All right. Great. Good afternoon, everybody. Welcome, this is the nineteenth annual KeyBank Capital Markets Emerging Technology Summit. My name is Jackson Ader, Senior Software Analyst here at KeyBank. Thrilled to have Semrush here. We have Eugene, President, Brian, CFO. So what we'll do, we'll let you guys kind of introduce yourselves and introduce the company. We'll go through some questions, and then I'll come to the audience. I'll ping you, you know, on occasion to see if anybody in the audience has any questions. But, that's kind of the format. That's what we'll do. So if you guys just take a minute, introduce yourselves, and then we'll get to it.
You want me to start?
Yeah.
Go ahead.
Name is Eugene. I've been with the company for eight years. So it's been quite a journey. When I joined, the company was at least 20 times smaller, was private, so we had a lot of growth and did an IPO and, you know, now my role is primarily, you know, talking to investors, driving the strategy and so on, and helping this guy. Before Semrush, I did a lot of VC deals. Did everything from seed stage, late stage, pre-IPO, but I, you know, I guess that tells you I understand the story from investor point of view as well, and I understand what is a good company, what is not a good company.
We'll hold you to that for the Q&A.
All right.
Brian?
Yeah, Brian Mulroy, CFO. I joined in April of 2022, so I'm just about rounding the corner into one year. Before this, I was at Microsoft. I was part of their commercial cloud division, and then prior to that, Nuance Communications, which is a voice technology company that specialized in automotive, enterprise, and in healthcare. Just to give a quick overview of the company, for those who are newer to the name, we are a SaaS recurring revenue business. Our revenue, we just achieved about $308 million in revenue, up 21% year-over-year. Our focus area is on top-of-funnel marketing. So think of it as, you know, providing tools, technology, and data that helps companies enhance their online awareness and visibility.
Companies of all sizes are looking to make their brands and their products known to consumers in the places where they are. And of course, most of us are online watching YouTube, TV, browsing around social media, and then searching on the internet. So, we specialize on search engine optimization and marketing, social media, competitive intelligence, and local marketing as some of the core areas of focus. In terms of the financials, as I said, we just completed our fiscal 2023, so over $300 million in revenue and very strong profitability. So we just hit 83.6% on gross margin. That's up 100 basis points year-over-year. And then on the non-GAAP net income side, $16.3 million, which is up over $40 million year-over-year.
So the company's really focused in on efficiencies and are driving efficient growth and really making sure that we're delivering strong, durable growth with efficient margins as well.
Do you mind if we just start off with, you know, the most recent quarter you just reported? Release Monday night-
Yeah.
call yesterday morning.
Yeah.
What are some of the highlights from the quarter and the guide? And then we'll get into some others.
Yeah. Yeah, just had a very strong quarter. So 21% growth in the fourth quarter. That's up from 19% in Q2, 20% in Q3. Our growth strategy, we have three main growth vectors, and that really played out for us in the fourth quarter. One is just to continue to get net new adds and get our brands and other more marketers and companies using our products. So we have nearly 108,000 paying customers and over 1 million free users that are hitting our platform every single week. And we believe that there's millions of marketers, millions of small and large business owners that have a need for our portfolio.
And we believe we're in the early innings of adoption and can get into the millions in terms of paying customers on our platform. Because we have such a broad reach, so 108,000 paying customers that we get through 150 different countries, all segments of the market. So we're working with small business owners, local businesses, all the way up to Fortune 500 accounts are using our platform. And we span all verticals and industries out there in the market, so there's really no place we don't play, and because of that, we are uniquely positioned to be able to cross-sell and upsell more of our platform and our expanding platform into that space.
So you'll see in our financials, our average ARR per paying customer just keeps ticking up every single quarter as more and more of that base continues to adopt our platform. And then finally, gross margins, as I said before, 83.6%. That's industry-leading just because of the talent and efficiency we have on delivering our technology and our data to such a broad set of users. And because of that, we have a structural advantage to be able to reinvest back into the company, into R&D, which we spend a little bit above 20% on each year to develop new products, enhance our portfolio, and just extending Semrush's platform to reach broader and broader into the top-of-funnel marketing space.
So we're well positioned for growth and really excited about what's ahead of us and pleased with what we've been able to accomplish to date.
Can we talk a bit about maybe the up-market motion? Eugene, you mind walking us through... You know, I understand the overall ARR per customer is around $3,000, but you're landing and expanding into $10,000, even $50,000 and, and plus. So, what is it about the motion, maybe up-market, that might be different from the SMB?
