Adobe Inc. (ADBE) Earnings Call Transcript & Summary

All right. Good morning, good afternoon, everybody. Thanks for joining us. I'm Tyler Radke, Citi's Co-Head of U.S. Software. We're very excited to be doing a fireside chat with Adobe. We have Jonathan Vaas, who is an Adobe legend and veteran at the company. Jonathan, I really appreciate you making the time for us. I know this is busy. You just reported earnings, you have a trip to Miami coming up for Adobe MAX that many of us are excited about. I was looking through the registration numbers, and I think we have close to 200 people registered. I saw about 10 different time zones. So thanks, everyone, in Asia and Europe for listening.

Maybe we could just kick off high level for folks that -- everyone knows who Adobe is, but just given the generative AI topic is super important for everybody out there. At a high level, how do you, at Adobe, think about generative AI, there's obviously a debate in the investment community. Is this good or bad for software companies? Is it good or bad for Adobe? How are you encouraging folks to think about it and the opportunity that it unlocks for Adobe?

Sure. Well, first of all, thanks for having me. I know you and I have been trying to do something together for a while now. So I'm glad we could make this happen. And hopefully, we'll run into each other down in Miami. When I talk about generative AI, I'm going to first try to harness how Shantanu talks about it at his altitude, which I think is a really interesting point of view. And then a little bit about how David does in terms of the -- more specifically to how we're developing products. Right now, we're seeing this fascinating inflection, primarily in the investments companies are making in training AI models. If you look at kind of across AI infrastructure, where it's mostly -- where this is being -- where the investments are occurring today, there's a massive amount of money, I think, certainly not an exaggeration to say tens of billions that's going into training AI models. And Shantanu's point of view is, if those investments turn out to have been good investments, that the investing in compute has to shift from training to inferencing. And by inferencing, we mean sort of the way these models are used by end users to actually give them outputs, whether it's language or a picture that you're trying to get or a video or an assistant or an agent. There's all sorts of different applications. And so we're right now in the heart of the training phase, I think, for most companies. And Shantanu's point of view is, this is a time where Adobe can really shine by showing how -- of the big incumbent tech companies, we are one of the first and largest to bring a lot of these capabilities across our product services to where they're really being used by millions of people. And the inferencing you see, we've been sharing this number since we got in the game of generative AI and it accelerated in Q3 and past $12 billion. So his point of view is, we really want to drive that. We think -- if we are the company that's really driving the usage of all these capabilities, over the long term, we're going to be the winner in the category and have a lot of different ways that the monetization can ramp for generative AI. What David will often talk about when meeting with investors is, there's a lot of different ways people are training models and the things they are focusing on. And I think where this started was sort of, like, hey, can we create a tool -- I'll just call it a robot. Can we create robots in our computers that can just do the work for us or write the essay for us, take the MCAT for us or come up with a video clip for us. Adobe's point of view, we have a strong creator lens on the world, but we think these are tools that enable more people to sort of get into the game of creating or authoring, and really need to be controlled. And it turns out that it's hard technology problems to solve. It's one thing to make a model that just produces an output that looks really, really amazing. But like whatever we, as TasteMakers of the world, decide looks amazing. But it's harder to make it controllable and toolable and give people the ability to iterate on it. And so I think Adobe right now is building a technology lead compared to the other companies in generative AI and really focusing on that problem. And I think over time, we'll see that more and more people will get used to using these as tools for what they're trying to do, just like we use a lot of other tools in our digital world. And I'll say this and then we'll jump into the more specifics. When Adobe -- when I think about the technology companies that have changed the world and become the behemoths in the market, in the early days of these inflections in markets, they focused on usage, period. I think of Google. I think of Facebook. And I think the point of view is, if everyone in the world uses this when they need to search for something or when they want to connect with friends or when they just have idle time on the train and they are on their phone, like if we can drive usage -- and we think of usage in billions, the monetization -- even Amazon, in some ways, fits this paradigm. The monetization opportunities are going to ramp over time. And of course, Adobe has a core business that's been growing and very, very profitable. But when you think about this new market of more casual creators, generative AI and even Adobe Express, the usage -- we think there's an opportunity for us to be the company that's used by billions of people. And then if we can nail that, there's so many different ways that, that will pay off on the P&L.

Awesome. Well, it's a great overview. And obviously, a lot of specific questions, and we've gotten a lot from the audience. But maybe sticking on just the AI monetization, and then we'll go into kind of the post-quarter questions, because I know we've received a lot of follow-ups on those. But as we think about that generation you talked about, accelerating momentum of images being generated, how do you think about when to start monetizing that, starting to charge for that usage? Is that something that you're thinking about for next year? How do you sort of approach that monetization conversation?

