Excellent. Thank you, everyone, for joining us. My name is Keith Weiss. I am the Head of the U.S. Software Research effort here at Morgan Stanley, and I'm very pleased to have with us from Atlassian Anu Bharadwaj, President of Atlassian. So thank you for joining us. Thank you so much for having me, Keith. Great to be here. I think it's a super exciting time to be talking to Atlassian for multiple reasons. One is kind of where you guys are in the product cycle, and I think it was a really kind of pivotal year of where the focus shifts from getting everyone onto the cloud, and there's still some work to be done there, but now you guys are starting to see the benefits, right? You're starting to get the yield on that investment, if you will. Maybe just to sort of start off from a high level, what were some of kind of the core achievements for Atlassian if we think about the most recent fiscal year? Sure. You're right. It is a particularly exciting time for us. Our patience has paid off in terms of investing in the long term, and the business is in a really great place. So over the last fiscal year, as you've seen, we've passed many milestones. We posted over $5 billion in ARR, which was a very important milestone for us. And we introduced all of our AI capabilities across search, chat, and agents through Rovo. And that went from inception to GA in a record amount of time. And since we've launched GA, we have over a million users, a million monthly active users that are using our Rovo and Atlassian Intelligence features, which has been phenomenal. Jira Service Management, which is one of our fastest growing businesses in the company, has just crossed over 55,000 customers across the board and continues to impress customers and drive a lot of cost savings and productivity benefits across the board, and I'm really pleased with all of the new product development we're doing on top of the cloud platform, so Jira Product Discovery just launched our Premium and enterprise editions, and we continue to explore more and more new product areas such that we're building up for a strong long-term revenue foundation. Got it. One of my theses about Atlassian, which I'd love to get your view on, you build out a cloud platform. It's a true multi-service cloud platform. That should enable the company to accelerate the pace of development, right? Because it's just so integrated. You develop services, and they're usable across the product portfolio. To what extent is that true? And sort of what type of improvements in velocity can you see now that you're on the other side of the hill? Yeah. So you can see the benefits come out in various different ways. So the important thing to understand about the Atlassian product development philosophy is that we're very much focused on the fundamentals of user adoption. For us, it's important that users are engaged and retained using our product. And then we build on top of that across cross-sell, upsell, expansion of seats and usage. What we have done is we've invested in the cloud platform pretty heavily over the years. As a result, what we can do is we can build more and more of a virtuous loop where the R&D investments in the platform can then be harvested across all of our different markets. So Jira Service Management, for instance, was one of the first products that we built on top of this cloud platform. And as you noted, it's one of our fastest growing businesses. We recently, over last year, I think we talked about how big of a business it is, well over $500 million. To start a business on top of the platform and grow it at a rapid pace takes investment in the platform, such as the common Teamwork Graph and the platform shared services that power these products really help us get faster and faster. One of the useful proof points to look at is the rate of velocity with which we announce newer products and innovation in the product line. Rovo was a great example. We've released Compass, Jira Product Discovery. You see us getting faster and accelerating our path to serve newer user personas across the software development lifecycle. The uniqueness of our customer base, such that we serve over 300,000 customers, many of whom are SMBs and many of whom are large enterprises, the common cloud platform really helps us create this enterprise-grade, hardened platform that can serve customers across that wide swath. Got it. Got it. So real tangible benefits on the product development side of the equation, the velocity of product development. You've also noted that cloud migrators expand seats 25% plus post-migration, are twice as likely to adopt Jira Service Management. So you're seeing benefits in terms of sort of that upsell, kind of cross-sell motion into the base. Why is that? Why does the cloud enable those gains as well? Yeah. That's an important part of the product strategy. And it was very intentional in the way we built it. Lots of companies claim to be platform companies. But the real test of the platform is, have you built something that is common, unique, and shareable across different teams in the organization? So for our customers, the ultimate test of value is using the Atlassian platform. Does it help all of their teams collectively get better? So it's not just a point solution of, did your developers get better, or did your customer support agents get better? The common cloud platform really enables us to bring these innovations