DBX - NASDAQ

Thank you for standing by, and welcome to the Dropbox Fourth Quarter 2025 Earnings Conference Call. [Operator Instructions] As a reminder, today's program is being recorded. And now I'd like to introduce your host for today's program, Peter Stabler, Head of Investor Relations. Please go ahead, sir.

Good afternoon, and welcome to Dropbox's Fourth Quarter 2025 Earnings Call. As a reminder, we will discuss non-GAAP financial measures on this call. Definitions and reconciliations between our GAAP and non-GAAP results can be found in our earnings release and our earnings presentation posted on our IR website at investors.dropbox.com. We will also make forward-looking statements on this call, including statements about our future outlook for our first quarter and fiscal year 2026 as well as our expectations regarding our business, assets, strategies and the macroeconomic environment. Such statements are subject to known and unknown risks and uncertainties that could cause actual results to differ materially from those described. Many of those risks and uncertainties are described in our SEC filings, including our most recent and forthcoming reports on Form 10-K. Forward-looking statements represent our beliefs and assumptions only as of the date such statements are made. We disclaim any obligation to update any forward-looking statements, except as required by law. I will now turn the call over to Dropbox's CEO and Co-Founder, Drew Houston.

Thanks, Peter, and good afternoon, everyone. Welcome to our Q4 2025 earnings call. Joining me today is Ross Tennenbaum, our Chief Financial Officer, who joined Dropbox in December. I'll start with a recap of the quarter and how we closed out 2025, and then I'll talk about how we're thinking about the business and our priorities going forward. Ross will then walk through our financial results and outlook. We closed out 2025 on a strong note. Fourth quarter revenue came in above the high end of our guidance and excluding the impact of our FormSwift wind down, constant currency revenue was flat for the quarter and the full year, which is a better-than-expected outcome. We also made meaningful progress on efficiency. Margin performance in Q4 exceeded our expectations, and we generated over $1 billion of unlevered free cash flow. At the same time, through our share repurchase program, we reduced diluted share count by more than 50 million shares in 2025. Taken together, Q4 was a good reflection of what we're working to do consistently, which is execute well, deliver against our plans and steadily improve the underlying trajectory of the business. And in 2025, our priorities were focused on strengthening our core business and scaling Dash in pursuit of returning to revenue growth. We're still executing on these objectives but now have proof points that these changes are starting to work. Coming into last year, our Core FSS business had strong fundamentals and scale, but execution velocity, product experience and our go-to-market motion had not kept pace with customer expectations. So in late 2024 and early '25, we did a leadership reset in Core FSS, bringing in a new general manager and rebuilding key leadership across product, engineering and go-to-market. Since then, we've made significant improvements in how decisions get made, how we prioritize customers and how we deliver value. And we're beginning to see positive signals. The team first focused on improving funnel quality, pricing and packaging, product fundamentals and retention drivers. And as a result, the individuals business saw steady growth across 2025. That matters because it demonstrates that the core product can still respond to focused innovation and better retention and growth are achievable with the right execution. And so our objective for 2026 is to maintain our momentum with the individuals business and return teams to positive net license growth. Work already underway includes simplified pricing and packaging, higher intent trials, reduced onboarding and admin friction and a sharper focus on retention. Some early tests in Q4 showed some promising signs, including improved teams trial conversion rates and higher first week engagement, and we began rolling these changes out more broadly in Q1. But in short, we're not simply maintaining our Core FSS business. Our goal is to bend the curve. In 2025, we delivered proof points and 2026 is about scaling that momentum. Our next focus area is what we call Dash and Dropbox, which represents the most important evolution of the Core FSS experience in years. Dash and Dropbox provides an AI intelligence layer directly inside our customers' everyday workflows with minimal setup and immediate relevance. In Q4, we launched embedded Dash capabilities inside our Teams plans, including semantic search, chat and stacks organization and sharing, and we're rolling it out in phases to eligible Dropbox Teams customers. We're seeing solid early engagement among the initial Dash and Dropbox cohorts. In Q4, over half of these active users returned multiple days per week, which is evidence that Dash is providing value and becoming a part of user workflows. And based on these results, we've begun scaling up our rollout to additional customer cohorts. Dash and Dropbox increases the value of Core FSS. It should further improve retention economics and serves as a natural on-ramp to broader Dash adoption. This is the most credible and immediate way that AI creates value for Dropbox FSS customers today. Now turning to our plans to scale the Dash stand-alone opportunity. And while it's true that we've introduced different iterations of Dash experience over the last 2 years, the sequencing of our rollout was intentional to ensure we build and scale the business and products thoughtfully. First, we focused our investment on building a best-in-class Dash product experience, including investments in its underlying infrastructure and performance. Then we focused on launching 2 growth motions for Dash, the sales-led motion that launched in late '24 and the self-serve version that launched in Q4 of last year. Now we're focused on engagement and adoption before we focus on monetization. The good news is we're seeing positive early signals of demand. At the same time, we're clear eyed that onboarding friction, time to value and the experience around connecting your apps need to improve. So in the first half of '26, we're focused on improving the new user experience to demonstrate connector value from first touch. We're investing in stacks as a sharing-driven growth engine, and we're compressing the time between sign-up and first value in your Dash experience. Next, historically, Dropbox has been primarily a product-led growth company. We have a sales-led motion today, but it needs meaningful improvement given our broader product portfolio. In December, we hired Eric Webster as our new Chief Business Officer. His mandate is to evolve and improve our existing sales-led motion into one capable of selling multiple products with the right funnel, process and enablement. That includes Core FSS, Dash, both stand-alone and bundled, Protect and Control, DocSend and other emerging products. Protect and Control is showing particular promise. As every company works to roll out AI tools safely, admins are confronting critical security challenges with overshared content and improper use of consumer AI tools. We're in a unique position to help these customers. By complementing our Dash offering with Protect and Control, we can both index customer data and use the underlying context engine to power capabilities that prevent authorized sharing and access beyond our secure perimeter. Capitalizing on this emerging demand, we closed a 6-figure international deal for Dash's protect and control features in Q4, and we expect Protect and Control to play an important role across our portfolio in years to come as AI data security emerges as both a stand-alone opportunity and an AI adoption enabler. Stepping back, here's how all this comes together. Our Core FSS business is stabilizing and showing credible paths back to growth. Dash is both a force multiplier for core and a stand-alone AI opportunity, while sales-led growth and AI data security expand our addressable market. Together, these vectors give us multiple paths to drive modest but meaningful growth and enough to shift the narrative to durability and progress. So in closing, 2025 laid the foundation. Now 2026 is about execution, scaling what's working, improving consistency. We're realistic about the work ahead, but confident in the direction, the team and the opportunity in front of us. Lastly, I'd like to acknowledge the many contributions of Tim Regan, our departing CFO, and thank him for making Ross' transition a smooth one. So with that, I'll turn over the call to Ross to walk through our fourth quarter results and our outlook.

