Welcome, everybody. Super excited to have you here. Anu from Atlassian has been so kind to join us. She has been helping lead Atlassian for several years. Before we get into Atlassian, just a really quick point. I've got an iPad here. This iPad gives me access to questions that you might have. I think there are some instructions, you know, on your card. You probably have done this over the last day or so. If you have questions, please feel free to enter them in Pigeonhole. I will do my best to bring them into the conversation here. Anu, I really appreciate you joining us. You know, maybe just introduce yourself 'cause you've actually had a really interesting—we were talking about our, some of our joint history.
You've been with Atlassian a long time, so you've seen a lot of things, and this is a company that's changed a lot over the years while being the same. Maybe give a little bit of that context about your experience that you're gonna draw on as we have this conversation.
Awesome. Yeah. Thank you so much for having me here. I am Anu. I'm the President of Atlassian. I have been with Atlassian for about 10 years, just over 10 years now. Before that, I was at Microsoft for 10 years building software for developer teams where you and I had an overlap. Atlassian is a company that originated in Australia. I joined as the head of product for Jira. I have a product engineering pedigree. The thing that really drew me to Atlassian, which was very small, maybe 300 people compared to Microsoft where I came from, a large company, was that this was a company that truly believed in a mission and was trying to solve problems in a very unique and interesting way.
Being in Australia, the company had a unique original point of view on how teamwork should work, how teams should be built. As someone who was a developer and had built products for developer companies and developer tools, I was really fascinated by the unique approach that Atlassian took. In Atlassian, I started out as a product manager, took various different roles over the years as the COO of the company. I drove our shift to distributed work. One of the interesting things about Atlassian is that we are over 13,000 people and fully distributed, in the sense that we do not mandate anybody to come into an office a number of days a week. In fact, well over 40% of the company lives about two hours away from any physical office.
We have had to live the future of work in order to invent the products that support the future of work. Most recently, as the President of the company, I am responsible for the overall top-line revenue, and I run our product design, strategy, BizOps, M&A teams. It has been fun.
Anu, it's, there's so much in there that we're gonna get into with the company, but it's about the company almost dogfooding itself. It has been at the forefront of building what it's building in a product and helping other people get there. Maybe actually help folks understand what Atlassian does, right? This is not a household name. This isn't even something like Salesforce where you, you may have, like, had some, you know, like, engagement with it. Maybe even start, like, literally from Bugzilla. Like, why did this thing get started to begin with, and how it morphed into this overall, like, project, you know, management toolset?
Sure. Atlassian's a 23-year-old company. We serve about 300,000 customers, which is an unusually broad swath for companies that serve B2B, that build B2B software. About 85% of Fortune 500 companies use our products, and we are very ubiquitous across teams of every kind. Atlassian's mission is to unleash the potential of every team. In fact, funnily, when I joined Atlassian 10 years ago, I joined thinking, "Oh, this is about Jira, which is a project management tool, bug tracking tool, right?
Mm-hmm.
Your Bugzilla reference. In the first week, I was shocked when I looked at product analytics that half—well over half of our customers, in fact—had nothing to do with development teams or bug tracking or any of that. Jira was being used for tasks as varied as hiring people, tracking recruitment loops, tracking financial workflows, tracking how customer support was getting handled, tickets with internal help desks. A lot of collaboration workflows. Fundamentally, we are a collaboration company, and it started out with developer teams. The original set of teams that used to adopt our products would be development teams, people who built software, so engineers and engineering leads. It spread to product managers, designers, and surrounding teams. Even now, across—we have a portfolio of about 14 different products and a cloud platform that supports all of these products.
Over half of our usage still comes from different kinds of teams, but not developer teams necessarily.
Right. Would a fair way to kind of, like, imagine what this means on a practical level for, for folks here? You know, I often describe, you know, there's, like, two levels of software. There's software used by, like, individuals, like, on their desktop as productivity tools, and then there's tools used by managers to help them, essentially plan out the work that needs to get done, allocate that work, track that it—that it's getting done, measure their productivity, create, you know, essentially ROI, you know. If you think of a Salesforce, it's that latter, right? Whereas if you think about, like, say, Office or Visual Studio in the case of, you know, a developer tool, it's that former. It's on the desktop for the individual.
