Loading…
Loading…
Good afternoon, and thank you for joining Atlassian's Earnings Conference Call for the Third Quarter of Fiscal Year 2025. As a reminder, this conference call is being recorded and will be available for replay on the Investor Relations section of Atlassian's website following this call. I will now hand the call over to Martin Lam, Atlassian's Head of Investor Relations.
Welcome to Atlassian's third quarter of fiscal year 2025 earnings call. Thank you for joining us today. On the call with me today, we have Atlassian's CEO and Co-Founder, Mike Cannon-Brookes; and Chief Financial Officer, Joe Binz. Earlier today, we published a shareholder letter and press release with our financial results and commentary for our third quarter of fiscal year 2025. The shareholder letter is available on Atlassian's Work Life blog and the Investor Relations section of our website, where you will also find our other earnings related materials, including the earnings press release and supplemental investor data sheet.
As always, our shareholder letter contains management's insight and commentary for the quarter. So during the call today, we'll have brief opening remarks and then focus our time on Q&A. This call will include forward-looking statements. Forward-looking statements involve known and unknown risks, uncertainties and assumptions.
If any such risks or uncertainties materialize, or if any of the assumptions prove incorrect, our results could differ materially from the results expressed or implied by the forward-looking statements we make. You should not rely upon forward-looking statements as predictions of future events. Forward-looking statements represent our management's beliefs and assumptions only as of the date such statements are made. And we undertake no obligation to update or revise such statements should they change or cease to be current.
Further information on these and other factors that could affect our business performance and financial results is included in filings we make with the Securities and Exchange Commission from time to time, including the section titled Risk Factors in our most recently filed annual and quarterly reports. During today's call, we may also discuss non-GAAP financial measures. These non-GAAP financial measures are in addition to and are not a substitute for or superior to measures of financial performance prepared in accordance with GAAP. A reconciliation between GAAP and non-GAAP financial measures is available in our shareholder letter, earnings release and investor data sheet on the Investor Relations section of our website.
We'd like to allow as many of you to participate in Q&A as possible. Out of respect for others on the call, we'll take one question at a time. With that, I'll turn the call over to Mike for opening remarks.
Mike Cannon-Brookes: Thank you all for joining us today. As you've already read in our shareholder letter, we delivered total revenue of $1.4 billion in Q3, driven by cloud revenue growth of 25% year-over-year and a free cash flow margin of 47% for the quarter. We continue to make great progress in our strategic priorities of serving the enterprise, making rapid advancements in AI, and connecting technology and business teams through the Atlassian System of Work. Earlier this month, we held our annual user conference, Team '25. Over 5,000 customers, partners and Atlassians gathered in Anaheim as we ushered in the next evolution of the System of Work, drove Rovo's incredible AI capability to the center of the Atlassian platform, introduced two new cloud offerings, and much, much more. Team is always such an energizing moment for me and all Atlassians as we get to see the reaction from our customers and partners to the incredible innovation that we're delivering and hear firsthand how they're using our solutions to transform the way they work. The next evolution of the Atlassian System of Work is officially here as we've started making the shift from standalone products to a vision of apps and agents with Rovo at the center of everything. By making Rovo included in all premium enterprise subscriptions of Jira, Confluence and Jira Service Management, with the standard edition soon to follow, we're democratizing the power of AI and accelerating human AI collaboration. We already have over 1.5 million monthly active users of AI across our platform, and we expect this number to continue to grow strongly week-on-week as we roll out and as more customers realize the power of Rovo's capabilities. If you'd like to hear more about how this sets us up to win in the AI era, check out the Loom that I just posted to our IR website. We continued to also make steady progress towards unlocking the Atlassian Cloud Platform for even more of our largest and most complex customers as we achieved FedRAMP Moderate authorization for our U.S. federal government customers and their industry partners. We expanded our cloud platform with the Atlassian Government Cloud and also announced our Atlassian Isolated Cloud, a single-tenant cloud solution for enterprises with highly sensitive data. All up more than 300,000 customers, including Mercedes, SAP, Workday and Xero rely on Atlassian to help their teams work better together and unleash enterprise knowledge across their organizations. We believe the progress we're making on our key strategic priorities will continue to fuel durable long-term growth and help us scale to $10 billion in revenue and beyond. With that, I'll pass the call to the operator for Q&A.