Yeah. Like Brian said, we play in both sides of the market. So, our product was always so good that it attracted a lot of interest from, you know, some of the brightest minds in marketing who work in the world's biggest corporations. The only issue was that because we were more of an SMB player, we would usually prioritize in our R&D backlog features for small businesses versus features for big businesses, because there are more small businesses in the world, so their voice is a bit louder when you look at the whole, kind of backlog of features you can build. So at some point, you know, I would come to the conference, and people would come to me and say, "You know, I've been a user of Semrush for so many years.
I'd love to pay more, but I need a couple extra things." And we decided eventually to have dedicated R&D resources, you know, people who build features specifically for those large corporations who are already our customers and already know and like the product. And, you know, as we have kind of finished building the first version of this product, we saw quite a lot of demand and interest. So right now it's in closed beta. So we've started onboarding first customers in the end of October, so it's been available for customers only for a couple of months.
But we already have a lot of really positive feedback, where people say, "Well, I've been using this only for a couple of months, and my traffic is already, you know, 10%-15% up." So right now, our key focus is to really build additional go-to-market infrastructure, so train our support reps to onboard customers. Onboarding for enterprise account is a little bit different than onboarding for a small business. You know, we're building systems that, you know, we don't need for credit card sale, but we need for kind of enterprise-level sales, such as, you know, CPQ, teams, you know, deal desk teams that works with procurement and security questionnaires. So as we add in all this infrastructure, we're getting ready for general availability that we expect somewhere in the first half of this year.
Yeah.
But like I said, so far, very encouraging traction, very good feedback, and, you know, ARR per customer for this kind of early cohort of accounts is at least 10-15x
Yeah
... higher than our company average.
Can you give us some examples of features that a larger business would need that, you know?
Yeah
... Jack's neighbor lemonade stand doesn't?
Hundred percent. So when you build product for a small business, your approach is more of a one-size-fits-all. You know, small businesses, they don't really have time to figure out everything, so often you need to give them something that works 80% of time and, you know, doesn't require a lot of input from them.
It's good enough.
Yeah.
Yeah.
So, so that's the approach you take with the SMB product. Now, enterprise product is totally different. Those people actually have very specific idea about what they need, and they can tell you what exactly they need, and they expect that your product will be very customizable. So number one thing is to build something that is very, very customizable, where one company can have one type of dashboard, one type of interface, and another company would have a different one. So what we did is, we effectively developed almost like different components that allow CSM reps to assemble any kind of solution from those building blocks, kind of like building with Lego.
Yeah.
You know, if customer wants a house, we can build a house. If customer wants a boat, we can build a boat. You know?
Configurable, not customized-
Yeah, con-
Right?
Yeah.
Okay.
It's all no code. So you can create something that is very unique for a particular customer, and you don't have to write a single line of code. So that's number one. The second one is workflow automation and AI workflows. So a good example would be our internal linking workflow. So a lot of websites have a problem that they don't know how... what is the best optimal way to connect different pages to each other, in order to, you know, rank well and provide good user experience.
So when you have a small website with, let's say, 20 pages, if you're an expert, you can just look at website and say, "Okay, I think this page connects here, and this page connects here," and it, it takes maybe 2 hours, but you can do a good job, you know, if you know what you're doing. But then the problem gets more complicated exponentially as you add more pages.
Mm.
A lot of our customers who are, you know, big companies, they have websites with hundreds of millions of pages, or, you know, millions of pages would be enough, really. And they don't know how to create this optimal internal linking structure. So we have created AI algorithm where customer can tell us what are the high-priority pages for them, that where, where they need more traffic.
Mm-hmm.
We'll give them instruction, "Okay, if you want to get more traffic to this page, connect this page to that page, this page to that page. Use this kind of anchor text," and so on. They can then give this instructions to their product team or content teams, or sometimes they even plug it in directly to their CMS system and apply those recommendations automatically. That's something you can do only with our enterprise product, where you have all those automated workflows, and we have a lot of them. So I just give you one example. We have, you know, 5, 7 different workflows that kind of have the same, the same common denominator, where, you know, small companies can do this manually, but big companies need a lot of AI and automation to achieve the goal.
Then the last component is professional services. So a lot of those companies, they want someone to help them out, maybe help to fine-tune the system to their needs, maybe provide some kind of second opinion, maybe help them to pitch the project or pitch the campaign to their manager. And, you know, we're a software company. We don't want to dilute our margins by providing services, but we have a network of experts who have been our customers for many years. And because we have a product that measures online visibility, we know who of them did a good job and who haven't done a good job, so we know who is a real expert. And we recruit them to work with us with our clients. Almost like an Upwork model, where our brand customer may have a specific project.
They would create a task, and then we would find the best person to help them out with this task in our expert network. So those are, those are three key components of, you know, what, what makes enterprise product enterprise in our case.
The double-edged sword of moving upmarket, though, is that you run into maybe some more sophisticated and, you know, more better resourced, more powerful competitors. How have you seen maybe head-to-head competition as you move up, you know, move into, you know, some new competitors, I guess?