to all of the different teams. So what happens with a cloud platform is, one, we have a deeper understanding of what each user persona, of what each team type is using our products to do. It's just inherently an advantage of the technology itself, and two, we are able to build these integration scenarios through the platform such that, for a very specific example, if you've tried to build a new product internally, Jira Product Discovery really helps you figure out what are ideas that are valuable for customers. Because we have Jira Service Management, where there have been tickets raised, and we also have Jira Software, where there have been previous bugs, previous products, previous features, all of that information aggregated. The Teamwork Graph that powers all of these products is immensely powerful enough to be able to say, oh, we can make PMs better using the same platform, and therefore, cross-sell becomes a very obvious thing to do for customers. The data becomes this connective tissue, if you will, across all these solutions, makes them all better at the same time, and also gives you guys more visibility to how the customers are using the products on a go-forward basis. Yeah. And an important distinction is that the connected platform is not just across our first-party applications, first-party products of Atlassian. But philosophically, we've always been an open platform. So over the last 20 years of building software and servicing different kinds of teams, we've made sure that we have built connectors at every step such that the work that gets done in Atlassian products, but also across third-party products like Slack, Salesforce, et cetera, we're able to bring knowledge of that right in the surface area of where users do their work. Yeah. It's an important point too. I've always thought of that as one of the strengths of Atlassian, is that it's not about trying to force consolidation or use of all of our tools. It's more about almost confederation, that you guys are going to be the organizing principles across these workflows. Some of it's going to be your first-party tools. Some of it's going to be third parties. Absolutely. Yeah, and that's what creates a durable strategy as well, and one that works across the wide base of 300,000 customers. Because some of our customers do do the consolidation of platforms because that's what suits their particular situation, but a lot of our customers use coexistence as well. Got it. Got it. As a software analyst, what I'm always looking for is things that are underappreciated in the marketplace or overappreciated, but in this case, underappreciated. And I think the expanse of the solution portfolio from Atlassian is one of those things that is underappreciated. You guys often talk to the System of Work as the Atlassian North Star. And it's a broad vision. It's about your customers using you as a workflow automation platform to automate teams across all different types of business functions. So I'm telling you, I think investors underappreciate it. Where are you in terms of customers appreciating it and understanding sort of that broader value proposition of Atlassian? Yeah. Yeah. So the System of Work is really our amalgamation of the Atlassian platform being useful for different kinds of companies across different industries and also across different phases of lifecycle of companies, right? One of the advantages of having 300,000 customers is that you get to see a wide variety of how different companies are run, how different businesses are run, and how best practices in some areas might not necessarily be best practices in other areas. So the System of Work is really our learning of over 20 years of serving these kinds of various industries, various sizes, various phases of customers to say, here is an opinionated way in which modern companies, modern tech-forward companies can work. And this is something that we get asked a lot. Literally, every customer conversation I have usually comes down to, so what do you do at Atlassian to solve this problem? Or tell me something that other companies in my category do. And how can you help us institute that kind of a system? So that kind of opinionated teamwork System of Work is really what we want to be able to provide for every customer. Now, there isn't going to be a one-size-fits-all. That's why the beauty of the Atlassian platform is that you can adapt it and make it your own. So think of it as we provide our customers a set of Lego blocks through the Atlassian System of Work. And they can then use those Lego blocks and construct the system that they want. And what we do through our go-to-market motion, through our channel partners, through our own self-service mechanisms, is we show you the different kinds of templates that you can build. Here's how you build a spaceship. Here's how you build a truck with those Lego blocks. And how do you put that together in a way that makes sense for your teams? And fundamentally, we believe that every knowledge worker across the company should really have a set of foundational products. Those are Jira, Confluence, Loom, and Rovo. And using that effectively helps make every single team, whether it's a technical team building software or a non-technical team trying to go sell or market whatever you're building out there in the field, it makes each of those teams better at what they do and better at the collaboration