[Operator Instructions] Your first question comes from Keith Weiss from Morgan Stanley. Please go ahead.
Yes. This is Sanjit Singh for Keith Weiss. Thank you for taking the questions and congrats on the 1.5 million in AI monthly active users. I wanted to talk about some of the decisions of the business, Mike, that you're making.
One being sort of embedding Rovo into the core products and that being potentially a trade-off in sort of near-term revenue growth. When we think about that maybe in relation to the mid-term outlook of sort of 20% CAGR, does that sort of move the needle from that perspective? And then when you think about some of the uncertainty in the macro, does that also sort of cause the TEAM to reassess that 20% growth outlook? Thank you.
Mike Cannon-Brookes: Sure, mate. I can take that one. Joe might want to add on at the end there. Look, the decision to embed Rovo in the core products, I would say, look, we are seeing continued growth and adoption of Rovo and our AI capabilities, right? As you called out, past 1.5 million monthly active users, continues to grow really strongly. Sales of premium enterprise additions are up over 40% year-on-year. And the cloud customers who have deployed Rovo agents to now tens of thousands of unique workflows connected to automations are seeing great value. So I would say that from that -- excuse me, position of strength, we feel very good about the capabilities that we've built and we want to get them into as many hands as possible and as fast as we can amongst our user base, right? We believe that by maximizing usage and adoption, we -- yes, we've always been patient with monetization, but we think we have a massive user expansion opportunity, both in terms of additions and also going wall-to-wall with more knowledge workers in those organizations. So that fits really well with the way that we think about building software in general. Again, we're always very thoughtful about balancing growth and capital usage. As we mentioned at the Investor Day, we've done some amazing work in R&D to make those capabilities able to be deployed at a reasonable cost, which is obviously a gateway in order to be able to make this sort of a move. The mindset you asked for with the decisions behind the decisions, look, we do think that the future of Teamwork is going to be about this sort of iterative human AI human agent collaboration. We are, I think, very forward-thinking in how we are doing that and implementing that. At the moment, we have a great offering. So we want to get that out to as many users as possible. We do see that it enhances our stickiness and promotes the wall-to-wall adoption of the Atlassian Cloud Platform when this works. Again, we have a huge amount of organizational knowledge that we have access to and giving them back to our customers in unique and interesting ways and naturally makes them come back to our platform, which is a great advantage. And this really removes the friction to get more customers started on that journey, which is really important that they can do that really quickly so that it can become a continued accelerant for Atlassian's continued growth in additions and more knowledge workers. And as I mentioned at the top, look, we have a really world-class R&D team that has built some amazing features in Rovo and all the capabilities and we hence want to make sure that they come through to as many customers as possible. We have a pretty good history, I think, of making these sorts of bold business decisions, you might say, everything from PLG to free editions to moving to the cloud. So I feel really comfortable that we are coming from a position of strength and that we have the team that can follow through and execute against this to deliver the maximum amount of user value we can. Maybe I'll let Joe follow on the CAGR and the long-term nature from a financial perspective.
Yes. Thanks, Mike. There's nothing in the macro that changes our thinking on our commitments that we made at the Investor Day. In general, I would say we remain confident and on track to the plans we laid out at that time.
From a revenue perspective, we do expect to tap into large market opportunities to drive healthy revenue growth across cloud and data center through our strategies that Mike talked about earlier around enterprise, AI and System of Work. But all of that culminates in our ability to drive revenue growth in excess of a 20% CAGR through FY '27. Rovo net helps that and we don't see anything in the macro right now that would prevent us from accomplishing that.
Your next question comes from Gregg Moskowitz from Mizuho. Please go ahead.
Great. Thank you very much. Two related revenue questions, if I may. Just first on cloud.