I think of Atlassian being like that Salesforce for, you know, any product team or any team that's trying to create something that's not—I mean, a sales team obviously has some tickets that they've gotta work through, but it's—that's their product, right? For all the other teams, Atlassian is a fantastic tool, to get it done. Is that a fair way of kind of thinking about Jira, you know, like, as its—as its central point? Obviously there's some adjacencies, you know, with Confluence and everything like this.
Sure. I would reframe it slightly as, you can think about what we build at Atlassian as different sets of products that serve different kinds of teams. So both the end users that you describe as well as the managers who need to report on them. Fundamentally, we are a collaboration company. We think about how do you create a system of work across the entire organization. The previous generation of tools used to be systems of record, right? You capture work and, you record it, and there is one source of customer info or one source of business information. Then we morphed into systems of intelligence where you could get insights on that data. Now we've come to an era where we've defined a system of work where work happens across all teams. Work looks like bug tracking for product managers.
It looks like sales deals and marketing campaigns for marketing and sales teams. The fundamental act of collaboration, of Peter and Anu work together in order to accom—accomplish a given outcome or goal, that thread remains common. That thread is what we address with Atlassian. The Atlassian system of work has tools for different sets of products, for different sets of teams. However, we connect all that with a central cloud-based platform which captures all that information. 'Cause fundamentally, our belief is that for any organization to work well, everybody needs to have access to the right kind of data and the right kind of context. We have customer support teams, product teams, developer teams, finance teams, exec teams. Everybody works off of a single platform and be able to unleash the power of, common knowledge.
You know, it's interesting if you look at the evolution of kind of your market share. If you—there are some product surveys out there that go on. You know, they would suggest that Jira's up to maybe 2/3 market share of teams working on it, and probably the next closest tool that people use is Microsoft Excel, you know, like, in the teens. In other words, like, a non-tool. It's like, you know, jamming something that doesn't really belong in that space. What is—what's allowed Atlassian to just kind of become the de facto standard that everybody uses to do this stuff?
It's not like over the years there haven't been other companies that have, like, tried to come at bits and pieces of it, but somehow Atlassian continues to just win share and nothing else seems to, you know, gain relative to you.
Yeah. That's a good observation. I live in San Francisco, and I feel like every time I drive down the 101 Highway, there's a new billboard of the new Jira killer and I've seen this over and over.
Yeah. It's like, "New Jira," you know, or whatever it is now.
The thing, uh, I think the thing that gives us staying power and endurance and how we've been able to scale to a $5 billion run rate is that we serve a wide swath of customers. If you think about our $5 billion run rate, our SAM is actually $67 billion. Even though we have 300,000 customers, we literally build products for every company. The number of customers or companies we can serve runs into the millions. We are just getting started, if you look at the possible addressable business for us. Fundamentally, Atlassian has always been a long-term business that is very patient with monetization and focused on bottoms-up selling. What makes us special is the land and expand model where we land our products self-serve.
We've traditionally, historically not had an enterprise sales team that goes and knocks on a company's door and says, "This is Atlassian. Would you like to buy us?" We've never done that. Instead, our products have been so good because our investment in R&D is also pretty atypical for a company. We invest more in R&D compared to sales and marketing that the product attracts users, in a self-serve organic way. Then it expands throughout the organization. Like I said, we serve different kinds of teams. Typically, in an organization, our products get used by a wide variety of different teams. It's not just Jira. It's Confluence. It's Trello. It's Bitbucket. It's Jira Service Management. HR teams use it. Finance teams use it. IT teams use it.
When we do have to go upmarket to an enterprise sale, classically, what we have to do is we have to land at the customer's place and say, "You know what? So many hundreds of your teams are using our products already. We're now going to consolidate that into one single enterprise license." It's almost an inversion of the model. What enables us to do that is our long-term deep investment in R&D, which has helped us build a technical moat with the cloud platform, which is very difficult for other competitors to catch up with. That has created compounding gains for us. We're able to serve the self-serve small, medium businesses, but also the enterprise businesses that need things like FedRAMP, moderate, that need things like isolated cloud, more data management and regulation.
We've constantly invested in that and made sure that we can address the full gap.