Yes. Yes. So we think of it as an economy where we kind of control the inflation and the usage. It started up -- very quickly, we integrated 2 capabilities in Photoshop, Generative Expand and Generative Fill, and like half of Photoshop users, within a very, very short period of time, just started using them in their normal workflows. And we saw sort of this ramp in generations, and then it was kind of stable. And what we said is, as we bring more GenAI capabilities across the suite of apps, you'll see the inflation. So it started -- one of the reasons it really accelerated in Q3, and I do think over time we'll probably get more precise about how we share some of these product-related usage KPIs. But in Q3, one reason it inflated was we brought it to Lightroom. At Lightroom Mobile, now people can very easily, like, circle somebody in the back of their photo at the theme park that they didn't want to be there and just remove them. And that really ramped generations. When we bring video capabilities to our products, that will really ramp the inflation of usage. So right now, it's sort of all about like how can we get more and more people hooked. I often use a coffee analogy of we're giving a lot of free drip coffee in all of these surfaces and getting everybody hooked and they're all loving. And eventually, they will sort of self-select into -- I'm ready to start drinking the fancy espresso drinks and they'll be lining up to buy the stuff because it's so cool and they're excited about it. So I think there are ways we're monetizing it today. We can go more into those. When advanced things like video come out, I think there will be a closer matching in time of how we monetize those SKUs with the video usage. But it's something that we can -- because we have such a good control on the cost of the inferencing, I think we can afford to be generous today to really get all these people hooked on the usage. And then over time, we see a bunch of different vectors of monetization.

Yes. Yes, maybe since you brought it up, like how do you think about those monetization vectors over time? And I know you're not specifically giving guidance or actions you're going to take next quarter or next year, but just philosophically between credits, higher-priced SKUs...

I think -- so I would start with new products and new SKUs, like the Firefly web app, like Express and like the AI Assistant, which is an add-on for Acrobat. We also have new SKUs in the Enterprise, and I'll talk about that in a minute. So there's new products and SKUs. When we add a lot of value to our existing products, from time to time, we might have a price increase that comes with the value we've added. Then there's going to be upsells in terms of, over time, the people that self-select into using a lot of these capabilities might need to buy a higher-tiered subscription that gives them as much as they need of the generative credits across applications. In the Enterprise, there's both sort of account expansion as enterprises step-up to the more sophisticated plans that come with all these GenAI features in their seats. But then there's sort of transformative offerings like GenStudio or Firefly Services or the ability to customize their own AI model. I didn't mention it, but folks probably already know, our AI is trained in a very commercially safe way, only on licensed assets. So then they know that they're not going to accidentally infringe somebody else's trademark, but they can bring their own license and copyright it and trademark assets into it and use it. So all of those things in the Enterprise are significant upsell of big value. That's going to save them money on their content supply chain and help them do things in a lot more streamlined fashion. So I think over time, the business becomes more enterprise-y as well. But those are a bunch of different ways we're thinking about it today.

Got it. Very helpful. And sticking on the price side, obviously, that's been very topical for Adobe following the actions just about a year ago that you announced in terms of kind of that initial way of monetizing the value that customers are getting with Firefly. I guess, where are we at in terms of -- first question is, where are we at in terms of the rollout of those pricing actions that have been announced in terms of those impacting the installed base? I think the last data point was around 60%, but would love to get an update there, and how you're thinking about additional SKUs that maybe haven't been announced yet.

Sure. Yes. Yes. And when we think about the growth drivers of the business, the biggest one is new customers coming in. But then over time, there's kind of 2 different motions that we want people to ultimately spend more with Adobe. And it's one, them opting into the upsell motions or from time to time, the prices have gone up. And that's been -- and there's been a pricing layer to the growth probably every year since 2018. In this particular wave you're asking about, we started when we integrated Firefly throughout Creative Cloud Apps in November of 2023, and it was a slower rollout. We try to feather in and out the pricing-led growth, so it's fairly smooth over time and not jarring to our customers either. But it started with not an entire world, it was, I think, North America and Europe primarily. And then through mid-2024, we expanded some of those pricing actions to the rest of the world in education. So the prices that we've announced are contributing to Adobe's growth in 2024, for the last month of 2023 even, and will flow well into 2025. So you'll have pricing-related growth sort of as that third leg after new subscriptions and upsell motions that benefit both 2024 and 2025. We often say pricing is a journey, and it's kind of core to how we operate the business. So we're also looking at how much we're driving and motivating upsell motions and what the new customer additions look like. But over time, I think for the next 5, 10 years, we would expect there to be a pricing layer to the growth and that would likely be the third ranking driver.

Got it. Okay. And this year, I think, was kind of interesting just in terms of the quarterly dynamics on pricing, because while you had this November rollout, I believe the comments were that pricing was actually a net headwind year-over-year, if you're looking at Creative Cloud net new ARR growth in the first half of the year, but actually by the second half that reverses to neutral to a slight tailwind. I guess could you run through how to sort of think about pricing between Q3, Q4 versus the year ago periods? And then how to think about pricing in FY '25 relative to FY '24?