The shareholder letter mentions that enterprise deals landed later than expected in the quarter. Joe, was that back-end linearity enough to impact the Q3 cloud revenue growth? And then secondly on data center. So tapping duration at one year clearly should provide longer-term benefits.
It is, however, causing some investor confusion right now. Just given that data center has meaningful upfront revenue, are you able to provide some context on what impact the duration change had to data center revenue growth for Q3 as well as what's being assumed for Q4?
Yes. Let me start with the cloud question. So we definitely did see deals fault more back-end loaded in the quarter than we expected. That did have impact on the cloud revenue growth rate that we delivered in the quarter.
As you know, from a billings perspective, that still is healthy and we expect that revenue to be recognized in Q4. We just recognized less in Q3 than we expected. From a data center perspective, I would just say growth in the quarter was primarily driven by pricing and it was partially offset by strong migrations to cloud and as you mentioned, fewer multiyear agreements. I would just highlight that other drivers for data centers such as customer retention, renewal rates and expansion were healthy. So overall, we felt it was a good quarter of fundamental underlying data center performance.
I would say if you think about the impact of multi-year agreements, we do recognize revenue in the period that we sign those agreements based on the total contract value. So when the duration of those agreements is shorter on average, we do recognize less revenue in the quarter and that was definitely an impact on the data center revenue performance that we had in Q3.
Your next question comes from Arjun Bhatia from William Blair. Please go ahead.
Perfect. Thank you. Joe, actually, if I could follow up on that question about deals landing later in the quarter on cloud. Is that just kind of idiosyncratic customer timing?
Is that migrations or is there something else at play that we should think about? And then maybe a broader question, just as we're thinking about cloud -- the cloud platform parity relative to DC, it seems like we're there yet. But I'm curious if there's still kind of hangups that customers have on migrating and those that are sticking around on DC. I'm curious what they're still waiting for outside of internal processes and change management, if anything at all.
Thank you.
Yes, thanks for the question. I'll go first. In terms of the deal timing and the linearity in the quarter on the enterprise side, I think it's really a function of the fact that we're doing larger, more complex deals. They're taking longer to close than we expected.
This is a muscle and capability we're building in the company. And so especially when you have the level of innovation and announcements and change that we're driving, customers are very interested in that. So we're spending more time with them, walking them through those options, helping them understand the value that we deliver and the value that we have to offer. So as a result of that, we are seeing some elongated deal cycles.
We don't attribute it to macro. It's strictly a function of the nature of the deals that we're trying to drive with our customers. Do you want to chime in on the back half of the question in terms of customers from data center to cloud?
Mike Cannon-Brookes: Sorry, I can chime in on the back-half of the question, of course. Look, the cloud platform parity with DC, I think we feel really great about where we sit. We every quarter continue to help customers to upgrade from DC to the cloud. We had major wins during the last quarter with the FedRAMP authorization and the announcement of Atlassian Government Cloud and we have a number of customers already in that early access program. So for obviously federal government customers in the United States or their partners, anyone who works with them, that's a big step for that particular group of customers. We also announced the Isolated Cloud, which helps for a different group of customers that want isolated compute, storage, networking and all of the services. We continue to work really hard on a lot of our other standardization. For example, Rovo is now ISO 27001 compliant, Rovo is now SOC2 compliant. So a whole series of continued compliance work that the team has done a really great job besides scale and performance. So I would say we feel really great about where our customers are. Again, cloud is a fundamentally better experience for Atlassian customers than data center is. And my experience from Team '25 talking to a whole series of different companies is that it's again not a when -- not an if they move or upgrade, it is a when. During the quarter, we closed a deal in cloud with one of the largest investment banks in the world, one of the global leaders in memory and chips and huge U.S. - a U.K., sorry, financial services company. So we had some massive cloud deals during the quarter as well. So feel really good that that's showing up in customer traction as well.
Your next question comes from Mark Cash from Raymond James. Please go ahead, Mark.
Yes. Thanks. This is Mark on for Adam. I just want to go back to the cloud migrations and data center moving to annualized only.