Thanks, Drew, and good afternoon, everyone. As many of you know, this is my first earnings call as CFO of Dropbox. Before I walk through our financial results and outlook, I wanted to share a brief perspective on how I think about the business and the opportunity ahead. What I'm about to share reflects my observations from my first couple of months in the role. It's not a new operating framework, and it doesn't represent a change in how we guide the business. But I believe it's useful context as you assess Dropbox's long-term value creation potential. What initially attracted me to Dropbox was the strength of the foundation. This is a company with a strong global brand and a large and loyal customer base of roughly 18 million paying users and 575,000 paying business teams and products that are deeply embedded in everyday workflows for both individuals and teams. That foundation is clearly reflected in the financial profile, a $2.5 billion revenue business with operating margins around 40%, approximately $1 billion of annual unlevered free cash flow and a 21% 3-year CAGR for unlevered free cash flow per share. That combination of scale, profitability and cash generation has proven to be durable and resilient over time. Our North Star is to grow free cash flow per share over time through a judicious capital allocation strategy. As CFO, my goal is to prioritize investments in the business where we see attractive returns, initiatives that drive sustainable revenue growth and margin. At this time, restoring revenue growth is our top priority. When our shares trade at compelling valuations, repurchasing stock remains a disciplined and efficient use of capital. Reducing share count under those conditions increases free cash flow per share and enhances long-term shareholder returns. What ultimately drew me to Dropbox was the opportunity to grow free cash flow itself, not just optimize the denominator by growing revenue and improving margins. Let me start with our current investment priority, growth. Naturally, since onboarding, I have been most focused on our initiatives to restore growth. While many discussions regard Dash, our opportunities to restore growth in our Core FSS business are also exciting. We recognize FSS operates in a mature and competitive market, and we're realistic about that backdrop. At the same time, over the past year, we've taken meaningful steps to strengthen the organization and evolve the product. In late 2024, we brought in new leadership to lead the core business who are experienced operators from large-scale tech companies. I've been genuinely impressed by both the caliber of talent we've been able to attract and the pace at which they're working to evolve the business across product, pricing, packaging and go-to-market motions. Last year, we focused on simplifying and strengthening our core business, which drove improvements in monetization and retention. That work continues. At the same time, the team has been integrating Dash AI capabilities into FSS, allowing customers to derive more value from the content they already store in Dropbox. From my perspective, this represents the most significant innovation to the Core FSS offering in a long time. Looking at the top of the funnel, one of the biggest surprises to me early on was the magnitude of gross new ARR that Core FSS still generates each year. Today, much of that is offset by churn. But by delivering more value through innovation like Dash and Dropbox, improved pricing and packaging and better end-to-end customer life cycle workflows, I believe there is a real opportunity to improve retention and grow net new ARR over time. Now turning to Dash. I see Dash as a genuinely valuable product and use it regularly in my day-to-day work. More importantly, nearly all Dropbox employees are weekly active users, and we're seeing strong engagement from active users in our early customer trials. We have an impressive engineering team rapidly innovating on an ambitious road map. At a minimum, I see Dash as a highly impactful evolution of our Core FSS offering and believe in addition to all our other efforts, it will help attract new customers, drive upsell and reduce churn. More optimistically, we will also drive adoption and later monetization of Dash as a stand-alone product. Regardless, anywhere along the spectrum, I see meaningful value creation potential for Dash and our AI product strategy. The third growth lever I'll touch on briefly is M&A. I don't view M&A as a silver bullet, and I know firsthand that not every transaction delivers as expected. But I also know that disciplined strategic acquisitions can meaningfully expand a product portfolio and contribute incremental ARR over time. Any acquisition we consider must meet a high bar for strategic fit and financial return. Over time, I see M&A as a lever that can accelerate product road maps, deepen our relevance with customers and complement the organic growth initiatives already underway. Taken together, Core FSS, Dash and M&A, these were the growth vectors I evaluated when deciding to join Dropbox. And after a couple of months inside the company, I see opportunity for each. Let me turn now to margins. The second driver of free cash flow growth is margin expansion. Should we someday decide to curtail our growth pursuits, I believe this business has the capacity to operate at margins meaningfully above current levels. That said, given the growth opportunities in front