm. Let's go talk to them. Mhmm. Right? Mhmm. Put it in a spreadsheet or database? [08:01] >> In the context of, like, post customer one and two, I think at that point we started seeing, okay, let's actually think about every possible customer that is using Stack Overflow, so we actually had that very specific list of the largest users company wise. And we typically went into companies and we said, you've got 2,000 users using Stack Overflow. And by the way, here are all the tags that they are using. They're using Python. They're looking up [08:25] >> content on Python. They're looking up content on PHP. And by the way, here's why you should think about having an enterprise license with us. It was very much a bottoms up, tops down, similar to a PLG motion, but this is more of a community led motion, if you will, or CLG sort of motion that we went after. [08:41] Okay, a lot of microsoft.com email addresses registering. You go to the person at Microsoft dot com who's the right person, you say this is how many people are already on your network, wouldn't you want to keep it private? Great, you close some sales. You told me earlier about SDRs. SDRs, sales development reps, what do they do? [08:56] >> Yeah, so obviously most SaaS companies have a great outbounding machine if you're actually good, and that I think is a foundational way in which you generate pipeline, so the motion of actually going and figuring out, hey, which company do you target and what do they actually use in terms of Stack Overflow, that whole work and actually putting out the outbound email or the call, that itself is the first initial activity. So I think it's just this [09:21] >> has been well documented over fifteen years of SaaS companies existing, you just need a machine to grow, and as long as they're productive in terms of generating MQLs and SQLs and SQOs, etcetera, in the funnel, which ultimately will translate and convert at some level to pipeline, then you start building the predictability motion, which is the obviously, one of the foundational characteristics of of SaaS businesses. [09:42] Meaning, somebody registers with an email address, your system checks to see what company they they're with Yeah. How big is the company. And if they're big enough email goes out saying, can we have a conversation about how we can help you more? [09:54] >> That's right, that's in the inbound SDR process. The outbound SDR process is we're saying, okay, what are big companies over 5,000 users that we can go after, and then in the context of getting them to pitch the product. And our SDRs go make those outbound calls and go do them systematically over a And period of then we measure that and we say how much that converts into pipeline. [10:16] Now this is something that's being done by many companies. Anyone out here doing this right now? Have an SDR checking to see the email addresses of the people who are registering for your site? Are. I'm seeing a couple of people here. For people who aren't doing it Yeah. What's a good way to get started? And then we'll talk about for people who are doing it how to get better. [10:34] >> Start with one or start with two. That's what I would say. As long as they are able to be productive on their own, there's no [10:41] difference Get one SDR. One, get one SDR. Look to see the email addresses of the people who are coming in Yeah. Use Clearbit or software to check to see who they are, and then start sending out offers to [10:51] >> get on calls. Correct. Basically, get the first meeting, and I think that you need you can do that on your own, but it's probably not the best use of your time, so in the context if you're a founder. But you can probably get started by doing it yourself just to see what is the attraction if you're an early stage founder. But then that really gets you to understand, okay, what is catching the attention of this prospective [11:11] >> company or this customer? Are you actually solving the right problem for them? Is it high enough on their list or do they care? And that gives you a lot of intelligence in terms of messaging, whether your product is relevant, is it must have, is it nice to have, how much should you pay for it, etcetera, etcetera. [11:28] Okay, and then for the people who are doing this, what have you discovered about improving this process? [11:33] >> Yeah, I think the landscape evolves pretty quickly, so I think to be able to add capacity only and when you know that these SDRs are productive, that the intersection between SDRs and your account executives, I think that's a very important intersection, because if they actually work collaboratively and they align together on which accounts to go after, that cohesive relationship is super important. So especially as you have a bunch of AEs along with those SDRs. So that [12:03] >> I think is probably the most important relationship, the SDR and the AE relationship, along with the demand generation organization who are obviously driving campaigns and you're doing all sorts of things around social media and LinkedIn and Google and all the things. But all that needs to come together between marketing AEs and SDRs. If it doesn't come together and it's somewhat of a disparate set of activities, you're not getting the biggest bang for your buck. You're probably [12:26] >> actually being quite inefficient with the buck. [12:29] Okay, all One of the things that I did before we got started was I talked to people who are here in the