Does this potentially increase the amount of cloud growth migrations in fiscal '26, maybe beyond the mid-single digit level? And just confirming that you anticipate data center to grow next year with the adjustment and the other blockers you've announced recently. Thank you.
Yes, I'd say to answer your first question, we definitely do believe that we're going to see better migration contribution to cloud revenue growth in FY '26 and FY '27. Consistent with what our guidance was at Investor Day last year, we said that those contributions would be in the mid to-high single-digit range over a three-year period and that we would see less contribution in FY '25 simply as a result of the roll-off of the server end of support. So that's how we're thinking about the migration piece. And the second part of the question? Yes, the question around data center in FY '26. It's really too early at this point to talk about FY '26 guidance.
What I can tell you is from that perspective we will -- we do expect to continue to have data center growth driven by pricing and expansion at the customers. That will be offset though, however, by increasing data center to cloud migrations.
Your next question comes from Michael Turrin from Wells Fargo. Please go ahead. Michael, are you on the line? We'll move to the next question. Go ahead.
Thanks very much. Appreciate you taking the question. Joe, the gross margin and the free cash flow margin stand out this quarter. I'm just curious if you can speak a bit to what you're seeing in terms of the overall efficiency gains, how you're thinking about your ability to continue to drive margin, particularly if the growth environment at all turns.
I know Atlassian has taken a countercyclical approach previously, but just curious how you're assessing those trade-offs currently. And just more commentary on margin as well is helpful. Thank you.
Yes. Thanks for the question. It was another great quarter on the gross margin front. As you can see, we posted gross margins of 86%.
That was solidly better than what we guided coming into the quarter. In cloud, the story remains the same, just more of it. We continue to benefit from price increases, upsell to premium edition, and importantly, engineering-driven investments we're making to optimize cloud infrastructure and support cost. And as I've said in the past, focusing on and managing cloud COGS efficiently right now is particularly important given the expected growth in that part of the business and the opportunity and strategy we have with Rovo in the AI space.
I'd also reiterate that this highlights one of the big advantages of the engineering investment model we have at Atlassian that we've talked about in the past, which is investing in world-class engineering talent that can work on unlocking efficiency improvements and cost-to-serve. And as I mentioned last quarter, some of the very best work in the company is happening in the space. So we believe there's more to come. So overall, we feel very good about the gross margin picture and trends and how trends are shaping up.
And we do believe those efficiency gains will be structural in nature and we'll be able to maintain them going forward.
Mike Cannon-Brookes: Okay. I just wanted to add on top of that. Look, we've always been incredibly prudent stewards of capital, I like to think. And balancing our approach to growth and profitability across our whole history, we've given some pretty clear long-term targets. We're executing really well, I would say, on our journey to those targets, both top line and bottom line targets that we've given out beforehand. The R&D team has done a fantastic job, as Joe mentioned, getting efficiency out of cloud at-scale and continuing to do so quarter-on-quarter. It's this sort of work that allows us to do things like including Rovo in our operations, right? It's not just about taking AI cost and some fantastical new features and making them significantly cheaper to run. It's also about the other savings we've been able to make in other areas of the cloud. So a great example maybe of how we're constantly trying to balance that growth and value delivery to customers along with the profitability and margin and doing that alongside each other. So really proud of how the team has executed here in the last few quarters.
Your next question comes from Alex Zukin from Wolfe Research. Please go ahead, Alex.
Hi. This is Arsenije for Alex. Thanks for taking the question. How has the outlook on cloud revenue changed given the back-half weighted enterprise dynamics you saw this quarter?
Was the expectation being below impacted more by better high grade deal momentum than expected? And also in the shareholder letter cross-sell adoption of high-value deals, top-of-funnel performance and retention was in line with expectations versus last quarter being in line to slightly ahead. Is there any clarification you can provide on where you outperformed the expectations last quarter versus this quarter outside of deals just landing later than expected? Thank you.
Yes. Thanks for the question. From a cloud perspective we did deliver 25% year-over-year revenue growth that was better than we expected coming into the quarter. The core underlying performance in our cloud business was very similar to Q2 with trends from Q2 continuing into Q3. The variance to our expectations was driven by two factors.