across the board. So, as a leader, it gives you the visibility around what exactly is happening in your company, how do you allocate capital effectively, where exactly are work bottlenecks, and what are good practices for tech and non-tech teams to really collaborate well. Got it. So it sounds like one of those templates is, on a go-forward basis, going to be how do you run a winning F1 racing team as well, right? Indeed. Yeah. We're very excited about the Williams partnership. It's a very cool partnership. I wanted to dig into the go-to-market side of the equation as well. Because I think one of the questions I get a lot from investors is, can the self-service motion get this done, right? Can the self-service motion really show and sort of get your customers over the line in terms of using the system more broadly across their organization to solve those more strategic problems? So does the go-to-market have to evolve? Does it have to change to be able to sort of meet the opportunity? Yeah. One of the great blessings of having the 300,000 strong customer base and having the bottoms-up flywheel motion that we've perfected over 20 years is that it gives us growth opportunity on both sides, on both the self-serve side and on the enterprise side. We've quantified before that we serve around 85% of the Fortune 500 companies today, but they only constitute about 10% of our revenue. So the System of Work gives us the base in which we can expand and go wall-to-wall within enterprise customers that we already have. And that we've quantified to be a $14 billion opportunity. So that will require some kind of GTM evolution, which is what we've invested in. And I'm really excited to have Brian Duffy on board as our new CRO. He continues to think about, as he comes on board, how do we help evolve our GTM tactics such that we can roll out the System of Work easily and in a way that's accessible to our larger enterprise customers. But the low-touch part of our business, as you know, is a substantial part of our business. And that self-service motion is working very well. So you've seen our SMB seat trends stabilize over the past couple of quarters. And SMBs continue to be a big portion of our business. And the System of Work, the way it manifests there, is a lot of our SMB customers already use more than one of our products. And like you noted, the cloud business really helps us do cross-sell and upsell in a much faster way, in a more organic way. So 25% of seat expansion happens in year one of migration, whether it's small or large companies. And so we don't really need GTM evolution there. But we do continue to invest in product improvements and product optimizations across cross-sell. So how do you use Jira, Confluence, JSM, Loom, all of this together? And why does that unlock superpowers that you did not have before? And AI is positioned perfectly well for us such that we can build these agents and cross-product search, et cetera, to really unleash the productivity of teams that are at the intersection of using these multiple products together. Got it. So Brian Duffy came on board as a new CRO. He's evaluating kind of the go-to-market motion. You've talked to us about he's in that evaluation mode. Software investors and even software analysts, we're not very patient. You don't think. Any hints, any clues you could give us in terms of perhaps sort of the magnitude of investment or potentially the magnitude of disruption that would come with changes that Brian is thinking about? Yeah. Like we've always said with Atlassian, you've seen us have a multi-year track record of being really patient and structured and thoughtful about execution. The flywheel model is the beating heart of the business model that Atlassian has, so we are being very thoughtful and measured about evolving GTM strategy, so it's not going to be a disruptive, "now we are going to change what we are doing fundamentally and start doing other things." It's really more of an evolution of how do we complement the model that we have through thoughtful incremental additions across not only sales reps. For example, we have a very small number of sales reps compared to the overall population of the company. We expect that to continue to grow. But we also expect that it continues to meet a high productivity bar that we've already established. There is a success team that we believe is going to be important for customers who adopt the System of Work such that we can handhold them across the board and help them implement this fully across their entire organization. There is the channel evolution that we continue to work on such that SIs can partner and be more of a strategic partner for a lot of our larger enterprise customers that expect us to elevate more of the service that we provide, but underlining all of these things, I think, is important to understand about the Atlassian business model that customers already use multiple of our products together and want to use more of these products across the organization. So in many ways, they're really asking us, help us standardize on your platform and help us adopt and roll out these platforms, this platform, which is a great place to be in. Got it. Thoughtful evolution. I like that. That makes me feel a lot more comfortable about my margin forecast on a go-forward basis. So I made it about 15 minutes without talking about AI. That's about as long as I can go. You