Sure. Sure. And one thing I've learned in having a lot of conversations about pricing, even in terms of the company, sometimes folks get confused that they hear headwind and tailwind. A lot of investors think a pricing headwind is if like the Big Mac has gotten too expensive and McDonald's will need to lower the price. It's deflationary [indiscernible] headwind. So I think as we've learned more about how people hear this, maybe the better way to describe it is, pricing is a tailwind to the business or a contributor to the growth. But if we zoom in on the close altitude of looking at the net new ARR, there might be tougher or easier comps depending on how much pricing contributed in the year ago. 2023, we had strong contribution from an Acrobat Pro price increase that created tougher comps for the first half of this year. And then we had easier comps on the back half of this year. And so if you're looking just at those little swings in net -- and we're not talking about hundreds of millions, we're may be talking about tougher or easier comps by tens of millions of net new ARR, but a tougher comp first half of the year, easier comps second half of the year. As we would look out to 2025, I would think that dynamic would reverse based on the changes we've announced today. Of course, we want to, again, drive more growth next year with some of those upsell motions. And if pricing is a journey, there's likely other products and SKUs that might -- where we might have added a lot of value and, at some point in time, make additional pricing changes.

Okay. So it sounds like second half better from a compare perspective on pricing than first half. And would you kind of think of Q3 and Q4 seeing about the same? Or is one of those quarters have slightly more tailwind?

Yes. Yes. Q4 would have a little bit more tailwind than Q3. Not dramatically, but a little bit more, yes.

Okay. And so then next year, you're not expecting any kind of net pricing headwinds just given that you won't have any of these like compares from Acrobat Pro and potentially there could be additional price rollouts to other SKUs that you haven't done. Is that the right way to think about next year?

Yes, I think that's right. And a lot of folks missed that the fact that Acrobat Pro is also one of the flagship Creative Apps. It's been in there since the dawn of Creative Cloud. Actually, the first time we made Acrobat a subscription was in the Creative business, and then we did it for the Document Cloud business. So a lot of folks, I think, missed that in their modeling in 2023. And so I don't think there's anything that should be unexpected as we look at 2025.

Right, right. Yes. I think that, that was -- when people saw the Acrobat increase, they all often just defaulted to thinking about Document Cloud, but it is a bigger chunk than I think even we realized is Creative Cloud.

While we're talking about it, my hope over time, I should say, is to maybe elevate the discussion to really the growth rate on our ARR as we look at -- is it flat? Is it accelerating? Is it decelerating? How are we growing? Right now, at the level of discussion of scrutiny on how much net new happens every single quarter, you'll see on Document Cloud, the pricing change for Acrobat rolled through our reseller channel from mid-summer '23 to mid-summer '24. And so we had a growth contributor that now is going to be a pretty small contributor in Q4, and that's something to think about as you look at the Document Cloud. The implied guide for that business in Q4 is that it doesn't have the pricing contributor that last year's Q4 did.

Right, right. And that is a good point. I think we did want to sort of ask you about the Document Cloud growth because that has been very robust, to your point. But as you think about that digital media guide for Q4, which we can get into the puts and takes and some of the reaction certainly on the stock post the quarter. But as we think about the comps, should we be kind of seeing more outsized growth in Creative and then declines in net new for document just because of those pricing actions you referenced. Any other finer point? Because I think if you imply some growth in Creative Cloud, you're looking at about a mid-teens decline in net new ARR in Document Cloud. So just curious if that's the right way to think about it.

Yes. And we did share on our call as well as at the Q2 call, an expectation that Creative net new ARR would return to growth in Q3, which it did. And that, that would again be true for Q4. And that implies, like you said, that the remainder of Document Cloud would be a little bit down. Now we'll see where we land, but I'll say this about the Document business. I think even notwithstanding some of the pricing-related growth that we saw earlier this year and even new growth drivers like the AI Assistant, that business is performing extremely well, primarily seat-driven growth, and I think, even without those drivers, would be Adobe's fastest-growing business. So I couldn't be more bullish on Adobe's Document business. And I hope investors start to spend more time on it, because PDF is a pretty powerful file format and Adobe is the very clear global leader in that franchise.

Yes. No, it's been a very, very strong growth driver over the most recent years. Going back to what we're talking about earlier around Firefly and some of the consumption services. One of the ways that you are monetizing, particularly at the high end, is around Firefly Services and Enterprise deals, you have GenStudio. But specifically around Firefly Services, can you help us understand, does that show up in both DX and digital media if it's attached to a DX? And then how do you sort of think about consumption as it relates to ARR?