of us, we believe it's prudent to maintain our current investment levels to pursue growth. At the same time, I do believe that over time, we can be more aggressive on cost discipline. We see the potential for additional margin upside driven by scale, continued cost discipline and productivity improvements. AI, in particular, offers great potential to automate many manual people-intensive processes across all functions, not just engineering or customer support. We believe that when employed, these initiatives will drive significant productivity gains. In addition, we continue to look for opportunities to operate more efficiently through better tooling and geographic mix, shifting more work to lower-cost regions. Taken together, we believe these efforts can generate savings, which we can elect to drive margin or reinvest in growth initiatives. To be clear, these are observations for my first couple of months. There's real work ahead to translate them into execution. So stepping back, this is how I see Dropbox today, a strong brand with a durable financial profile, significant free cash flow generation and multiple avenues for long-term value creation. And as I look at how the business is trending, we're making progress toward returning to growth while optimizing for efficiency. In that context, I see meaningful optionality in the business that I believe is underappreciated by the market, reinforcing share repurchases as an important part of our strategy. With that, let me turn to our fourth quarter financial results and our outlook going forward. In Q4, revenue declined 110 basis points year-over-year to $636 million, but increased 40 basis points year-over-year when excluding FormSwift, which acted as a 150 basis point headwind to revenue. Constant currency revenue declined 160 basis points year-over-year to $633 million, but was roughly flat year-over-year, excluding the 150 basis point headwind from FormSwift. Relative to our guidance, revenue outperformance was driven primarily by retention improvements across our self-serve SKUs. Total ARR was $2.526 billion, down 190 basis points year-over-year and excluding the impact of FormSwift, which was a 160 basis point headwind, ARR was down 30 basis points year-over-year. Total ARR declined 170 basis points on a constant currency basis. We exited the quarter with 18.08 million paying users, a sequential increase of approximately 10,000 paying users. The quarter's paying user growth was primarily driven by momentum in our Simple plan. Average revenue per paying user is $139.68 as compared to $139.07 in the prior quarter. ARPU increased sequentially primarily due to FX tailwinds as well as an overall mix shift from annual to monthly plans. Before we continue with further discussion of our P&L, I would like to note that unless otherwise indicated, all income statement figures mentioned are non-GAAP and exclude stock-based compensation, amortization of purchased intangibles, certain acquisition-related expenses, net gains and losses on real estate assets, workforce reduction expenses and net losses on equity investments. Our non-GAAP income also includes the income tax effect of the aforementioned adjustments. Gross margin was 80.8% for the quarter, down 230 basis points from the year ago period, reflecting higher depreciation associated with our hardware refresh and ongoing data center build-outs as well as increased infrastructure costs associated with the expansion of Dash trials. Operating margin was 38.2%, ahead of our guidance of 37% and up roughly 130 basis points from the year ago period. Operating margin increased year-over-year largely due to lower headcount following our risk in 2024 and elimination of marketing support for FormSwift. Compared to our guidance, operating margin benefited primarily from revenue outperformance as well as lower outside services and marketing spend. Net income for the fourth quarter was $174 million. Diluted EPS for the fourth quarter was $0.68 based on 254 million diluted weighted average shares outstanding compared to $0.73 in the year ago quarter. The decrease was largely due to higher interest expense. Moving on to our cash flow and balance sheet. Cash flow from operations was $235 million, an increase of 10% versus the year ago period, primarily due to payments related to our reduction in force in Q4 '24. Q4 '25 also included $26 million of interest payments, net of the associated tax benefit related to amounts drawn under our term loan facility. Unlevered free cash flow was $251 million or $0.99 per share, up 44% year-over-year. Capital expenditures were $11 million in the quarter, primarily related to data center build-outs. In the quarter, we also added $34 million to our finance leases for data center equipment, marking the end of elevated spend for our hardware refresh cycle. And now I'll provide a brief update on our real estate strategy as we continue to actively pursue subleases across our real estate portfolio. Last month, we executed a sublease of all remaining available square footage in our current San Francisco headquarters, including the portion of the space we were occupying over a 3-year term. We also executed an extension and expansion of an existing sublease. As a result of these 2 subleases, we expect to generate approximately $97 million in total future cash payments over the remaining term of our lease through 2033, net of the cost to lease a smaller San Francisco headquarters, given