It was better-than-expected, paid seat expansion and data center migrations. All the other drivers in this part of our business, whether it's cross-sell of additional products or adoption of higher-value additions or top-of-funnel performance or customer retention, those were all in line with our expectations going into the quarter. In terms of looking at the level of beats relative to the guidance in Q3 versus first half, I would just start by saying that in both Q1 and Q2, those were exceptionally strong quarters where we executed really well against a very robust set of opportunities in pipeline.
Then I think fundamentally, Q3 is simply a seasonally slower quarter for us. So there was less opportunity for the level of outperformance you saw in the first half of the year. And then I know not specific to cloud, but more generally, there were some mechanical dynamics, particularly in data center and marketplace that negatively impacted revenue growth on the margin. And I think taking all of this into account helps explain most of the difference between our H1 performance to guidance in Q3.
Your next question comes from Kash Rangan from Goldman Sachs. Please go ahead.
Thank you, Mike and Joe. I'm on a plane. So I'll try to keep this quite brief. But curious if you can talk about the effect of tariffs on your customers' businesses.
Maybe it's a little premature. But what are you hearing from your best customers as to how they're prioritizing Atlassian during these times? And also if you could talk about the CRO transition. I know you had a new CRO join at the start of this quarter.
Could there have been some settling in and redirection of the go-to-market effort that could have resulted in the back-ended nature of the business although it sounds, [indiscernible] but just curious to get your more thoughts there. Thank you so much.
Mike Cannon-Brookes: Sure. Kash, let me take the first part and I hope you have a great flight. I'm impressed by dialing in from a plane there. Look, the effect of tariffs on the business. First thing I would call out is I think that we're in an incredibly healthy position today. We feel the business remains really healthy. So nothing to call out on a sort of macro impact at this point. Pipeline looks strong, customers continue to want to upgrade to the cloud. They're expanding their usage of Atlassian. Team '25 had everybody in the business on a high in terms of their customer reaction to what we had done. So we feel we're in a really healthy and strong position as a business. It doesn't mean we're unaware that there is a complex environment out there and we continue to remain vigilant about where that flows. I guess is probably the easiest way to say it. We are obviously -- well, I say we -- I'm now quite experienced in leading Atlassian for 23-plus years through various different cycles and periods of uncertainty, right? We had this in 2008, we had it in 2020. So we have a really strong exec team with a broad experience. And obviously I've -- we've done this before. Every time of these is different, but you can learn a lot from that experience. From a resilience point of view, look, I think we have more than 300,000 customers, pretty diverse in terms of the industries and the countries that they come from. Our go-to-market evolution and our customer base has changed since, for example, that 2020 period. We're a lot more balanced, I would say, now, right? As we shared at Investor Day, in 2020, our enterprise customers represented 15% of sales, 1-5, and today that's more than 40% of sales in 2025. So a big change in the last five years, which I think gives us a lot more balance and that's where we are there. I think, look, our focus has to maintain through any of these environments on the customers, right? Getting closer to the customers, understanding their challenges, understanding opportunities we have, right? Whatever that environment is will probably affect our customers much more than it affects us. But that doesn't mean we are not aware of that and we aren't careful of that and stay close to those customers. At the same time, we've historically done a good job in a thoughtful way, I would say, about gaining market share in those difficult times. We see a lot of customers consolidating onto the Atlassian Platform from many, many vendors. We have an example of Breville that moved from 10 different tools to the Atlassian Platform, et cetera. So there are potentially very good opportunities for us in that in terms of our competitive prices and gaining share, or however you want to phrase it during those periods of time. So feel really healthy about where we are now, really confident in the strategy and priorities we have. So we're feeling very good. On the CRO transition, look, Brian is doing an excellent job. He has been in the building, what, three months now? There's no doubt he has a huge amount of experience. He's bringing that to bear already. There's still a lot of learning to do, obviously about Atlassian. We're a large and complex business with a lot of history. I will say we have again a fantastic executive team that are leaning in to help every which way we can and a really excellent sales and success team that is already coming from a position of strength. So feel good about those transitions we have. Again, our focus areas in sales continue to be largely the same and we continue to work on the high-velocity parts of our model all the way up to the strategic customers and making it an integrated process of how we go to market across all of our 300,000 customers.