guys have rolled out a lot of AI within the portfolio. I think there's over 50 Atlassian Intelligence capabilities. You've talked about over a million monthly active users of that AI. We talked about Rovo as more of an agentic framework. You've talked about customer usage interactions of 25 times year- on- year. So a lot of activity taking place, a lot of positive feedback we hear from the customer base on this AI functionality. At the end of the day, we're all investors. How do we monetize this? So how are you guys thinking about sort of the value proposition of what Atlassian Intelligence and Rovo brings? And what's the right monetization avenues for that? Yeah. The way we think about monetization in AI is very consistent with the way we think about monetization of value propositions across the portfolio, which is first we want to focus on adoption and engagement. Because ultimately, our business works by are our users happy, our customers using our products, and are they situated in a way that they can expand and use more of our products across more teams, but also more and more of our portfolio itself? So our approach with AI has been exactly that. We are effectively focused on how do we maximize adoption, how do we discover the specific use cases that actually make people's lives better, that actually make them more productive. And by virtue of the broad portfolio that we have, the kind of cross-product search, the kind of cross-product agents that we have are quite discrete and different. So for example, one of the more popular agents that we have that get used internally a lot is also one of the ones that are popular externally called, where you already have a Jira ticket that expresses a user story or a spec or a requirement. And the agent basically, with one click of a button, is able to generate a bug fix or is able to generate code to implement said story. And you can directly integrate that into your deployment lifecycle. Now, that is an important way in which you can make a developer productive. But we don't just serve developers, right? We serve designers. We serve program managers, product managers, customer support people, service people. So we've created different agents for each of these different personas. Our intent has very much been to discover which of these agents get used the most and what is the one that is most valuable. Because some of these are more automatable than others. And I think it varies with time. So our focus will very much be on figuring out how we take the 1 million MAU and keep increasing that more and then figure out, once we have the value proposition really clear, how do we then monetize it? So there's two channels to monetize it. One is the classic seat-based pricing that we currently have. The other is consumption-based pricing. I expect consumption-based pricing to become more important for us because just the nature of technology and the nature of usage of AI capabilities. But we already have about six different ways in which we do consumption-based pricing in the product today outside AI. So we have JSM assets, JSM agents. We have build pipeline minutes. So this is not a model that I worry about. Do customers understand this? They are quite familiar with this. And we are currently using CBP on a lot of our AI-based functionality. But we're not in a hurry to max out the CBP channel. We expect that this will become more important. And we'll see how it evolves over time. Got it. When I think about your kind of pricing model, there's more Atlassian Intelligence functionality in some of the higher value SKUs. In the most recent quarter, you talked about that movement to higher tiers increasing by 40% on a year-on-year basis. Am I wrong in thinking that that's also kind of another monetization channel for AI of just getting people to come to those higher value solutions to get to the good stuff? Oh, yeah, absolutely. I think that is one great way in which we can monetize. But that's the beauty of the Atlassian portfolio. We have so much breadth as well as depth that there are several points. Think of it as a matrix of different points at which we can monetize functionality. That's why we are hyper-focused on how do we continue to use our impressive R&D engine, one that we make a lot of investment in and we continue to make a lot of investment in over the long term to figure out what are future monetization points we can build. For example, there is a customer on Rovo, Procore, that use a lot of our agents. And one of the ways that they use agents in, which was a pleasant surprise for us, was in terms of user-defined agents. So not only are they using the first-party agents that we have shipped, like the work breakdown and Confluence summaries, but also they're building their own agents such that they're making use of their own Atlassian-powered Teamwork Graph. And they're very sticky. So really, there are future avenues through which we can figure out which of these are the most valuable and where should we monetize. But creating that kind of sticky adoption is going to be the primary focus. And we believe very strongly that if we are able to provide that kind of value and stickiness, monetization should follow. Got it. So if you guys rolled out Rovo, which is both sort of your collection of first-party agents, but also a platform for customers to build agents on a go-forward basis, I think Atlassian first