we will vacate our current headquarters. From a cash perspective, the 2026 impact is immaterial due to lease structure and investments we plan to make later this year in our new San Francisco headquarters. From a P&L standpoint, we expect a modest benefit in 2026. The impact of both of these new agreements has been factored into the guidance we'll provide today. As we move beyond 2026, both the cash flow and earnings benefits become more meaningful as the sublease income builds. Turning to the balance sheet. We ended the quarter with cash and short-term investments of $1.04 billion. In the fourth quarter, we repurchased approximately 14 million shares, spending approximately $415 million. As of the end of the fourth quarter, we had approximately $1.17 billion remaining under our existing share repurchase authorization and $1.2 billion of additional term loan liquidity with $700 million allocated to retire our March 2026 convertible notes. I'll now offer our outlook for Q1 and the full year 2026. For the first quarter of 2026, we expect revenue to be in the range of $618 million to $621 million. Excluding FormSwift, this implies 0.4% growth year-over-year at the midpoint. We are expecting a currency tailwind of approximately $8 million. On a constant currency revenue basis, we expect revenue to be in the range of $610 million to $613 million. We expect our non-GAAP operating margin to be approximately 38% Finally, we expect diluted weighted average shares outstanding to be in the range of 241 million to 246 million shares based on our 30-day trailing average share price. For the full year 2026, we expect revenue to be in the range of $2.485 billion to $2.5 billion. Excluding FormSwift, this implies roughly flat growth year-over-year at the midpoint. We are expecting a currency tailwind of approximately $27 million. On a constant currency revenue basis, we expect revenue to be in the range of $2.458 billion to $2.473 billion. Gross margin to be in the range of 81.5% to 82%, non-GAAP operating margin in the range of 39% to 39.5%. We expect unlevered free cash flow to be at or above $1.040 billion. We expect cash interest expense net of tax benefits of approximately $190 million. We expect CapEx to be in the range of $20 million to $25 million and additions to finance lease lines to be approximately 4% of revenue. Finally, we expect diluted weighted average shares outstanding to be in the range of 227 million to 232 million shares. I'll now share some additional perspective on this guidance for 2026. Excluding FormSwift, we are guiding to a flat revenue year in 2026 while continuing to invest. That reflects a disciplined approach as we validate execution, refine go-to-market motions and ensure that improvements translate into measurable results. Our guidance reflects that balance. We see long-term opportunity, but we are pairing that conviction with near-term prudence. Regarding revenue, following the elimination of marketing support for FormSwift at the beginning of last year, the business has experienced gradual user decline each quarter and will continue to be a modest headwind this year. Further, we have made the decision to sunset FormSwift by the end of the year. For paying users, last year, we offered directional commentary because of strategic decisions we made, including the wind down of the FormSwift business. Looking ahead to 2026, we expect modestly negative net new paying users in Q1, largely due to seasonality and FormSwift headwinds with roughly flat paying user growth for the remainder of the year. On gross margin, we expect modest pressure this year as we scale Dash trials, partially offset by ongoing structural infrastructure improvements. For operating margins, as we mentioned last quarter, we do not expect this to be a year of margin expansion. We remain confident in our ability to execute and believe it is prudent to invest in near-term growth opportunities. Our margin outlook reflects material investment in Dash as we expand trials across both new customers and a larger segment of our FSS user base. We expect these investments to be partially offset by ongoing cost discipline and efficiency initiatives. Regarding finance leases, this quarter marks the end of elevated spend for our latest hardware refresh cycle. And as a result, we expect materially lower infrastructure investment this year with finance lease activity more heavily weighted towards the second half. As a reminder, we typically refresh our infrastructure every 5 years. Regarding CapEx, we expect a slight increase in CapEx as a result of a onetime incremental investment related to the build-out of our new San Francisco headquarters. Excluding that investment, CapEx will be down year-over-year as we have completed our hardware refresh cycle. Our unlevered free cash flow guidance reflects a benefit this year from lower cash taxes related to the One Big Beautiful Bill Act, along with the absence of onetime cash outflows we had in 2025 related to the San Francisco lease buyout and reduction in force. Our interest expense outlook assumes we draw the remaining balance on our term loans. Once drawn, total outstanding term loan debt will equal $2.7 billion. Lastly, we expect our weighted average shares outstanding to decrease to approximately 227 million to 232 million shares, which assumes we exhaust the remaining balance on our share repurchase authorization. With that, operator, please open the line for questions.