So we definitely know all the names in this space, 'cause, you know... In a way, we've been draining revenue from them for a while. 'Cause like I said, even though we never had enterprise version of the product-
Mm-hmm
Our SMB version was so good that a lot of, you know, best people in search engine marketing, they used it anyway. Maybe it was not 100% it fit, but they would use, you know, internal analytics teams. They would use our APIs to sort of fine-tune, you know, our SMB product to their needs. If you think about number of logos, so we have at least 5,000 accounts who are big corporations with more than 500 employees, and they are existing Semrush customer, very happy, very loyal. Those other companies that position themselves as, let's say, a, you know, enterprise SEO vendors, you know, one of them has 400 customers, another one has maybe around 1,000 customers. So we actually have more customers who are big companies than those players.
You know, we constantly see people switching from them to our SMB product, so we think as we bring upmarket product, we will see even more switching. But that's not the primary audience for us. The primary audience for us is those 5,000-plus existing Semrush customers who just want more sophisticated product, more customizable product, you know, access to automated workflows, and so on.
Yeah, so that's, that's the second pillar, right? When you talk about kind of your three things of growth, and we talked about land and, and trying to land a little bit larger customers, maybe, you know, implementing the deal desk, laying the foundation for all these things on the land side. But, Brian, can we talk about maybe net retention and these 5,000, you know, type of customers where you can begin to expand within them? When should we start to see that show up in the numbers?
Yep. Yeah, absolutely. So just kind of building on everything that Eugene's been talking about, we're really bullish about this opportunity and our position in the market, mostly because we have 5,000 enterprise accounts and the enterprise go-to-market capabilities already. So we've already engaged with procurement organizations in these companies to create those custom terms and conditions and agreements. We have the deal desk, and then we have the enterprise-class seller who's focused more on developing relationships and building a strong partnership versus facilitating a transaction. And because of that, we believe there's an ARPU inflection point for the business with this enterprise play. As you mentioned before, it's about a 10-15x uplift. So right now, our average ARR per paying customer is $3,000.
Our price point for the enterprise product will be more like $30,000-$50,000 to start, and we can ramp up from there. So really excited about the opportunity and the potential that it creates for net retention and expansion and, of course, a continuation of the really strong, long trend that we've had of consistently growing from quarter to quarter, our average ARPU.
So the final piece, the new products, we've talked about but you, Eugene mentioned Generative AI. I'll get out ahead of it this time. On the Generative AI piece, what are you developing? Why does it make sense for you to invest your R&D dollars to develop your own IP in artificial intelligence rather than maybe licensing or, you know, wrapping yourselves around something else? So what are you doing, and why?
100%. So the, the way we're thinking about AI is that, you know, it is, it is very important technology. It's a transformational technology, but we should be thinking about it the same way we thought about previous generations of new technology. So for example, I like to talk about cloud computing, right?
Mm-hmm.
So cloud computing entirely changed the way people think about scalability of their server infrastructure, and entirely changed the way people build products and ship products and what kind of margins people can have with cloud products. And with AI, it enables us to build new things. But the real question: How does AI contribute in a positive way to the solution of problems that customers have? And that's what we ask every single team at Semrush to think about. Like, you know, disregard what is technologically possible today, but where do you think AI can help the customer? And couple quick wins that we already have are, of course, our content creation products, especially text content. We had those products for a while, but of course, as large language models improve, demand for those products have increased dramatically.
What we're doing now is primarily educating the market about what is the best use of this kind of generative AI capabilities when it comes to marketing. So, we even have a lot of free products, even products that you can get without registration. For example, paraphrasing tool. And we have millions of people coming to those kind of free products and using them. Maybe they're not professional marketers today, but they're learning what can AI do for them from marketing point of view. And at the same time, some portion of them are professional marketers, and they want to create more text content, so they start using our premium content creation tools that have been growing, you know, very rapidly recently. Another good example is how we use AI in our local product to help small businesses reply to reviews.
So we have, let's say, clients, and they get review, and then usually don't have time to do anything, so review, review stays unanswered for maybe a week until they have some time on a weekend to answer. So with AI, they can, if they want, they can put it on autopilot, and it will just reply automatically. And most of the time, it's a better answer than they can create on their own because when we train the model, we use all their previous replies, replies from similar, you know, service providers, as well as best industry practices about how to handle online reputation management. So this is another feature that has very high take rate and also helps us to drive monetization for our local product because we price it as a premium functionality.