started talking about it a lot in the New York-Barcelona conference last summer. Investors started getting excited about it. I started getting excited about it. And now everyone is talking about agents. And the market seems to have gotten very crowded and very confused about sort of what are agents, what aren't agents. What's the Atlassian view? How are you guys defining Rovo? What's Rovo trying to accomplish for the end customer? And why does Atlassian have the right to win when it comes to that agentic architecture? Yeah. Yeah, that's a great question. I guess in all of the different tech kind of hype cycles, some terms get popular. Some terms start getting overloaded. And I think agents. You can't keep us penny off. I think agents are one of those where the term is used quite effusively. At Atlassian, our definition of agent is very discrete and clear. We believe that agents need to have certain goals, so it's not just a piece of feature or functionality that exists in product. It needs to have certain goals. It needs to have the ability to take action. We also believe it needs to have personality. It's one of the quirkiness of the product strategy at Atlassian, where we strongly believe that because we are focused on teamwork, if AI agents are going to be part of your team, we believe that they will have a personality of their own. They will have goals of their own. They will have a certain degree of autonomy that you can control as the person who orchestrates the agent, operates the agent. But our vision very much is that there will exist agents that help partially or fully automate certain jobs that people do. As an example, if you are an HR person that's trying to answer questions about legal policy or relocation policy, we can help you automate that down either partially or fully, depending on the deployment that you have. So agents can then answer those questions for you, approve workflows for you, et cetera. But there can be several AI-based productivity enhancements that are not at all agents, but that are still built into the platform because of the powerful Teamwork Graph that understands objects across different contexts and has a certain amount of intelligence end to end, right? So our vision very much is that we will continue to build agents for specific jobs that specific user personas do. We will make them available as part of the products that we currently have, and we will also give you the ability to create your own agents. Like I said, with the Procore example, the stickiness there comes from different companies have different processes, different ways of working. They have different personalities, right, and so we want to give our customers the biggest canvas to then be able to create their own agents and give them the kind of power and autonomy that they can control. Because I think that's very important in a B2B setting where trust is a first-order bit, and we want to make sure that agents come with a requisite level of trust and security and your ability to control what those agents can do. In the future, really, we think of teamwork is going to be consisting of teams that have humans as well as these AI agents. Got it. On the risk side of the equation when it comes to generative AI, and particularly when we think about Atlassian, while you have a much broader business than just developers, developers is a very big business for Atlassian, you have a lot of seat-based licenses to your software based upon the number of developers. I was talking to Sam Altman this morning, and he threw out some wild figures. He's expecting software development to get 10x, 100x more productive on a go-forward basis. I mean, I'm not hearing about that in the wild, but it does create a concern for investors of, does that part of your business remain durable on a go-forward basis? Are we going to have as many software developers on a go-forward? Or does Atlassian have offsets that will enable you to continue to monetize the software development process as effectively if we don't have as many software developers? So how do you think about that dynamic? Yeah. So I think it's difficult to predict how quickly the productivity gains will come. But we firmly believe the productivity gains will come. There is ample evidence that people are going to get a lot more productive. I think developers, specifically with respect to coding, we have enough evidence to see productivity gains. So they're 10x, 100x more productive. We welcome that scenario. I would love to get there as soon as we can because it's a great thing for Atlassian. Because what we do is really we help teams create things. We help teams build whatever they think is valuable and then help serve their customers with this thing of value. And in that whole lifecycle of building something of value, coding is just one part of it. And in fact, coding represents a fair amount of work and time and effort and money today. And the more that gets simplified and gets cheaper and more productive, the more software gets built. The more tech-forward companies can do things that they never imagined. And if you look at the Atlassian platform and portfolio, we serve the end-to-end process of software creation. So if the coding part of it gets more productive, or if one role in the software development lifecycle, one role in the team gets better, that's phenomenal for us. That's awesome. So we'll make that persona as productive as we can. And then we'll help the rest of the team get even more productive around that gain. Got it. Just to make sure I'm thinking about this correctly, the value proposition of what you're bringing to your end customer with this software development platform isn't automating any single software developer, but it's the process, right? And if we're doing more software development, if there's more process getting done, perhaps we have to change the pricing model to accrue that value. But you are providing a lot more value. And you have confidence in Atlassian's ability to figure out the new pricing model that aligns to what you're driving for the customer. 