audience. Sandy Robinson, who's a speaker who will speak later at the conference, said, Talk about compensation. So when it comes to sales compensation, when it comes to SDR compensation, when it comes to even the CRO that you hire, how do you think about compensation? [12:47] >> Yeah, think it's mostly, it's pay for performance. I think you wanna make sure that people have a base level of compensation on the base that people feel like they can pay the bills at a minimum level, of course, and that's important, but most of the, it's carrots more than sticks, I mean, it's really making sure that you have tremendous amount of accelerators, you have the opportunity to over perform, and you can make a lot of money [13:07] >> as an STR or an AE. But I think with salespeople, STRs, they're successful, they have a machine, they think about diversifying their pipeline with big companies and big prospects and middle scale, and then smaller companies, as we talk about the pebbles, rocks and boulders combination, then you've got the discipline and the predictability for them to start saying, okay, you know what, I'm gonna plan ahead, three months ahead, so that I hit my base number, and then [13:33] >> anything on top of that, if I hit a whale or I hit one of these bigger companies, I'm gonna overachieve and my accelerators are gonna kick in 1.5 times or two times. And then the most successful reps and the most successful STRs will plan ahead, two quarters, three quarters ahead. And they have that machine going, so there's a massive pipeline just building, and they're coming into the week with plenty of things to go do, and they're [13:54] >> not starting from scratch. So I think you need a little bit of maybe a couple of months, three or four months of them to ramp, obviously, when they get started to build the machine and to get going so the flywheel is running. But once you do that, then I think you're on your way. [14:06] Are these your customers that we're looking at here? Siemens, Bloomberg, Microsoft That is correct. Again, based on 2022 and some of the numbers. And by [14:15] >> the way, it's not spying, it's paying. So but yes. That's correct. [14:19] But we're looking at $250,000 for an enterprise customer. How long does it take to close a deal with [14:27] >> an enterprise customer that's paying that much? Yeah, probably between nine to twelve months for the largest strategics, like the Microsofts. The enterprise is probably closer to six months, and an SMB customer is probably anywhere between one and three months. But you're going [14:43] to the top and getting the deal with the company, you're not having them start to use it and then going into the company. [14:48] >> We've done sort of both over time, but as I've talked about already, 100,000,000 people use Stack Overflow, invariably everybody within companies use Stack Overflow because they've used it as part of their work historically. So that is in many ways our bottoms up PLG motion, or if you think community led motion, And so we leverage that as our bottoms up motion, which is very similar to a lot of SaaS companies who provide people free trials and freemium [15:13] >> offers, etcetera. This is our free offer, our public community. People have been using it, same format exists, etcetera, in the private product. Yeah. [15:21] Okay. Alright. So now the next thing that you told me that you did was you said, look, we've started going after our existing users and seeing if some of them would get their companies to convert into the Teams product. Then you went into lead generation, you started building marketing assets, campaigns, messaging, what worked for you? [15:37] >> Yeah, think having a very specific ideal customer persona, as many people here know, ICP, very important, making sure that the value proposition or what you're actually talking to them about, actually solving an important problem for them versus just shooting and spraying and praying and hoping that something sticks. But the more targeted you are to the people that actually matter with the problems that you actually are gonna be solving for them in a very crystallized fashion by [16:04] >> leveraging your strengths. That whole process of Wait. This sounds very corporate y. Yeah. [16:09] Yeah. What are we talking about? Like, when you're saying that marketing has actually worked to pull people in, what's worked? Where do I see it? A lot of us here didn't even know that you had a SaaS product. [16:18] >> Yeah, yeah, that's fair. Branding, so clearly we've got to work on that one. But I would say the biggest thing that has worked has been the bottoms up motion along with engaging with the CTOs and the CIOs and VPs of engineering that we typically engage with. And the specific problems that they are trying to solve, whether that is transforming into the cloud, leveraging AI tools these days, or whether that is becoming more efficient with their developers [16:46] >> in terms of productivity, or if it's onboarding people during a time when they're hiring a lot of people very easily, how do you get people you've got to zone in on the top three problems that they actually care about. And that messaging goes out very specifically to the personas that we care about. Right? Okay. Specifically, the CTOs and CI. [17:02] I'm gonna spend real quick just on partnerships, and then I'm gonna get to AI because we've done some big things there. Yeah. Partnerships you've got partnerships with Microsoft and