Sure. And we'll talk more at MAX about sort of GenStudio and that bit. GenStudio is this exciting realization of a decade-long vision of bringing content creation truly into the CXM business of how you deliver and measure and optimize and monetize all that content. GenStudio does include Creative products and Creative SKUs as well as DX. Firefly Services is one of them, but Firefly Services is purely creative. So as the Firefly Services component of one of these big deals -- I mean, first of all, the rev rec team will look at the deal, and depending on how that enterprise chose what pieces they want, they would allocate ARR and revenue accordingly. But Firefly Services all falls within the Creative business. And when a company buys it, they're committing essentially to a volume of API calls. So it's a subscription sort of a metric where they're saying we're going to use -- it's probably a band of this amount of API calls. If they use less, they're still committed to it. If they use more, our likely playbook would be to go to them and recut the deal at a higher-tiered subscription because we want to be driving ARR. There could be, theoretically in our pricing model, overage sort of revenue, which you wouldn't see in ARR, but we would much prefer to go back to them and have it be a friendlier deal to them, where it's just a known commitment. And for us, then it shows up in recurring revenue, which is our primary KPI. If I think about like a GenStudio transformative enterprise deal, like holistically, I think 60-40 is a good way to think about how it might typically be allocated with 60% of the value likely being DX, 40% of the value likely being Digital Media.

Got it. That's a helpful way of framing it, and I'm sure we'll learn a lot more in Miami next month. But we've had a few follow-up questions come in just around some of the topics we were hitting on earlier. And maybe to sort of combine a few questions here into one. But let's just go back to last quarter. And maybe just start with -- on one hand, look, you put up a very healthy beat again in the quarter. You raised the full year Digital Media net new ARR, which is positive. Obviously, the stock reaction was not positive. So what do you think is -- I guess, what do you think folks are missing? Or is there anything that you think you want to get out there that maybe didn't get appreciated? And then we can dive into some of the specific questions.

Yes, so I got to start with how encouraged and excited we were about the Q3 numbers. Our last record for net new business in Digital Media was $464 million. And to put up a 5 handle on a business that's usually a little seasonally down in the summer. So our summer quarter, it's June, July, August, a lot of people are on vacation. So we were really excited about a significant record. A $40 million record was a pretty big breakout from trends. And so -- I think after Q2, we had raised the year to $1.95 billion on net new ARR, ,and we took half of the Q3 beat effectively and added it to the annual guide with our implied Q4 guide to $1.97 billion. So we feel like it's an execution story really of us doing what we said we were going to do. And I think -- I mentioned earlier, I would really love to sort of elevate the level of discussion on the business to really talking about the growth rate of the ARR and where we see that going. And little swings between our quarters can actually hit right now, because we're talking about this very close altitude of how much net new hits each quarter really swing. So the last investor I talked to this morning said, yes, your Q4 guide was $20 million below expectations, and that had a 500x impact on your market cap. Now, I didn't check his math. I don't know if that's true. That was something he said off the cuff. But so I think a lot of people are really extrapolating from these near-term trend lines to long-term trajectory. And I would go back to where I started. On the long-term opportunity here, we really believe we're at the front end of the technology inflection that impacts billions over time. I think in 10 years, every student in the world, every knowledge worker in the world will start with generative AI to get every image for their school paper or for their lemonade stand or for their Board presentation, whereas we all grew up in a world where we were using Google Image search. So I think it's that big of an opportunity. I love where we're positioned. But the reality of how we've been talking about the business is the scrutiny on how much it's each quarter is sort of where the discussion is. So I would love to expand the aperture.

Yes. Okay. And I guess just as you think about that Q4 guide being a bit sub-seasonal, but obviously, there's offsets to that being Cyber Monday timing, also some of the favorable timing that you saw in terms of some larger deals in Q3. I guess was there anything that changed about the -- you talked about record growth in Q3, would you say that positive description of like record growth, is that sort of true about the pipeline? Or anything that leads you to tell investors to be a little bit more cautious as they're thinking about next year?

Sure. So first of all, in Q4, I think Shantanu said on the live call, we hope to have an all-time record in Q4. I think if we do that, we'll hit $2 billion for our first time for a net new ARR in a year, which would be exciting. And so -- I think Adobe has a good track record of delivering on our targets, which we take very seriously. And hopefully, we'll do that. As we looked at -- and we're not unaware that investors really do hyper-scrutinize the trend lines between each quarter and the next quarter's guide and the next quarter's actuals. And when we looked at it, I mean, we expected a great Q3. But right on the last couple of days, there were, I think, 3 Document Cloud deals, big enterprise deals or public sector. We kind of tend to bundle together enterprise and public sector. That to good execution, the customers were like, we're ready to go, we're ready to sign. And of course, then we put them through and we signed them. We had profiled those as Q4 deals. And when we looked at the trend lines compared to last year, we realized that just those 3 deals, we mentioned those strong enterprise execution in the prepared remarks for Doc Cloud. But those 3 deals, in combination with Cyber Monday -- like once every 5 or 6 years, that Cyber Monday week, which is one of our biggest weeks of the year, falls into Q1. And that holistically more than explains the kind of $20 million deviation from last year's trend line.