Maybe we, we push on that R&D point 'cause I think you've emphasized it several times about why that's so powerful. I think that there are two pieces of this that you've alluded to. One is kind of that product-led motion. We should talk about what that is, what you called the flywheel. I think you're starting to talk about product-led, although I think of them slightly different for Atlassian than, say, like, you know, a Datadog or some of these other businesses 'cause I think the setup's slightly different. The other side is the R&D spend is part of what makes it hard for anybody else to come in, right? Let's start with the first one because I'm gonna ask you a pointed question on the latter.
Help describe that flywheel and how sales and marketing actually is kind of a product function in this type of mentality. Your sales and marketing as a portion of revenue has been like 10%-12%, right? Very low. I have many other software companies that are 20%-30%, right? Your R&D is much higher in the 20%-30% relative. Some of it is about taking that budget and putting it there. How does that work on a practical manner in your product team to drive essentially that commercial engine?
Yeah. It's a great one. Like I said, we have, at Atlassian, a cloud platform which supports multiple products. The way to think about the product-led motion is typically our customers start with one of the 14 different products. It could be an engineering team using Jira for bug tracking. It could be a customer support team using Jira Service Management for customer service. It could be an HR team using Confluence for performance management and recruiting. What happens is because we have so many different products and different landing points, any given team in an organization could start using our products. Typically what happens is more than one team starts using more than one of our products because once you start using our products, we've engineered it in such a way that it's really easy to add on adjacencies.
For example, we have 60% overlap between Jira and Confluence, and we have about 60,000 customers using Jira Service Management. These are also customers of Confluence, for example, many of them. What happens when a team starts using more than one of our products is that they start storing workflows and data that is common across many different workflows. The more they use our products, the more powerful it becomes. The more useful it becomes, the more productivity it unlocks for teams. There is a natural attraction and a natural gravity to keep adding more than one product. We think of that as cross-sell. Typically, teams on cloud, using one of our products, tend to add the second, third product way faster than, for example, behind the firewall or on-premise software. On top of that, we have different editions.
Because we serve 300,000 customers, we've built it in such a way that there is a freemium tier and a very accessible, accessibly priced standard tier. We go all the way up to premium and enterprise, which tends to have very sophisticated data controls, compliance, security, performance management, SLAs, which tend to be priced higher and higher. In many ways, we provide organizations not just with a kind of ladder to go up, but think of it almost as a net. You can go laterally, add more products, which will be cross-sell, and you can go upwards, which is upsell in terms of adopting more and more additions. Because we have that entire mesh across the board, we really track how do customers respond to adding more products? What can we do to help them discover more of our products, help them upgrade to future additions?
Like I said, we attach the enterprise sales engine on top of that. Like you described, sales and marketing is very much a function of how do power product users, how do we take them further up the chain?
Yeah. Yeah. Maybe we go to the other side, which is, you know, one thing that I think is very powerful about Atlassian, but it costs a lot from an R&D standpoint. It's a little bit like, why does Amazon as a retailer win? It's the long tail.
Yeah.
What you have embraced is the churn of all these little tools that people in product teams and any function want to use and basically said, "I am going to integrate those into our product and make it feel as if it was just a natural part of that." Whereas if I think of other competitors that have tried to come into your space, it is like that vanilla, like, "Yeah. You can do this very simply." It is like, "I need it to be customized for the stuff I am doing and all the little tools." Maybe talk about, number one, why that drives cost, but obviously why that is so hard for other people to replicate, you know, like, say, if Monday wants to come horizontally into this or Asana or whatever, they just remain so vanilla relative to what Atlassian provides.
Yeah. The R&D investment has basically allowed us to create a common cloud platform, which is extremely hard to imitate because the kind of ingredients you need to create a platform is, one, you need usage across multiple kinds of workflows, which over 20 years we have established usage across different kinds of teams. Therefore, the platform is able to serve different use cases, different personas pretty effectively. If we want to enter a new market, for example, we just announced a Jira Service Management app called Customer Service Management. With Customer Service Management, we are now starting at a point thanks to the platform where we have all of the primary workflows covered, and then we add customization and AI on top of it. We already have 25% of JSM customers using our CSM product.