You know, we're also adding a lot of AI-powered products in our App Center through partnerships with other leading players. Actually, just yesterday, we've launched a new app from company called AdCreative. So this is a leading player in ad creation space, so they use AI as well as some data from Semrush, you know, about different ads that work well. They teach AI on best-performing ads and then help other participants, you know, their clients to create ads in, you know, in line both with best practices, but also is what is trending right now. Like I said, we keep adding more and more AI-powered features to help people, you know, do their daily tasks faster and achieve higher productivity.
But also, I think we're still at the stage where people are still learning, like, what AI can do for them.
Yeah.
So that's kind of how we approach AI. I think it's very exciting. We're already seeing a lot of good results, but there is... It will be even more exciting in the future.
I have a follow-up on, on the monetization strategy, and then I'll come to the audience in case anybody has any questions. But just to follow up on, on how are we gonna make money on AI, right? Like, how does it turn into dollars?
Great question. So I think you need to, like in any type of business, you need to provide added value, right?
Mm-hmm.
If what you do is just a wrapper around GPT, then people can just go and buy it from GPT. So the question is, like, what is added value? And in our case, we have models that are highly customized for a particular use case. For example, why our content writing tool is different from ChatGPT. ChatGPT does not necessarily know anything about your business.
Mm-hmm.
So if you ask it to create some content for you, it just creates very basic, generic text that anyone else can create, and in a competitive market, it probably doesn't perform that well. So what we do is we'll look at, okay, if you want to write about this topic, who else writes about this topic? Who gets the most traffic? Out of people who get the most traffic, what's common thing? Like, what do they talk about? What kind of sentiment they use, what kind of facts they use. Do they have same data sources that they link to? Are those, you know, are these data sources very important to be mentioned here? And then we use all this market intelligence data and create prompt. So we do, we engineer prompt for a person. It's a huge prompt.
For our average blog post, the prompt is actually bigger than the blog post because all the information you need to give to GPT to create high-performing content exceeds the amount of content that it provides in the output. 'Cause you want to provide all the facts about the industry, all the recommendations about tone of voice, about keywords they need to use, and so on, structure of the text.
Mm-hmm.
And then we feed this engineered prompt with additional data from Semrush, and the output is totally different than what people can get on their own if they just go to ChatGPT. And then when we do some tests and see, okay, if we publish just GPT output with no extra training, and then we publish content from Semrush content writing tool, our content has much higher probability to rank for a particular keyword versus just generic output from GPT. So that's just one example. With the reply to review feature, same thing. We train the model to answer specific questions.
Mm-hmm.
We don't use just, you know, the model that is available out of the box. So I think you have to provide this added value, and ideally, you need to have unique data assets like, you know, what Semrush has.
... so I, I have a final question for Brian, but I want to be fair to the audience. Anybody have any quick ones before? Okay, there's one. Oh, sure. Go ahead.
Yeah, just can you describe how the market is structured? Is it fragmented? Do customers use homegrown solutions, your competitors, market share, that type of information if you know?
Just an overview of the market structure and fragmentation, if you can.
100%. MarTech is one of the most fragmented software markets. Our space is not an exclusion. So, the big differentiator for us is that we are a platform, so we combine many, many different product solutions under one umbrella. And, if you go, for example, to review website like G2, we're leaders in 20 different product categories in online visibility management space. So, if you combine this with this fragmented market dynamic, what it means is that other people provide point solutions, and we have those solutions as well, but we also give them access to additional functionality that allows them to solve much broader scope of their problems and achieve much better results. So in general, platforms are much more efficient than MarTech versus point solution. Biggest companies in the space, like, you know, HubSpot, Salesforce, they're all platforms.
So that's a, you know, differentiation from business model point of view.
Real quick, I promise. I know it's blinking. You know, efficient frontier-
Yeah.
In terms of balancing, right, revenue and expense growth, you know, I think that's an ode to Harry Markowitz. Maybe it's not.
Yep.
But, what does that mean for investors?
Yeah.
Yeah.
Yeah, good question. So, make no mistake, Semrush is a growth company. We're here to stay. We've got a really good, extensive market opportunity, a lot of strategic priorities that are firing on all cylinders and, and driving success for us, and we believe Semrush has an opportunity over the long run to drive really strong, durable growth. In 2024, we are doing that as efficiently as possible, but reality is, you know, Semrush could easily be a 25%-30% operating margin company. We can do that today, but we've got a lot of opportunities to be able to invest in. It doesn't take much to run the existing operations of the business, given our really strong gross margin.
So as I was saying earlier, we have an unbelievably strong structural advantage to be able to reinvest back in the business, and we are doing that. And even after doing that, we're still delivering 10%-11% operating margin and then 7%-8% free cash flow margin, but that's after we're investing over 20% in R&D, continuing to build out our enterprise go-to-market function and making sure that our infrastructure is ready to support the next phase of growth for Semrush.
Great. Eugene, Brian, thank you so much.
You got it.
Thank you.
Thanks for having us.