[Operator Instructions] Our first question comes from the line of Mark Murphy from JPMorgan.

This is Jaiden Patel on for Mark Murphy. It was great working with you, Tim, and welcome, Ross. You've talked about Dash for a few quarters now, goal and pipeline building. Can you give us any quantitative framework around Dash seats, attach rates or ARR contribution at this point? And then looking forward to the guide, again, we've heard some of these strong proof points and very much understand that the focus is first on adoption, but would love to hear any assumptions you're baking into the guide?

Sure. I can start. And well, I'll start with just like the conceptual framework. I mean we start by focusing on product quality and building the capabilities, the infrastructure to index the known universe of SaaS apps and build a private search index and all the other things that go into building a product like Dash. But then we focus on engagement and make sure onboarding is good and that there's repeat use, then we scale it up to our user base and focus on driving adoption and then monetization. So we'll be able to share more specific metrics and targets and so on as we continue on that progression. I'd say where we are right now, as I shared in my remarks earlier, is that we spent a lot of last year really building that experience and focusing on product quality and building those integrations, the Dash AI integrations natively into the Dropbox experience and also building the self-serve version of Dash, which we need to reach our self-serve Dropbox FSS customers and beyond. And then now we're focused on -- and we've seen a good early signal there. So good repeat use of the integrations within Dropbox. Lots of -- we're focusing on tuning up the onboarding experience. And then now we're turning towards driving adoption in the first half by scaling up the integrations to more of our Dropbox Business customers for the Dash integrations and then rolling out the Dash stand-alone products more broadly in the first half. And then the second half will start to phase in more around monetization. So probably second half will be a better time to share more of the specifics around attach rates and ARR contribution and things like that.