We start very much from a position of strength because of the platform already containing existing usage patterns, understanding existing customers, as well as existing data from organizations. It's extremely hard to go mimic that because it takes years of compounding investments to make that work. Typically what we see with competitors is, for example, Breville is a customer recently that adopted the Atlassian system of work. When we say system of work, it's really the entire portfolio of products. There were 30+ teams using our products that consolidated on the Atlassian platform, and they displaced about 10+ competitive tools. You can see how it's kind of a surround and conquer strategy where different teams adopt different parts of our product, but the common platform really connects all these teams together.
We are able to really bring together teams across the entire company like no other competitor can.
I'm gonna put a pin in something so we can come back to it 'cause yesterday I hosted ServiceNow, and I think the two of you, and part of the reason I've liked both of these businesses, I think have similar ways of how this is evolving in kind of like some parallel market spaces where I don't think you have to take each other out, but you're getting some of that same long-term value. We should come back to that. I do want you to pull a thread of something that's been very important for your business that you've alluded to, but we should talk about directly, which is the cloud versus server and data center. This has been a big evolution for folks in the audience who may not have been following Atlassian for a longer period of time.
They have historically had, kind of a self-hosted on-premise, what they would have called the server product back in the day. They more recently have built a cloud version.
Yeah.
You've been alluding to some of the values of cloud. First, maybe help the audience understand where are you in that migration of trying to get pretty much every customer to cloud and why is it so powerful to get every customer to cloud, even though there was a risk and pain of churn of that customer base when you were kind of like extracting them from kind of that self-hosted version.
Sure. Atlassian historically had a server business. Think of this as a perpetual license model. We would basically have customers buy the server software and really only pay as a maintenance fee thereon. Several years ago, we decided to switch to a cloud model for two reasons. One, technologically, cloud would allow us to offer innovation that server technology would not. Second, we wanted to shift to a subscription model from the perpetual license model. We began this journey in 2020, and I would describe it as the final leg of that journey. We have been phenomenally successful with that transition. We have done better than what we had internally modeled in terms of customer churn. We have done better in terms of what we expect expansion to look like.
The fundamental thing to understand about the business is that the shift from perpetual to subscription business is complete for us in terms of we no longer have a perpetual license model. We end of life the server product, and data center and cloud both have a subscription model. Customers who adopt our cloud products, like I said before, show higher degrees of expansion, that is, adding more seats as well as higher degrees of cross-sell, adding more products. We have a lot more cloud products than we do server products naturally because server was the legacy business. Over the years, we've managed to successfully migrate a lot of our server customers.
We went from the business being entirely server in DC to well over 90% of our customers now are on cloud, and 90% of our customers now choose cloud. The set of customers I would describe our journey as the leg that is about to come is that we have a set of enterprise customers that are some of our largest and most complex customers that have regulatory rules that have constraints that need them to run hybrid environments. Those are the customers that we're focused on over the next couple of years to bring them over to the cloud platform. I would describe the migration overall as having been extremely successful and better than what we expected.
Yeah. And to enumerate that, I think early on you were pretty open with investors basically saying, "Hey, there's a good chance we're gonna churn some of these customers." How has that actually ended up?
We've actually done better on churn.
Yeah.
Than we had initially modeled.
Very little, right?
It speaks to the staying power of the products and just the competitive differentiation because the time to value is really low. Customers can immediately see value. Second, we are priced very fairly and with a night to value. Churn happens to be very, very little for us.
I can, I saw this firsthand because at Alliance Bernstein, we were a server customer. When this change occurred, we were like, "We should look at getting off." You know, we're not sure that we wanna do that. When we looked at all the alternatives, not only were they worse, they were more expensive. It actually reinforced, I think, this process, while frustrating to many customers, helped them like almost re, it's like reasserting their vows in some way. It's like, "Wow, okay. I now remember why I'm sticking with Atlassian." You know, maybe we take this in two different places. We need to talk about AI.
Mm-hmm.
There are two things we need to talk about with AI: the threat and the opportunity. Okay? Let's start with the opportunity. Okay? You have started shipping over the last couple of years Atlassian Intelligence, Rovo. Help the audience understand what role AI is playing from a feature functionality standpoint, for the business right now.
Yeah. AI for us is very much built into the platform. The way we think about AI is it's very additive to the business. As a technologist, I'm very excited and optimistic about what AI can bring for us, and as a businesswoman, I'm very optimistic about what this means for Atlassian's business. Technologically, what we've done is we've infused artificial intelligence into the cloud platform so that every one of our products, every single one of our products, bring AI into your daily workflow. We are not asking people to go buy a new product, learn a new thing in order to use AI. It very much plays into the land and expand model where we are very focused on usage and adoption first and being patient with monetization.