Your next question comes from Keith Bachman from BMO. Please go ahead.
Hi. Good afternoon. Thank you. I also wanted to focus on the question of growth.
And Joe, part A is when I read the shareholder letter, you called out a few things that caused you to risk adjust the guidance. And there are more macro this time than past. Are you implying that there's more conservatism in the Q4 guide than you might otherwise bring to bear? And the second question is candidly probably more important is attending the event a month ago, there was a number of questions about the impact of collections.
And more specifically that the bundled pricing, if you will, of collections was -- it's a pretty significant discount to the adding of the individual pieces. And so just wondering if you wanted to offer any thoughts on how we as investors should be, so to speak, tuning our model on how that price versus seats, how the collections may impact the growth as we look out over the horizon given the pretty significant discount implied by the bundling. That's it for me. Thank you.
Yes. Great questions, Keith. So from a Q4 guidance approach, overall, there's no change and we've taken the same approach to our guide for Q4 that we've taken all year, which is we are accounting for potential impacts of macroeconomic uncertainty and execution risk related to our enterprise go to market sales motion. I'd highlight that data center and marketplace, there is greater variability just given revenue recognition and the underlying dynamics in that part of the business around cloud migrations, multi-year deal mix and the transactional nature of marketplace revenue.
I'd also highlight that as you think about our FY '26 outlook, keep in mind, we will continue to take a conservative and risk-adjusted approach to our guidance for FY '26, just as we have throughout FY '25. In terms of the Teamwork Collection, while the Teamwork Collection is not a driver to our Q4 guidance and it will have a limited impact on FY '26, we think it is a powerful part of our strategy in transforming Atlassian. And the reason we think it makes sense is because there's an untapped attach opportunity to over 10 million Jira users. And we think we have a real opportunity to drive long-term revenue growth with this through additional attach of Confluence and Loom to Jira, additional attach of premium and enterprise versions since all users must be on the same version, incremental new seats because we have an opportunity to drive wall-to-wall and displace alternative tools.
And then lastly, AI credits will drive AI usage and that's going to strategically strengthen our structural competitive position broadly, and that will provide an uplift to our overall business. So overall we're very bullish on the opportunity for TWC, Teamwork Collection, to drive growth over the long term.
Mike Cannon-Brookes: I just wanted to add, Keith, from my point of view, maybe slightly more philosophical than Joe. Look, collections are about being as customer-first as we can and we always try to do that at Atlassian. It simplifies how our customers can buy and grow with our offerings. That simplicity, that reduction in friction does result in, in our experience, a long-term growth driver that appears over time as customers increasingly want to buy larger amounts of software from Atlassian that they want to not buy it in smaller and smaller pieces, if that makes any sense. They want to buy it in large amounts. That enables us to create quite some significant simplicity in the purchasing behavior, but also simplicity in the deployment and the rollout behavior. As Joe mentioned, that simplicity comes from having a single edition, having a single seat count and any new offerings that are introduced like Rovo, for example, into the Teamwork Collection will just appear for those users, which is a much easier experience for the customer. I believe over time that will lead to continued growth. That's part of the reason, but also lead to customer stickiness and customer happiness as more of our functions are exposed to more people within that organization and it gives us a larger opportunity to go after the knowledge workers that we don't have access to in many of our customers at the moment or who don't have access to our products. So philosophically, I think it's about simplifying things for our customers in terms of their purchasing behavior.
Your next question comes from Rob Oliver from Baird. Please go ahead.
Great. Thanks. Good afternoon, guys. Mike, for you and then, Joe, quick follow-up for you.
Just on the three cloud strategy, which seems fairly straightforward around commercial and government cloud. I'd be curious on the isolated cloud just to hear a little bit more from you, Mike, on what the strategy is there? Is that a new type of customer? Is it a customer that otherwise might not have been as willing to move to an Atlassian managed cloud that's currently in DC?