Jaiden, this is Ross. Thanks for the question. I just wanted to add, I agree with everything Drew said. We are excited for what we're doing around the core business and our ability to do things that drive us to a growth posture as well as for what we can do with Dash. And we are focused very much on engagement adoption this year. And to your guidance question, we leverage all information we have in front of us and just given the size of the core business, you can assume that, that has the most weighting and influence on how we think about our guidance for the year.

And our next question comes from the line of Rishi Jaluria from RBC.

Maybe to start with, I want to continue pulling on the thread of Dash. Look, I'm in total agreement that you have to drive utilization and ultimately, customer value before we can really worry about monetization in a big way. And you've given us bits and pieces over the years. But maybe what sort of metrics can you give us around engagement with Dash, whether that's anything like people spending more time in Dash, percentage of paying business users using it, even like what impact does this have on gross retention? Any sort of metrics you can give us around like engagement and adoption and feedback around Dash would be helpful. And I've got a quick follow-up.

Rishi, it's Ross. I'll start here. I just -- I know coming on, I'm looking at how all this has played out, and I know we've been talking about Dash for a while, and we've been very focused on investing in building the product, which I think is a great product that gives me a lot of value. I think that what we're focused on now is like, just remember, we launched this in Q4 to the Dash and Dropbox solution into our core. We think that, that is a tremendous evolution of our core product set. It drives a lot of value for our users. We launched it to a small number of users, and we've seen some really good results from that in terms of those core users adopting and using Dash and returning to continue to use Dash week-over-week. And those results exceeded our expectations and has given us the confidence to go forward and accelerate our rollout of Dash to more of our users in this year. So I think we're seeing some nice results on that side. And again, I think that Dash is a significant enhancement of our Core FSS product line and something that we can use to drive value for our users. And then on the self-serve side, and Drew talked about this in his prepared remarks, we've seen some good results in top of funnel in our Q4 launch. There is work that we need to do to just showcase time to value faster for these customers, and we're working on that. But overall, I think we're pleased with where we are with adoption. It's allowing us to accelerate that rollout to our core business. And as we see more adoption and get into the monetization phase, we will talk about it more, and we'll introduce metrics as appropriate to help you track it better.

All right. No, that's really helpful. And then maybe just continuing on Dash, but I want to think about a broader kind of more medium-term strategy. Drew, I know the idea of having kind of this connectivity of knowledge and content has been a thing you've been focused on for a very long time and Dash totally fits in with that. Maybe what's the longer-term opportunity for you to not only leverage kind of this idea of universal search and knowledge attainment, but even get that a little bit more workflow integrated and turn Dash into maybe being more of a platform where your more power users have the ability to actually build content-specific agents on top of Dropbox that can automate a lot of that workflow and actually get a lot of work done given kind of the content you have system of record. Maybe how are you thinking about your opportunity and investments there?