We've included Rovo with every single edition of our products, and we've seen phenomenal adoption of it. In fact, just within a quarter, we went from 1 million users of AI on our products to 1.5 million . We've managed to increase usage and adoption by 50% just over the last quarter. Business-wise, why this is exciting for Atlassian is that AI as a technology trend basically means that more software gets created. Earlier, you needed to be a highly technical person writing code in order to create software, whereas now that same power of creation is available to everybody. You don't have to have written a line of code in the past, but you're now able to build applications, build software, despite what your background is.
That is a great thing for Atlassian because what that means is as more software gets created, more people get involved in this entire creation process. Fundamentally, we are in the business of teamwork. Whether it's humans collaborating with other humans, humans collaborating with AI, all of those workflows need management, and that's where we come in. In fact, there are a lot of our customers who have openly talked about how AI is going to impact their teams and their workforce, and we continue to see their licenses expand. Why? Because our tools serve every single profile that's involved in creation and not just the people writing code. In fact, the code generation tools, which I think perhaps command more share of the news than is want, is, 20% of a developer's time on average is consumed by coding.
The remaining 80% is, how do I work with the designer? How do I work with the customer support person to resolve bugs? What about the remaining 80%? That is what we are focused on. How do we make that efficient? I think there is tremendous opportunity and upside there for Atlassian's business overall to capture those workflows.
I think part of what you were getting at there is also gonna be that, like, demand side, the number of seats and humans. Let's pause on that one for a second. If I had to make this, like, tangible for folks in the audience, would you agree with my positioning on this basically to begin with, what does GenAI do? It honestly only does a couple of things, right? It can do search and synthesis, and it can do creativity things, right? If you've got a workflow that has some of those issues associated with it, you can add value, right? If you think about a project management tool, either for leadership or individual users of it, there's a bunch of places where they run into those types of activities, right?
If you're an end user of something like Atlassian, you've got to do status updates. You've gotta search for who's doing other things. You've gotta do all of those types of activities. Plus, you have, like, annoying, like, daily updates and stuff like this, or you've gotta, like, check for things on a regular basis. You have historically offered scripting tools to do a bunch of that. You've got search capabilities. My experience with, you know, Atlassian Intelligence particularly, but I haven't done some of the scripting stuff with Rovo yet, is that it helps people do the stuff they're already doing, but much faster, right? This is like the biggest pain points in somebody's daily life.
Just like developers, you know, can bring down their time using coding tools to, you know, search for code more easily, this is allowing essentially the process management side of, you know, running any project run more efficiently. Is that a good way of, like, articulating where that value capture and why, like, people get this, like, natural, like, "I need this" type attitude and why you've gotten to 1.5 million customers or individual users already on the platform?
Yeah. What AI helps with, like you said, is, in terms of my daily job for any given persona, for any given role, it has the ability to supercharge and make that workflow more productive and save time. For example, even just internally in Atlassian, we see about 2.5 hours per week saved on average across Atlassians just by using different sets of tools. With Rovo, our Atlassian AI offering, we introduced Rovo about 18 months ago. Rovo is fundamentally, think of it as an offering for all kinds of teams to be able to build AI agents and search and chat. AI search and chat, like you described, is very useful for every single persona because typically in an organization, what happens is knowledge is dispersed across different sources.
A classic enterprise might use Google Docs, might use Confluence, might use Jira, might use Zendesk. There's a variety of different Salesforce. There's a variety of different apps that get used, but knowledge is siloed across all of these. What Rovo Search does is we've built about 50 connectors to all of these different apps such that all that knowledge is available at one single place. That use case is very useful and productive, and it's managed to save time and make search faster and more effective for practically every user across the organization. The second thing you can do with Rovo is you can build agents that, to your point, perform these different jobs and activities that a user can do day to day much faster in an automated way. What used to be scripting and writing code is now available to everybody.