Sure, sure. Right, right. And to be clear, though, the Cyber Monday guide was -- those dynamics were already incorporated in the initial full year guidance. It's more just a comment Q3 versus Q4 seasonality?

Exactly. Yes. We know the timing of Cyber Monday for the next 20 years. I mean, that's known. It's sort of just a way of explaining like, yes, this is why this year's trend lines might be slightly different from last year. And of course, we'll see how we execute and where we land.

Yes. Okay. So one of the other dynamics of the quarter beyond the ARR noisiness, which I'm sure you're sick of talking about at this point. But just on revenue, I think there were some questions on that, too. To your point, record Q3, favorable timing of some large deals and full year net new Digital Media ARR was increased. So mechanically, you would think those would be more positive for revenue in Digital Media. But the Digital Media revenue guide was basically held flat and there was a big beat in the quarter. So Q4 Digital Media revenue came down. Can you walk us through why that happened? Because it's not really intuitive, I think, for a lot of investors. If you have better ARR and timing, usually that would translate to better revenue.

Yes. And this theme is something I've been learning a lot about ever since I took on this role in COVID. So we think -- we look really closely and internally, we scrutinize what I would call the revenue yield from ARR or, frankly, any bookings-based metric will have -- there will be some delta to the yield. And sometimes, for example, if you have a huge tailwind from FX, the yield might be greater than expected. But there's always a leakage component. And when you look at leakage, it's fairly complex FP&A stuff, but it's the linearity of the timing of the bookings that come in versus linearity of your attrition base is one of the things. But I would say, in choppier macro times, you see more leakage. And like an example of something that would stay in your book of business, but that would not come in terms of revenue would be, I think of when I was a Comcast subscriber and they've increased my price, and I call and I say like, "Hey, I'm going to cancel, I don't want to go up to the higher price." Then Comcast would say, "Hey, like hold on. We'll give you a few months free and we'll renew you at a new discount." And then I'm happy. I'd stay in the book of business, but there's a few months that they didn't get revenue from me. And so Adobe has programs of running discounts and offers to subscribers or there's things with enterprises that might result in leakage. But that's one of the things. Perpetual is another thing. So if you look at the trends, you see it on our income statement, it's the product revenue line item. We are -- there's not much perpetual left in our business. It's mostly Acrobat, but that's been on a declining trend. It's on strategy that impacts the Q4. And then FX where it is compared to our expectations when we set the initial guide based on the forward rates we used, it has -- so all 3 of those things kind of compound to being a year where ARR was a beat, but our revenue for the segment has kind of stayed in line in the range.

Got it. Okay. And the -- I guess on FX, the dollar has weakened, I guess, over the last 90 days. But should we think about it -- I guess, was the revenue guide kind of not updated for FX last quarter? I guess just thinking about it from a 90-day ago perspective, it's not as intuitive as to why FX would be a headwind because it's been a bit of an incremental tailwind for others out there?

Yes. And it's really the dollar where it stands compared to our expectations when we set the initial guide coming into the year based on the forward rate that's being used is stronger than we initially had forecast. Now we have a sophisticated hedging program to help protect us on the downside. But all things considered, I would say the FX component to this discussion hasn't been material. But if you look at the range we guide to Digital Media revenue, it's a pretty tight range. So we've -- despite some ARR outperformance, we're still kind of in that original range we set.

Okay. Okay. Makes sense. Right. And then I guess as we think about the margins, which the first time we're bringing up profitability 35 minutes into the call, but it's one of the great things about Adobe is the margin structure. Headcount, I believe, was down a little bit sequentially, and I think there's some seasonal timing and whatnot there. But how do you kind of think about the margin expansion picture from here. You mentioned earlier, you have some really great efficiency in terms of driving some of these GPUs for the images. But how should investors think about kind of margin expansion from here?

Yes. And I'll try -- I'll hit -- well, first, very quickly on the headcount. You see a bump every year in Q2, that's our summer interns. And then you usually see it normalize in Q3, and that's because they've gone back to school. So that's all you see in sort of that dynamic. But I'll hone Dan Durn and try to hit margins and how we think about it on the gross margin line and sort of then the operating margin line. On gross, our intention with bringing all of these cloud capabilities to Digital Media and inferencing is to spend a little bit more on the hosted side for Digital Media. I think it's -- as you actually look at the trend lines of Digital Media COGS over the last 2 years, like the downward pressure is -- you have to squint pretty close to even see it. But we're pleased that we are seeing real inferencing and we are bringing more products to web browsers, but we're also pleased that the technology keeps gaining in efficiency. And I think over time, we even believe that a lot of inferencing will move to the edge of desktop and even mobile devices. But over time, like let's talk about next year or 2, I think you'll see a very slight trend line on COGS going up in the Digital Media business. And now I'll go over to Digital Experience. And that gross margin is expanding. It broke through 70% for the first time in Q2, even higher in Q3. And I think internal to the company, there's a real pride in our Digital Experience products becoming more cloud efficient, scaling. And the point of view is, they effectively should fund through efficiencies and scale what Digital Media needs to spend on inferencing to hold total Adobe gross margins where they are, which is world-class. If we go into the operating line item, today, our biggest investment is training the models. That hits R&D. Pretty much the places we're investing on the operating expenses are the talent that's building AI as well as things like our data platform and Express and then training the models, which is a lot of GPU expense. Other than that, we feel like through prioritization and being really disciplined operators, we should be able to grow, for example, G&A slower than revenue growth. And once all of these new web-based capabilities are in market, there's also going to be sales and marketing efficiencies. Usage and retention goes up, we know that from Photoshop last year. So there's a lot of gain in the operating line item from what you can just call product-led growth. So over time, we think there's plenty of margin expansion ahead of us.