Sure. I think it's a great question and something we're very focused on. So we talk a lot about building Dash itself, but I think in a lot of ways, what we've really been building and why this investment has been over many years instead of a couple of quarters is because we're building or we've built a completely new generation of technical infrastructure that we internally call our context engine. And so what that is, is shifting the value we're providing at the platform level from basically like really scaled and cost-effective storage to building this context layer for AI that indexes the known universe of SaaS applications, builds kind of a private search engine and then connects -- basically formats all of that content in a way that a language model or an agent can work with it. And we've -- and to your question of like, well, are we going to shift from sort of informational use cases like search or chat to helping people get the work done, you're starting to see us do that across the portfolio beyond Dash. And we'll do it within Dash too. But just to give a couple of examples, part of what really resonates with customers with the Dash integrations in the Dropbox is for the first time, they can talk to their Dropbox in natural language, and it kind of blows their mind. So if they want to find a photo of a red sunset, it doesn't have to be called red sunset.GPG anymore. If someone says red sunset in a video, that search can find it. And increasingly, we'll be shifting on all of our surfaces from kind of informational queries like that to actually automating workflows and helping you get stuff done. Security is another example that I talked about. So every company is trying to figure out how do we roll out AI safely and confront a lot of new issues. The first is overshared content. So to some extent, every company has documents floating around with a broader permission set than it should. And while that's -- and customers have been talking to us about that problem for a few years. And in response, we -- that's one of the reasons why we bought a company called Nira last year and have been building on that since with what we now call Protect and Control. But we find that customers are really struggling how do we deal with this overshared content. And that might have been a theoretical problem a few years ago, but with enterprise search, with these AI tools, suddenly employees can literally just ask for sensitive or bad stuff and find all of it, making it a lot more dangerous. Second, as companies have -- as CEOs or companies have encouraged AI adoption, and tried to hurry that up. What they're also seeing is that something like 30% or 40% of queries that first employees are using consumer AI tools like ChatGPT to answer these questions or at work and like 30% to 40% of those queries have people pacing PDFs or of really sensitive customer or company IP or just sensitive material that they shouldn't be sharing and customers have no way of dealing with that. And then to your point about agents, on the one hand, this whole coding revolution, agent coding revolution that we saw last year is really not only itself kicking into overdrive, but when you look at things like OpenClaw or Cowork or others. Now there's a lot of excitement about, all right, can we bring this paradigm to knowledge work in general. But what you see there is that opens up a whole new -- that kicks these security concerns and overdrive at an even bigger level. And so these are all big opportunities for us beyond just the basics of AI search and chat, and we're trying to strike the right balance because on the one hand, we're as excited as everybody else about all the transformative things you can do with AI as you give it more agency. At the same time, the median -- when I talk to sort of the median Dropbox customer, their biggest pain points are like are still more basic where it's like I have 10 search boxes when I really want one or like, yes, I'm using AI, but when I use ChatGPT, it doesn't know anything about me or my company or my work. And so there's still a lot of low-hanging fruit and just providing these kind of more basic levels of value in addition to the more workflow-oriented and like workflow automation pieces on top. But this and a bit of a longer answer to sort of address the spirit of some of the prior questions, too, is that like we've really been building a new generation of infrastructure that builds on top of 10-plus years of infrastructure before that, which is really tuned for storage. But as a result, when you look at our price points for Dash or other things, we're able to provide a product that really no one else has been able to provide, where it's unlike some enterprise search competitors or similar folks in the space. If a typical Dropbox customer wants to adopt one of those things, they usually are facing a $50,000 setup fee. They have to provision their own custom cloud infrastructure to run it. It's like a 3-month pilot. There's a lot of friction. And so both the opportunity and the challenge we've had to date is like how do we box all that up and put it in a package that anyone can just download with an app and be up and running in a few minutes. And it's really exciting that we're really close to the finish line there. And this half is when we'll really start scaling that up. So really building a next generation of technical infrastructure, lots of manifestations at the application level from better FSS to Dash to security, but I think it's an important kind of framework for how to think about these investments.

And our next question comes from the line of George Kurosawa from Citi.

I'm on for Steve Enders. It was good to see paid users return to sequential growth. I think you alluded to some improvements in retention. I'd just like to double-click on what do you feel like drove some of those improvements and how we should think about kind of sustainability and maybe further improvements you can make going into this year?

Sure. Well, as I said in my prepared remarks, I mean, the first thing that has really driven these improvements in a more fundamental way is bringing in a new generation of skilled leadership and who have in turn really elevated each of their functions and leadership teams as well. And I think that what you saw in Q4 is a reflection of a lot of that work starting to pay dividends. So the improvements have been across the funnel. I think last year, you've seen us really continue to drive steady improvements in retention and just improvements across the funnel. The individual business has done well, and you've seen steady growth across 2025. And I think what you -- but the bigger picture of what you see there in addition to the specific gains of different funnel metrics is really that we -- with the right execution and the right leadership, we can demonstrably drive sustained improvements in retention and growth. And then turning to '26, a lot of our focus is on the Teams business, where we faced various downsell pressure over the last couple of years since we launched a price increase a few years ago. But there's a lot of improvements we've been making there, too. So everything from basic stuff like improving churn and downsell directly by redesign and cancellation flows and better communicating the value we're providing, sort of no regrets, things like that, improvements to conversion. So a lot of the pricing -- we're investing a lot in simplifying our plans and tuning pricing and packaging, improving our trial flow. And so we've seen that paired with improving conversion rates on the way in. On the onboarding experience of setting up a new team, reducing just a lot of, again, common sense stuff like just taking -- sanding down all the rough edges and reducing staffs and friction and getting your team up and running, that's been paying dividends. But as you'd imagine, one of the things we're most excited about is just building a better product experience and taking the FSS product, a new generation ahead with the integration of all these capabilities from Dash and being able to talk to your Dropbox and being able to automate a lot of the work that you're already doing there. So we see a lot of room to continue improving across the funnel and across the portfolio. And it's been good to see some of those proof points become stronger in Q4.