Rovo Studio basically is something that any person can use. In fact, a lot of our customers tend to be non-technical in nature. HarperCollins is one of the customers of Rovo that has seen 4x time savings just by deploying Rovo agents to do things like editing, summarization, workflow, compression, process management, all of that. Internally also at Atlassian, we have over 500 agents. Most of them are written by people on the marketing team or on the comms team or on the support team, people who have never written code, but they see the value of time savings and productivity, and therefore Rovo gets adopted. The way I would think about Atlassian's AI investments is that they're spread across the entire organization and bring teams together in a more productive way.
Why Atlassian? Obviously, there's a lot of people bringing out AI tools. In fact, one thesis out there is that AI tools are gonna just replace all existing software. Why does Atlassian get to provide that tool, and why aren't people gonna just bring in other agents and layer it on top of Atlassian?
Yeah, that's a great question. Easy answer. There's two distinguishing traits about Atlassian that makes this interesting. One is we've always believed in an open ecosystem, and that has helped us gain compounding advantages over time. What does that mean? It means that we believe that every single company is always going to be in a situation where they use multiple different apps, multiple different pieces of software to manage different things. We've taken the approach of building connectors and integrating with as many applications as possible. In fact, 50+ connectors is no easy feat. This requires deep technical investment that we've done over several years such that it respects security and permissions. That's the difficult part because in an organization, not everybody has access to the same data, right?
One of the things about AI is that you wanna be very careful about the answer you give out. The answer has to be respecting permissions, has to be secure, has to be the right set of data for every user. We take the approach of open ecosystem of connecting with all of these apps and respecting permissions such that you can build AI systems that work for you based on the permissions in your organization. That open ecosystem is very unique about Atlassian. The second distinguishing thing about Atlassian is just the breadth of teams that we address. You look at any other player in market, they might have developer tools, they might have testing tools, they might have project management tools, they might have documentation and knowledge management tools. We literally have tools across the entire knowledge worker landscape.
We serve product teams, we serve development teams, we serve project teams, portfolio teams, sales teams, marketing teams, customer support teams, the entire swath of service management work, project work, as well as knowledge management work. We are pretty much the only game in town that has that kind of breadth.
You know, and one of the things that I find very interesting is there's this concept out there that if you own the data, that's who's protected. I think you're emphasizing what I believe as well, which is if you own the context, you're protected 'cause the context is what's important. Data's easy to access. Like, particularly structured data, you can train anything to go deal with it. When the context is at your level and you've done this across the entire ecosystem, you can't use any third-party tool because it doesn't have any of the context to make any use and do anything useful for your organization. That's what you own, and that's like incredibly powerful. Let's talk about the risk now.
I think you and I may share a similar opinion on this, and maybe it's just 'cause, like, we come out of the product world, and so of course we want more of us in the world. I think there's a big narrative out there that I'm fighting daily around, hey, all of these co-pilots are gonna make it doesn't matter if it's software development, you know, you're NASA, you're whoever, you're gonna need fewer people on your product teams because we're gonna have AI building everything. And because you're a seat-based business that largely is in the, you know, ecosystem around a product team and all the adjacencies to that product team, that's gonna be bad. How do you think about that?
What gives you comfort and excitement for the future of the scale of number of humans that are gonna be using your collaboration tools in a world that they are using co-pilots of whatever type to make their job more efficient?
Yeah. This goes back to the point around, what happens in the future. What do you believe AI will help people do? Fundamentally, everything we've seen so far, all the evidence points to a place where more things get built, more content gets created, more software gets created, more people are able to create things than we were able to do before. Now, Atlassian is in the business of teamwork. How do you help people create these new things? How do you help people create new software? Is it gonna be the same people as we used to serve before? No. That's a good thing 'cause it's expanding the number of people we can serve, the kind of people we can serve. We see that evidence already. It shows up in the business today already.
Jira Service Management's one of our fastest growing business, over $600 million. Jira Service Management gets used by a lot of different teams who have nothing to do with software development. These are HR teams, these are marketing teams. The teams that are actually supercharged by AI, they use our products because it makes them better. The licensing model, specifically, we have consumption-based pricing, and we've had this for quite a while. I do think consumption-based pricing will get more interesting and important in an AI world because as people unlock more productivity and as people learn to do more things with AI, they will also consume more of these kinds of capabilities. We already have about 70 different kinds of consumption-based pricing levers built into the product in Atlassian, into the product portfolio.