100%. Because in the long run, what really matters is the value proposition that you provide as a vendor to your customer. And you've seen now 20 years of evidence that with the powerful R&D investment that we have, which is very contrarian in the market, what we do is we continue to stay ahead of the curve in terms of what is valuable for our customers. So we want to be a strategic partner for them to help lead them into this new era of what software development means in an AI-native era, in an AI-enabled world. How can we help our customers get that much more productive? And the mechanism of how we monetize that might be through seat-based consumption, seat-based pricing, consumption-based pricing, whatever. Those are the mechanics of it. But the fundamental thing to understand is how do we help people create more things of value? How do we help teams operate better? And that's just going to increase over time. Got it. If we turn the discussion around to sort of internal to Atlassian, you guys are the primary or not the primary, but the first users of all your solutions as well. You're using Rovo internally. You're using these developer productivity tools internally. Can you talk to us about maybe some of the initial sort of leverage and efficiency gains you've been able to yield from these tools? And help us relate to sort of the margin targets that Joe's put out there. How important is this in terms of you guys improving the operating margins on a go-forward basis? Yeah. So, straight up, I would say we are committed to the 25% operating margin that we said we would hit by FY27. That continues to be a goal. Nothing has changed. Whatever happens in AI or doesn't happen in AI, we're committed to hitting that goal. In terms of internal usage of agents, we've had a longer lead time because what we do at Atlassian is we build a lot of products for ourselves, try it out in anger first, and then we productize it. And Rovo is no different. So, three areas where Rovo usage has been really impactful for us inside the company. One is the developer set of agents, where we use AutoDev, AutoFix, AutoReview internally. It saves our developers a good couple of hours each week across usage of all of these agents. That has been very helpful. In fact, I just saw a Loom video of one of our developers showing how they fixed 11 accessibility bugs in a row using one of these agents. And all it took was a few clicks. So that's been phenomenal. A second category of agents are around the service agents. So we use a lot of internal IT help desks. And that has helped us automate a significant part of our service tickets, of our internal requests, all of those. And customer support's another place where we have seen phenomenal productivity gains. And our Customer 360 agent, which has actually it's one of our customer agents that we built for ourselves, where we have our salespeople use that regularly to figure out what's the footprint of a customer and how their site health is, how their account health is. That has also saved several thousands of hours for us internally. So we continue to be excited. And we are a tech-forward company. So a lot of our employees continue to be early adopters of this technology. OK. Outstanding. So I'm going to try to sneak in one last one. So as we look ahead into 2025, what are some of the key milestones or metrics that we as investors should be looking at within Atlassian to understand how well you're progressing against these goals and these innovation targets? So three things I would look at. One is our continued success in enterprise adoption. And we express that in terms of number of customers that spend over $1 million with us. You saw we had a record number of those over the previous timeline. And just penetration into enterprise going wall-to-wall, I think that'll be one important stream to look at. AI is another important stream to look at. And there, I would point more towards adoption. And we'll think about more and more of these public metrics that we can put out. But that's effectively what we're laser-focused on. How can we make sure that we have useful applied cases of AI all across the portfolio? And the third one, and one of the most important ones, is what we call the System of Work. How many of our customers are able to deploy the System of Work successfully? To what extent? And how do we support them both across SMBs and enterprises? And again, there, we will continue to evolve what kind of public metrics we can put out in terms of how we are making progress on that journey. All right. I'm looking forward to those metrics. Atlassian continues to be a very dynamic, very exciting story. Thank you for coming and sharing it with us. Thank you so much, Keith. My pleasure.