And then, Joe relative to your comments earlier on cloud COGS and cost-to-serve how should we think about as that ramps in '26 and certainly not asking for guidance there? But as we think about that ramp, how a single tenant solution option in cloud could potentially impact cost-to-serve? Thanks, guys.
Mike Cannon-Brookes: Sure, Rob. Let me take the first part of that. The strategy around Isolated Cloud is about continuing to make sure that we're serving the entire breadth of our customer base. As we scale in the enterprise, as we scale in ever larger and more complex customers, some of those customers, some very small amount of them, have particular needs. And so Isolated Cloud, with its own sort of dedicated storage, dedicated compute and dedicated networking for the customer, certainly appeals to some amount of the largest and most complex organizations and also those that have particular needs and choices. We want to make sure we're offering a set of services to all of our customers that did meet their particular needs. A lot of the architectural changes and improvements we've made over the last three years in cloud and in the cloud platform allow us to do this. Things like data residency, which you can think about deploying our cloud in different regions of the world, and then FedRAMP, which has led to Atlassian Government Cloud, allowing us to meet various conditions that are required for those federal customers and their partners. You can think about Isolated Cloud as kind of the next stage in our journey architecturally to be able to deploy, think of it as conceptually an entire Atlassian Cloud, all of the services for a single customer, and to be able to do that both at an effective R&D speed perspective, so that we can get the features out to those customers in the same way they are getting a cloud software. And secondly, being able to do that from a cost management perspective, both that the customer is willing to pay and sees value in. And secondly, that Atlassian can deliver it at the right margin and price.
And then I'll take the second part of that, and thanks for the question. I'll start by talking a little bit about the long-term structural view of gross margins and then I'll dive into your specific question around Rovo and the impact that's going to have. I'd say the savings and efficiencies to the question we answered earlier that you see this quarter, they are structural and sustainable. But as Mike mentioned, keep in mind, we'll continue to add more and more value and related costs to our cloud products to strengthen our structural competitive position.
And the strategy with Rovo is a great example of that. So while we create capacity with savings and efficiency, we also backfill with additional costs over time. So in terms of our longer-term guidance, company gross margins are going to be primarily impacted by the factors we've talked about in the past. So that's things like our revenue mix, which will increasingly skew towards the cloud and progress on our cloud gross margin improvements.
And just from our perspective, our goal is to drive year-over-year cloud gross margin improvements as we scale such that we can mitigate as much of the impact from that revenue mix shift to cloud as possible. And as you've seen over the last few quarters, we've been very successful at doing that. So that's the overall mindset we have as we look to shaping future gross margin trends. And when you net all that out, we do continue to expect blended gross margins to decline over the next two years, consistent with what we shared with you at Investor Day last May.
In terms of AI COGS in the near term, it is early days and the impact from Rovo and AI-related COGS is pretty small and we're managing it well. Obviously with the change in our Rovo pricing and packaging strategy, there will be a significant investment in COGS. But as you read in our shareholder letter, we believe that decision is absolutely the right thing to do to strengthen our structural competitive position and drive long-term growth. Over the long term, we do believe AI-related costs will come down the curve as multiple vendors compete in that space and we'll continue to optimize our AI infrastructure and support requirements just as we're doing on the other aspects of our cloud platform.
And as you can see from the results this quarter on cloud gross margin, we are driving savings in cloud COGS that creates that capacity. And so in the future, we feel we are really well-positioned to be able to manage this change in strategy from a margin perspective.
Your next question comes from Raimo Lenschow from Barclays. Please go ahead.
Thank you. I wanted to go back to Kash's question on Brian earlier. Obviously, you're kind of planning for the new financial year that is coming up. How should we think about changes to the go-to-market because Brian coming from SAP, and he's bringing like an enterprise focus that might complement what you had before.
But do we need to think about like changes to your go-to-market in that respect? And then I had one follow up.