Rovo is our AI offering, also comes with a bunch of consumption-based meters. I think those things will get more interesting in terms of how does the business evolve. Fundamentally, as usage increases, the business is going to benefit from this overall.
Yeah. And I think what you're saying is we can make an argument, and I wanna continue it slightly longer here that the seat side might be fine, but in addition to that, you're adding an additional pricing lever that may actually give you further upside. It's like, you know, a box, right? If we just have a line, like, you're only growing so far, but now we're actually having area, right? Price and quantity could be going up because of this, and that could actually be a double tailwind for the long-term growth of Atlassian.
Absolutely.
Now, let's push on that seat bit a little bit. The way that I have described this, I find that functions in an organization are either profit centers or cost centers. A cost center is one where you have a service level, there's a certain level of output you wanna do, and you're gonna do it at the lowest cost. On those types of organizations, yes, you might have fewer humans, right? Those tend to be back office tasks, whether or not it's IT or it's finance or it's HR. On the other hand, there are profit centers. Profit centers are usually driven under an ROI type model, right? It's like, how do you decide to build your next, you know, function?
It's 'cause you have a, you know, some business case for that, and you're like, this is the cost, this is the benefit, you know, I'm gonna get to do that. The interesting thing is, AI can reduce the cost of creating the next unit, so more things are profitable.
Yeah.
You actually end up with what we've seen over the last 30 years, which is actually an expanding desire for product, not a compressing one. The idea that right now AI is uniquely like adding productivity to product development is shocking to me. I mean, I remember when I got started in the 1990s, you had to have a master's degree to be a software developer. You could be a collegiate degree, you know, a bachelor's degree in the 2000s. You could have an associate's degree in the 2010s. We have been like trying, and we've added so much productivity. I don't even see this as a massive step change relative to things that we've been doing for 30 years. What it's done is it's actually increased the number of people we want doing product. That's my argument.
Like, where would you shoot holes in that argument for like why product actually expands as opposed to contracts in this type of environment?
If you think about software, it's fundamentally supply constrained. It's not demand constrained.
Yeah.
Right? So I've never met an IT team across the 300,000 customers we serve that said, oh, we're just sitting around with nothing to do. As we increase productivity and automate some of the tasks that were not able to be automated in the past, but with AI, we now can, what happens is teams move through backlogs faster. We see this all the time with customers. For instance, at Atlassian internally, when our teams have used our own products, we've cut down PR cycle time by 45%. Those benefits accrue, and now we are able to really deploy those teams to build more products. If you see the pace of innovation coming out of Atlassian, it's markedly accelerated over the last couple of years. This is what we see with a lot of our customers as well.
I think as we make each of these creation teams a lot more powerful, more software gets created, and therefore we will see more and more usage of our products. Going back to your point of, there are some things that are profit centers and some things that are cost centers. At Atlassian, we make it a point to serve all those kinds of teams. The kind of behavior that we see is that as we provide things like customer service management, like Jira Service Management for automatic ticket handling such that you do not have to have a human handling the same kinds of questions that come up, IT help desk, new or higher onboarding questions, we are able to go deflect that and use AI to automate that.
We see a lot of those teams getting a lot more productive, and they're willing to really pay for that productivity step up, right? That's a good thing. On the revenue generation side, the creative side, we're able to supercharge the teams that can create these new things, can deploy their creativity to create gen revenue-generating objects. Those also step up in terms of now new personas enter that world, and more software gets created, and therefore we are able to serve more numbers of those customers. I think on both sides, we are seeing this. One concrete piece of evidence that we see is, you know, you hear a lot about smaller companies, AI-native companies starting out with really small numbers of people and being able to do a lot more with fewer people.
Because we serve 300,000 customers, we have a lot of SMBs within our customer base as well. When I compare some of the small-medium businesses that are highly technical, compared to the small-medium businesses that are not necessarily in technical areas, I see the expansion rate continue to be consistent across both.
Yeah. It's very interesting. Like, one of the things that I've really—it's impressed me about Atlassian, I think it's allowed it to be so durable, is because you operate at the level of coordinating how people get stuff done. As the way we work changes, you still need Atlassian, right?
Absolutely.