Got it. That's helpful in the framing kind of between the different businesses and everything. I guess an investor question that came in here is, I guess, sort of related to margins in a way, but just capital allocation, like how are you thinking about buyback versus M&A. Obviously, there's been some large M&A that you have announced with Figma that ultimately got blocked. But do you think you're -- just given the momentum that you now have with Firefly, like you're leaning more towards buybacks here versus larger M&A?

We always start conversations on capital allocation with -- it all starts with a growth mindset. We're going to invest to grow the business. Primarily that's an organic. And from time to time, we've brought in some inorganic as well. I think our focus on growth in the near to midterm is going to have a heavy focus on the organic, just because there's so many growth drivers in document business. AI, we just talked about is sort of a once-in-a-generation inflection in Digital Media. And now with the monetary policy finally becoming more accommodating, maybe we'll start to see a dynamic in the Enterprise, where enterprise customers start to invest more to drive growth, which should be a good thing for Digital Experience where we built that platform from scratch. So I think it's mostly organic growth as we look at, and Adobe over time has done a good job with smaller tech tuck-ins, talent deals. Even Photoshop was an acquisition of a 2-person company with a great product. So I think we'll probably get back to our heritage of looking at like smaller deals that can be additive to the business, drive organic growth. And then it's return the excess cash, which is a good source of profitability to Adobe. There's a lot of cash. We did $2.5 billion of share buyback the last 2 quarters. I expect us to file our 10-Q sometime in probably the next week, and then you'll see what we did for Q3. But there's a strong point of view with the growth mindset that share buyback is the best value and the best way to return capital to the stockholders. We've driven the share count significantly down over the past decade. I think more than 10%, which is good to see that it's not just offsetting dilution, but truly driving down share count.

Yes. Makes sense. Okay. One question that we got from the audience. Just going back to thinking about next year's growth, is there a way to sort of -- and kind of in the context of product cycles and everything. But is there a way to stack rank the impact or maybe call out what are the biggest contributors to new growth from the new products as we think about Firefly Services, video, Acrobat AI, Express, or Express for Business and kind of consumption credits and others. Or maybe there's some that we're not thinking of, but anything that you'd sort of put at the top of that list?

Yes, if -- so any time people ask me about growth drivers, I can't help myself but to say, in the near term, certainly for 2025, my expectation is that Adobe's large-scaled flagship businesses drive most of the growth just because of the scale, and they're all growth businesses. So I think Creative Cloud -- all Apps and Photoshop and Acrobat and Adobe Experience Manager and Adobe Experience Platform or DX, those will be the biggest sources of growth. The newer businesses, I think of as accelerants to growth, that are still working on getting to scale, right? And so I think if I had to sort of -- just based on my -- what I see in my own excitement about them, rank what's coming for 2025, I think enterprises making these transformative commitments to some of these new things like GenStudio and Firefly Services is very high for me. Because these are -- if you get one enterprise that's willing to pay you $20 million a year, there's a lot of $10 a month subscriptions you have to sell to get to $20 million. So it's -- the transformative deals really have -- they move the needle on your recurring revenue sources. Based on how quickly we see people adopting Acrobat AI Assistant in our checkout flows, that's one that's up there for me. And then just Express, which is scaling up. It's still primarily a usage-based game for Adobe as we want just massive audience of people adopting it. But I think on the monetization, we see people starting to buy in emerging geographies where we struggled historically to break through. And I think we had our best quarter in Q3 for Express monetization. So Express is going to come. And then I think there's going to be a lot of new upsell SKUs for things like video that should be exciting.

Got it. And as you think about the Express monetization and some of the momentum, I think you talked about goals earlier of getting to 1 billion plus users and everything like that. How do you sort of think about the competitive landscape today? Obviously, Adobe historically has dominated the high end of the design category, but you do have design companies out there with real scale and growth, Canva and then there's a number of others that are really adopting generative AI. But as you look internally, what are some of the metrics that you track that gives you conviction that you still kind of have that competitive advantage that's sustainable?