Mike Cannon-Brookes: Thanks, Raimo. I think we're always changing our go-to-market is probably what I would say. We're evolving every quarter, certainly every year, how we go to market as our customer mix changes and our needs change. We've seen this many times in the past. We've moved from how we sold to adding data center, for example, we've moved to selling cloud and subscriptions. We've adapted to selling to the largest enterprises in the world. Again, well more than 500 customers now spend more than $1 million a year with Atlassian and we've grown that enterprise business, as we mentioned, to over -- from 15% of the business to over 40% of the business in sales and, obviously, cloud additions in free, but also in premium and enterprise. So what I would say is we're continuing to evolve our go-to-market motion almost every quarter, but certainly every year as we look forward and see what the customers need and what the customer base needs and where we think the biggest opportunities are. There's continued experience, obviously, that comes with Brian, from a large global enterprise customer base to some of our largest customers and how we can continue to serve them very well. At the same time, we continue to learn, both Brian, myself and everybody in the team, about the large-scale frictionless end of the model that we're running, right, with 300,000 plus customers nowadays and many, many offerings. New offerings like Loom and Trello have very different sales dynamics to some of our traditional offerings. So I would say we keep learning about everything from a marketing point of view all the way through sales and customer success. We want to be a continual learning organization. But there's no doubt Brian brings a huge amount of experience and we're super happy to have him on board. Really excited about the future in this world as we continue to evolve.
Your next question comes from Jason Celino from KeyBanc. Please go ahead.
Hi, guys. This is Billy on for Jason Celino. Mike, it sounds like feedback and demand for Rovo has certainly been very positive, but of course, we're still in early days. I just wanted to get a pulse check on where you think we are in enterprise, agentic AI adoption and maybe what some of the barriers out there still are to that adoption curve.
Mike Cannon-Brookes: Thanks, Billy. Great, great question. Look, I think there -- we're very early days. There's no doubt that the easy answer is extremely early days in terms of agentic adoption in enterprises. I would say that a lot of them are experimenting, a lot of them are trying to learn and trying to understand fundamentally what the technology can do. They are understandably cautious with enterprise knowledge. We get a lot of questions about Rovo's commissioning and structure and how it works, who can access what content, how it all sort of strings together architecturally. I will say customers are very happy with how thoughtful we've been on all of those areas. But it's natural that this is a new area of technology. But they see clearly the upsides, the possibilities, but they're trying to be very careful and cautious about deployment and also working out where it actually adds ROI to their business. And that's a lot of what we spend time with customers is helping them to understand that. There is certainly a gap, I would say, between the awareness of AI and the usage of AI at the moment. Part of where we're trying to make sure we get across that gap is certainly with the Rovo inclusion with the various offerings that allows them to try Rovo to connect data and to learn. And as we mentioned, we have various ways to deliver that value, but also for Atlassian to gain from that value in time from higher editions, but also consumption-based pricing and credits and various things like that. So it's an evolving area, I suppose, I would say. Where I would say customers are most excited is the breadth of agents that we supply and also the familiarity of those agents in their interaction. So what's very different about Rovo is the way that an agent interacts inside the software, as another colleague in your company would. That makes it much more familiar to the usage pattern than some of the other approaches to AI, which has gotten pretty great reactions from customers. And the ability to iterate, as I said, between humans and agents and back-and-forth and obviously multiple times two, three, four loops to get a task done, that also resonates really strongly with our approach. It feels more like adding extra people to your team, as we would phrase it, than it does someone sort of replacing what you're doing, which creates huge positivity in our customer base. Generally, super early in our journey here, but huge level of customer excitement and our goal is to get that into as many hands as possible. I think we're making pretty good traction. As we said, we've gone from 1 million AI now to over 1.5 million AI now in a quarter and we'll continue to see that number rise. I think it's growing really strongly.
Thank you. That's all the questions we have time for today. I will now turn the call over to Mike for closing remarks.
Mike Cannon-Brookes: Thanks everyone for joining our call today. As always, appreciate the thoughtful questions and continued support. It's great to see a whole lot of you on the call at Team '25 just a couple of weeks ago in Anaheim. Have a kickass day and we'll talk to you in next quarter.