others. What are they doing? How are they selling Stack Overflow teams? [17:13] >> Yeah, so I think as part of the scale journey, we obviously started with our own user base, we went into these big companies, we built up an SDR team, we had a demand generation function that got built up, then we had a more mature AE organization, and then we surrounded, we hired a lot of people who actually had SaaS backgrounds to go and kick off this entire transformation. And then we said, okay, at some point it [17:36] >> makes sense to then engage with big companies as we think about big brother partnerships. We are a smaller company, we have somewhat of an outsized impact in terms of our user base, but we wanted to partner with one of big ecosystem companies. So Microsoft is an example, one of the biggest companies in the space. We are very big partners of theirs. We actually engage with them their Azure partnership area, if you call it the ISP program. [18:00] >> I see. And we align with their reps to be able to go and sell into accounts jointly, so that's another way for us to get the attention of a CTO or CIO of a very large company. And so that is a [18:12] They bring your salespeople in when they're making Mutually. When they're closing. Got it. [18:15] >> Yeah. It has to be done mutually. [18:17] Yeah. Now let's talk about AI. Dude, when I was looking you up last night, the big criticism that I kept seeing was Yeah. People aren't gonna use Stack Overflow anymore because developers are already using AI as part of the software they're using, and you and I talked about what you did with Google Gemini. Can you talk about that? And then we'll talk about beyond. Yeah, so [18:33] >> I think our vision in an AI world is all about bringing together the power of AI and humans on the public platform so you have this really high trust product and set of answers. Because when we went and did a bunch of surveys in our user base, 100,000 people responded and said 70% of them wanted to use AI tools for software development or for something related to technology development, but only 40% of them trusted what was [18:58] >> coming out of AI models because hallucinations and all the things that you know about, which absolutely the case. And so we say, our role should be bringing humans and AI together on our platform on Stack Overflow. What does that mean? It means humans do amazing things, creating new knowledge, validation of information. The whole community asks, we've been in the upvoting, downvoting game for fifteen years of reinforcement learning, if you will. And then you've got AI that [19:24] >> does all the things we know about, summarization, semantic search, and all the things that you've read about. So bringing those two things together delivers highly trustworthy, highly accurate, highly attributed and cited information that are, you know again, they're leveraging Gen AI, and you're probably getting a Gen [19:41] AI on the platform. And at the same time, you made a deal with Google's Gemini where they can take data from Stack Overflow and use it to inform their AI and get links back Yeah. As a way of having people see the sources so that there's no concern about about Right. Bad information being on there. Have you made a deal with OpenAI yet? Can't comment on [20:02] >> it yet, but we are [20:04] But it's a possibility. [20:05] >> Absolutely. All the big AI companies, we went and basically got a bunch of inbounds from them when we announced our overflow API program, which is basically to say, how do we make sure they're focused on responsible AI, so they're not just sort of scraping the web off of our web, which is what's been happening for the past several years? [20:21] Meaning And they've been scraping your data without your knowledge and consent? [20:25] >> Historically, yes. But now when we saw that, we said, look, there's responsible way of engaging with us, and so we launched this program called Overflow API, and the first company that basically engaged with us is Google. And so now all Stack Overflow content, when used within Gemini's AI capability, will show up in their AI tools with links to Stack Overflow back to the users who get attributed, the work that all of you, if you contributed to [20:48] >> Stack, your names will show up when a Gemini answer is And by the way, reputation accrues back to the user. If you're the best JavaScript programmer, your answer was used by Gemini thousands of times, and you get the credit for it. What's the deal with? Did you make a deal with Anthropic? We have many announcements coming, and you can expect some here in the next very short order, actually, but more than Google, So alright. Yes. [21:12] Thank you very much. [21:13] >> Thank you, Andrew. Yeah. Give him [21:14] a big round. Yes. [21:20] Hey, folks. If we haven't met yet, my name is Nathan Latka. I launched and sold my first software company back in 2015 and went on to write a book about it, which you guys made a Wall Street Journal bestseller purchasing over 30,000 copies. Thank you so much for that. After the book, I launched this show and one went on to create founderpath.com. I raised a large fund to do non dilutive deals with B2B software founders. So [21:47] far, we've invested in over 400 software founders totaling $150,000,000 Here in 2024, we're doing three to four new deals per week. So if you're looking for capital and don't wanna give up equity, go sign up at founderpath.com for free to get your offer.