Yes. First of all, I do think -- I think of that space as a new market for Adobe and just an explosive new market. I think most of the people in it never paid Adobe for Creative Services in the past. I think of them as mostly Google Image Search copy and pasters for the past 20 years or even Clip Art users in the Microsoft suite of products and other places. So I think we're now -- we live in a world where people do want to create content and short-form videos that stand out, and it's a pretty big explosive market. We saw that market evolve a lot over the past 5, 10 years through templates, editable templates. And Adobe is in that game, too, with Express. But we think -- first of all, it's going to inflect and it's going to be AI to create design templates that then you can then change, which is much faster and more expansive. So we think we have an advantage in reframing it as an AI-first conversation, and that's where our technology shines better in the editability. I think we have such a huge footprint with the PDF users that we will integrate increasingly creativity into PDF and that's a big opportunity for Adobe. And then in terms of competition, I think there's 2 really focus areas where we feel like we have some competitive advantage. One is from the enterprise down. The sorts of problems we're solving for enterprises and automating their content supply chain and driving true personalization, really complex stuff that not really any other single company can offer. So there's a value proposition in the enterprise and already a relationship and a trust that moving from enterprise down into mid-market and SMB is a direction you'll see us kind of go after the market opportunity. And then from education up, and this is how Adobe built the Illustrator and Photoshop business back in the '90s when I first learned them was back then, education primarily meant university, but now it might mean third graders. I have a third grader, which is why I think of that. And so we're getting Express in the hands of teachers and provisioned for free for students, and there's a lot. So we're kind of moving education up and then enterprise down, and that's kind of the commercial focus right now.

Got it. Going back to the enterprise side. Just a couple of questions there. I mean, number one, the most recent, I think, price increases that you announced in November of last year -- I guess you announced it a little bit before that. But largely, those were not applied to enterprise, I don't think, or we didn't see at least publicly kind of the ETLA price increases or anything like that. But how do you think about the AI-related price increases to the enterprise SKU? And then just pricing generally within ETLAs, like how much more runway do you think you have? Just what I imagine is like very attractive volume discounts to incentivize adoption. But just how do you kind of think about the growth potential of ETLAs from pricing?

Yes. First of all, when we add value to the products and have a pricing change, it's for the digital customers, it's primarily a 1-year dynamic of when they come up for renewal, they step up. For enterprises, they tend to be on 3-year or 4-year deals. So it will be a longer tail of discussions. And there are 2 things going on with enterprises. So it's not as simple as just a monthly credit card change. For an enterprise, it's a discussion. There is an upsell motion for the seats. So think of enterprise as 2 things. P times Q, what's the price per how many seats they have for things like Creative Cloud. And there is an upsell discussion when they move to -- I think it's like enterprise CC4 is the new version that has all the Firefly entitlements. So we see an upsell moat. We both see seat expansion and a price expansion within the enterprise. But we would much rather talk to our enterprise customers about buying a much bigger set of goods from Adobe versus just the prices increases. So at renewal, we'll also look for product expansion within the enterprise. But then there's these big transformative value propositions like GenStudio or Firefly services and custom models. And that's not seat driven, that's more -- this is -- you're going to pay us $5 million a year for this expected set of hosted storage of assets in your content hub and API calls to use Firefly for variant creation. So there's a lot of that. And I think we'll have tailwinds for the near to midterm with enterprises where -- I said earlier, I think the creative business becomes more "enterprise-y". I know that's not a word, but I really do think we can save them so much money from traditional, like video and photo shoots and content production cycles of what we can automate for them, where I think there's a really big opportunity?

Got it. Got it. Okay. A couple of other questions that have come in. I guess going back to some things that you mentioned. The first one is you talked about kind of goals of trying to elevate the discussion from net new ARR to more overall ARR growth. Like what do you mean by that? Are you thinking of maybe guiding any different? And I guess if you report just ARR, you can kind of back into net new, but just any thoughts on that?

Yes. like when I look at -- and you've increasingly seen this in our prepared remarks, we'll mention how ending ARR grew year-over-year, that's a constant currency calculation. But I do think that's the right way to think -- we get in these discussions, and they're like, "Well, wait, so was it $5 million or $10 million that fell into Q3 from Q4?" And I say like, first of all, this isn't the altitude we should be talking about the business on. I think our guide for 2024 implied the ARR would grow around 12% to 12.5% year-over-year, and I think our actuals have grown around 13% year-over-year. And I think that's sort of the right way to think about it. You certainly want every business to have the incentive of getting every deal done as soon as you can, not sort of like, oh, wait, that should be a Q4 deal. And so I think that's the right attitude to be talking about it. And I think you'll continue to hear more of that focus from Adobe, is just what's the growth rate of that book of business. And of course, just like our peers that report on CRPO, you can always calculate from the numbers, how much they've added every quarter. But really, the focus should be on the growth.