Great. That's helpful color. I also wanted to ask about ARR. The last few quarters, we've seen revenue show good signs of stabilization. It seems like ARR seems to be diverting a little bit weaker, a little bit of a divergence there. Can you just talk us through the mechanics of why we might be seeing that and how we should think about those -- the delta between those 2 metrics this year?

Yes. Sure, George. This is Ross. I'll take it. I think it's an astute observation. We had some Q4 positives around paying users and ARPU and revenue and ARR was a little bit lighter. I think that they should be moving in the same direction, but I would just let everybody realize that we're talking about like a slight divergence around a neutral middle line. So it's not very far off in either direction around 0. So I think ultimately, those things -- the ARR should move in the same direction. I think in any given quarter, there's some discrepancies. In particular, in Q4, your ARPU and your ARPU metric has FX positives embedded where ARR is on a constant currency basis. And there's also some timing-related differences that impacts those metrics definitely. So ultimately, we're optimistic about the business and return to a growth posture and would expect that those start to move together in the future.

And our next question comes from the line of Matt Bullock from Bank of America.

I wanted to ask about the paying user growth assumptions embedded into the guide this year. it's encouraging to hear that we're targeting Teams license growth this year. But maybe help us think about what's embedded in terms of the full year paying user guidance, how we should expect that to evolve throughout the year across individuals and Teams plans and with the sunsetting of FormSwift as well, that would be helpful.

Yes. Thanks, Matt. I mean -- and just to reiterate for everyone, what we said is Q4, we are very pleased with the result of having positive net new paying users. And I think that, that's because it's the net number is driven by both the improvements we put into place around retention, which we hope will continue this year. And also, we're also targeting the whole customer journey and improvements for gross adds as well as upsell in addition to retention. So we're very pleased with the result in Q4. And as we look forward into 2026, I said earlier that we should expect some seasonality in Q1 such that net new paying users will decline in Q1. That's our expectation. And then for the full year, we expect it to be flat year-over-year in net new paying users, which I think compared to the last several quarters and years, is a positive result. And I think, again, that is a reflection of what Drew talked about, really, what I'm most excited about is like we've got a great team in Core, and they're rapidly iterating on some really cool initiatives that are intended to drive better retention and improvements across the customer journey to return the Core FSS business into a growth posture. And we're excited about Dash. And so I think that's reflected in the guidance around net new paying users. In terms of how it goes by quarter, I would just focus on -- we expect to be negative in Q1 and then make it up for the rest of the year to be flat for the year. I don't want to get too specific on each quarter thereafter.

Understood. And then one quick follow-up, if I could. I wanted to ask about potential M&A strategy. Which key areas would you potentially be evaluating opportunities to expand the product portfolio? Just trying to think through potential bolt-ons here going forward.

Sure. I can start. So I mean M&A has been a really valuable tool in our kit for scaling the company since the beginning and ranging from bringing in talent to bringing in early-stage products like things like Nira that I mentioned earlier and scaling them up to bringing in established businesses like HelloSign or DocSend. So we've had success across all 3, and we have -- and we continue to be very active in looking for opportunity -- M&A opportunities. And I think Nira is a good recent example. There have been others on the talent front where we've been able to bring in some really great AI talent, folks like Mobius Labs who have really deep capabilities in multimodal understanding. So like processing -- they're using AI to process large quantities of images and video and audio. I imagine, is very relevant for us. And then looking ahead, it kind of dovetails with what I said before, we've got this really powerful -- we see the big bottleneck in AI generally as this gap between AI tooling and your company's context. We've been building that missing context layer for AI. We built a whole new generation of technical infrastructure to facilitate that. As the world starts turning towards more agentic capabilities or people having their own agents, then there's a lot of new opportunities for that context engine to help make those agents actually able to connect to your work context, like to connect to your Gmail and your Salesforce and your Dropbox and everything else, not just the local files and your computer, which is the current limitation for a lot of these things. And then security, like securing and building a secure perimeter around your company for rolling out AI safely and agent safely, particularly in areas around content. So across both the infrastructure and the application layer, there's a lot of interesting opportunities, and we'll have more to share as the year progresses.

This does conclude the question-and-answer session of today's program. I'd like to hand the program back to Peter Stabler for any further remarks.

Thanks, everyone, for joining us today. We look forward to speaking with you next quarter. Have a great afternoon.