Like, if we needed fewer developers and way more product managers on teams, it doesn't matter. You still need to collaborate that work. You still need to assign it. You still need to track it. You still need to do all of that stuff. You're in this perfect spot. I have other companies in my own coverage that are at the tooling. If we change how we do things, it might blow up their tooling, but it doesn't blow up the fact that we need to have, you know, project management across all of this. You know, we only have a little bit of time left here. The one thing that you are emphasizing is how many customers you have. If I had to bring out the other negative narrative, maybe all the growth is gone.
Like, now it's just about trying to like get deeper penetration in the existing customers 'cause there's no incremental company out there who could possibly, you know, adopt Atlassian. You know, I've talked about you guys having, you know, 2/3 market share within like product, you know, teams, and if you take a line, you know, that number gets, you know, maybe 70%-75% of teams. Like, are you, are you done, you know, adding like new customers? Or like, if you're not, like where who, who possibly like could, could adopt Atlassian still?
Yeah, that's interesting 'cause the way we look at it, the SAM for the things we build is about $67 billion, and our annual run rate is right now at $5 billion. There's definitely a lot more that we can go grab. Like I said earlier, the 300,000 customer companies we serve, there's millions of companies in the world, and our product set really is able to serve both technical and non-technical teams. We really have a lot of upside. Very quantitatively, if you wanna look at it, I think our unique land and expand model, the business model helps us be very efficient with how we go to market. When we went back and analyzed just within the 300,000 customers, let's say we didn't acquire a single new customer, what happens? What's the opportunity there?
There's an $18 billion opportunity just within existing customers, even if we didn't go out and acquire a single new customer. And $14 billion of that is within enterprises. This is why, adding, hyper-efficient enterprise go-to-market motion for us is a clear winner, right? That we can go and access a lot more within our existing base. Having said that, new customer growth continues to be healthy for us, and we expect it to continue to grow and expand simply because of the wide variety of customers that we serve and the power of having a platform that lives in the future.
Let's finish with the enterprise bit. I did dinner with Bill from ServiceNow a month ago at their Analyst Day, and you, you've clearly poked the bear. It was very funny 'cause Gina was like, "No, we, we gotta stay focused on the largest enterprises." Bill was like, "I gotta go after, you know, Atlassian in the, in the mid-market." How do you see that, that like battle play out between ServiceNow and Atlassian that, you know, are coming in what could be perceived as more and more direct competition, particularly at this interface around enterprise customers? Is it something where only one can win or you can both coexist? Like, how do you think about that competitive setup?
Yeah. If you think about service management as a market overall, we think it's about $18 billion overall. Our Jira Service Management business is about $600 million and growing really fast. We have about 60,000 customers of Jira Service Management today. Atlassian's unique point of view is that we don't wanna be focused on just the Fortune 500 or the Global 2000. We wanna be focused on the wide variety of mid-market, small-medium businesses as well as enterprises. We often come up, in a lot of customer cases where we see displacement of incumbents. Typically, that leads to wall-to-wall deployment of Jira Service Management, but it doesn't always have to be the case. We see clear coexistence patterns, but we also see displacement scenarios.
For instance, recently, a customer that deployed Jira Service Management saw 60% cost savings across the board as they moved from an incumbent to the Jira Service Management solution. Typically, why customers choose Jira Service Management over other players is, one, the time to value. We do not really ask people to bring in a large army of consultants or people to just deploy the product. They are able to see immediate time to value, matters in days, not weeks or months. Second, the cost-effectiveness of the product is amazing. Any day of the week, we are able to argue how the value-to-cost is immensely high and in a league of its own for Jira Service Management. Third, Jira Service Management has Atlassian Intelligence built into the platform.
We see very high rates of resolution and automation simply because we are a tech-forward company that has invested in R&D for so long. We are able to really serve that wave of AI adoption inside service management, which is usually disruptive for incumbent businesses. For us, it is very additive because we are bottoms- up thanks to the land and expand model. We are able to continue to really grow and expand there. In fact, Forrester recently published a report where they said there are two dominant players in the service management market, which has center of gravity and has real pull. I am really glad to say that we are one of them.
I really appreciate you taking the time and joining us, and I know that you've got a packed schedule coming up here today. If you have any questions, please feel free to reach out to me directly. You know, I know there's probably a ton of additional context. Happy to be helpful as you kind of explore Atlassian over the coming months.
Thank you so much.