Yes. Okay. Very helpful. And back to your coffee and espresso analogy, which hopefully you have a nice espresso next to you there. What do you need to see to stop selling the coffee and start charging for espressos? And I guess, is that something you're expecting to do next year? Or what do you need to see?

Yes. Yes. For us, a lot of the espresso is in development. We teased our video models. I think there's some premium models that are coming out, the ability to auto-dub video into different languages with the audio and video matching, the end-to-end video suite, what we're going to do in 3D. Also, the imaging models are becoming more and more complex, and there's just going to be more buttons in our different apps that trigger a generative credit. So I think the -- once we have the full menu that's out there, that in and of itself, we're going to see people starting to self-select into these higher related categories. And we talked about pushing the distribution curve of usage to the right on this chart, where we can see the long tail of users that are going really, really, really deep in these capabilities. And we want to sort of delight them so much and get so many people used to this that the whole distribution curve of who uses a few, who is using a lot and who is using the most is to push to the right to where that opportunity to start up-charging in different ways for -- there might be a video-only SKU, there might be the Creative Cloud All Apps -- more Generative credit SKU. But I think we'll see in the data and the usage once we have that full menu in market when the right times are to pull those levers.

Got it. Okay. That makes sense. Probably more to get announced next month at MAX, too, I imagine.

Yes.

And I guess on the topic of MAX and not that you would ever want to front-run Shantanu or David in terms of the exciting things they're going to announce. But what should investors expect to hear from Adobe at MAX? I know you are doing some investor sessions, I believe, more Q&A focused, but to just kind of level set for folks because we get the question of what should I expect at MAX. So give us a little bit of a preview of what's...

Yes. I think -- first of all, I think this will be a big one in terms of the demand we're seeing from customers. The pandemic now feels like a long time ago. There's a ton of demand for people to come together. And I think there will be a heavy focus on how we're building AI for creators. And like -- kind of like where I started, it's really a tool for ideation and iteration, and I think that's going to be front and center across more and more media types. And Adobe, we now are innovating so quickly that we don't save up innovation for the whole year and announce it at MAX. In fact, we've had these regional MAX. We had one in London not too long ago with new announcements. So I think you'll even see announcements between now and MAX, but it's just going to be constantly teaching the community what we're doing, what's coming and how they can go deeper in it. And then yes, we're going to have a Q&A session with the investors. As a reminder for folks, this year, our big investor event or financial analysts event was in March at Summit, and that will be the new normal. So when you get all of the additional disaggregations and disclosures from the company, it's after we've completed the fiscal year. And we used to do a Q&A in Summit. Now we'll do the Q&A at MAX.

Got it. So we shouldn't expect any formal targets.

No, Shantanu shared that on our Q3 call. You'll get our 2025 targets in December with earnings. But if there's any changes in how we are starting to think about KPIs and targeting the business, we would share that at MAX, so there was an opportunity for folks to ask questions and discuss before December.

Okay. Okay. Great. Awesome. Well, I know we're running up against time, but maybe I'll turn it back over to you, if there was anything you wanted to hit on, and any message that you want to get out to investors just as they're thinking about Adobe into MAX and the rest of the year.

Sure. I'll say, keep leaning in and paying attention to what Adobe is sharing on the usage of all these new capabilities, because I do think that is the best leading indicator of what the opportunity is. And so again, we've shared now since we got in the Firefly game, it's over 12 billion generations. But the pace of that significantly accelerated in Q3. And I think that's exciting. And to go back to where I started, we have a fundamental belief that if a company becomes the platform that potentially billions of people are using, the opportunities for growth that are going to stem from that are huge. And I think Adobe is very encouraged by the way we're seeing -- I mean, we've been in the business for 30 years of creativity and seeing what typical trend lines look like of teaching people to use new tools. And GenAI is like nothing we've ever seen in terms of you put something in the tools and suddenly half of the users, like that quarter, have adopted it and are utilizing it. So I think you'll hear Adobe increasingly start to -- like kind of firm up some of these usage-based KPIs and share and why that's excited, because we want investors to be excited about that as well. And if we nail that, the opportunity in front of us is massive.

Okay. Yes. So I guess we should maybe think about other KPIs being introduced around images generated, et cetera?

Yes. Things like images generated -- one of our big KPIs for Express is total exports, because that implies, if somebody exported something, they completed a project and they used it for something. That's something we talk about a lot. Of course, MAU, we talk -- I think we've shared in the past that MAU for Digital Media was free and paid over $650 million, it's grown since then. So we have aspirations of MAU breaking through $1 billion. And I think those are important leading indicators on the business as well.

Awesome. Well, I think we were right at the top of the hour. This is a quick hour, we covered a lot of ground, everything in the weeds to high level. Jonathan, thanks for spending the time with us. And thank you, everyone, for dialing in. And hopefully, we'll see everybody in Miami next month.

My pleasure. Thanks, everyone.