Good afternoon, ladies and gentlemen. Thank you for joining Dropbox's third quarter 2018 earnings conference call. All participants are in a listen-only mode. Should you need assistance, please signal a conference specialist by pressing the star key followed by zero. After today's presentation, there will be an opportunity to ask questions. As a reminder, this conference call is being recorded and will be available for replay from the investor relations section of Dropbox's website following this call. I will now hand the call over to Darren Yip from Dropbox's investor relations team. Please go ahead.
Thank you. Good afternoon, and welcome to Dropbox's third quarter 2018 earnings call. Today, Dropbox will discuss the quarterly financial results that were distributed earlier. Statements on this call include forward-looking statements, including statements relating to the expected performance of our business, future financial results, strategy, long-term growth, and overall future prospects. These statements are subject to known and unknown risks and uncertainties that could cause actual results to differ materially from those projected or implied during this call. In particular, those described in our risk factors included in our Form 10-Q for the quarter ended June 30th, 2018, as well as the risk factors that will be included in our Form 10-Q for the quarter ended September 30th, 2018. You should not rely on our forward-looking statements as predictions of future events.
All forward-looking statements that we make on this call are based on assumptions and beliefs as of today. We undertake no obligation to update them except as required by law. Our discussion today will include non-GAAP financial measures. These non-GAAP measures should be considered in addition to, and not as a substitute for, or in isolation from, our GAAP results. A reconciliation of GAAP to non-GAAP results may be found in our earnings release, which was furnished with our Form 8-K filed today with the SEC and may also be found in the supplemental investor materials posted on our investor relations website at investors.dropbox.com. I would now like to turn the call over to Dropbox's Co-Founder and Chief Executive Officer, Drew Houston. Drew?
All right. Thanks, Darren. Good afternoon, everyone, and welcome to our earnings call. On the call with me is Ajay Vashee, our Chief Financial Officer, and Yamini Rangan, our Chief Customer Officer, will also join us during Q&A. Today, I'll talk about our business and product highlights and provide you with an update on our go-to-market strategy and ecosystem. Ajay will review our Q3 financial results and provide guidance for the remainder of fiscal 2018. Q3 was another strong quarter, with revenue growth of 26% year-over-year, driven by continued paying user growth and ARPU expansion. We also generated a strong non-GAAP operating margin, which helped us deliver a 33% free cash flow margin. These results continue to demonstrate the strength of our global collaboration platform, our efficient go-to-market strategy, and our operational discipline. We hit significant milestones this past quarter that I'll speak to in a moment.
Before I do that, I want to thank our team for a great first three quarters as a public company. We've built a strong, sustainable business based on a deep understanding of our customers and the tools they need to do their best work. We're operating at scale with over 12 million paying users, and we have healthy margins that continue to improve because of our infrastructure efficiencies. We're also executing against our mission of designing a more enlightened way of working. We're building a modern platform for work and collaboration by bringing together content and all the communication and coordination around it. Users love our new product features and updates, and we're continuing to expand our open ecosystem to reduce the friction between the tools people use at work. Moving on to the latest on our products.
We delivered a number of innovations in Q3 that make the Dropbox experience even better for our users, teams, and admins. One of the most important ways we're improving our platform is through our machine intelligence efforts, which we call DBXi. Our workdays are getting noisier and more disorganized. Our content and the conversations around it are scattered across many applications, between emails and text messages and documents and spreadsheets, making it harder and harder to find what you're looking for. We see a lot of room for improvement here. In Q3, we introduced two new machine intelligence initiatives to help solve these problems. First, we upgraded our search engine with an improved architecture called Nautilus that we've built from the ground up. Nautilus harnesses the power of machine intelligence to deliver a completely personalized search experience for our hundreds of millions of users.
It surfaces the most relevant files quickly and efficiently so our users can focus on their most important work. Second, we've made big strides in making images searchable in Dropbox by launching new optical character recognition, or OCR technology, into early access for select business subscribers. There are billions of documents and PDF files on Dropbox that are actually a photo or a scan of original document and are therefore stored in image form as opposed to text. This presents a problem because most tools we use today can't find text in images, so when you do a typical search, it turns up no results. As a result, we noticed that many of our customers have workflows that involve manually combing through a bunch of JPEG files or PDF scans.
Beginning in Q3, we began automatically scanning all images and PDFs that are stored in Dropbox, making the text within them easily searchable. In going forward, this machine intelligence-enabled feature will be available to users on our professional, advanced, and enterprise plans. We also noticed that today's workflows are often highly fragmented, with users toggling between several different applications when working on a single project. To help solve this problem, we announced Dropbox Extensions so that users can initiate and manage workflows with category-leading partner applications such as Adobe, Autodesk, DocuSign, HelloSign, Adobe Sign, Vimeo, and more, all within Dropbox. Now users can seamlessly perform actions like edit PDFs or annotate videos, request signatures, and more right from Dropbox.
For example, our users will have the option to start content-based workflows, such as taking a contract from first draft to final draft and requesting a signature without ever having to download or scan a document. Extensions automatically save a user's content back to their Dropbox folder, that keeps it organized, in sync, and in one place. In this way, individuals and teams can utilize the functionality of external applications and maintain a centralized home for their most important work. For Dropbox, extensions will help drive continued engagement and stickiness for our platform. Turning to Paper. Paper continues to be an important part of our platform. Not only does it help drive better conversion and retention of team subscribers, it also serves as a testing ground to help us design productivity features that will help shape the future of work.
We've been observing how our early adopters use Paper are continuously adding new features and functionality to better streamline their workflows. In Q3, we launched Paper Timelines, an interactive tool for project management. With Timelines, teams can add due dates and milestones and share status updates to stay aligned. From aggregating key files to assigning tasks, teams can easily oversee all of their projects and corresponding activity so nothing gets overlooked. Paper's also a powerful tool for unifying content with coordination. We recently partnered with several best-in-class companies, including Adobe, Airtable, and Lucidchart, that our users can embed design files, spreadsheets, and graphics directly from these applications into one consolidated Paper doc. Finally, we also enhanced the core Dropbox experience and enabled Paper folder previews.
With this added functionality, users can paste a link into a Dropbox folder that their team can navigate through without ever leaving a Paper doc. Next is the admin experience. With more than one billion pieces of content saved to Dropbox every day and over 300,000 teams paying for Dropbox Business, our platform is optimized for performance and security at scale. Our customers use a variety of endpoints, apps, and products to manage how their employees and teams work together. To help empower administrators to use the tools they need to facilitate collaboration, we announced partnerships with Google Cloud Identity, BetterCloud, Coronet, SailPoint, and more. These integrations make Dropbox a secure, unified home for work, allowing admins to configure the security tools they use seamlessly with Dropbox. Now, let's move on to the infrastructure that powers our platform.
We have over half a billion registered users who trust us with over an exabyte of data. For Dropbox, this means millions of HTTP requests and terabits of network traffic. To support all that, we've built an extensive network of points of presence around the world that we call Edge. We just announced a new point of presence in Toronto, we'll be adding two more in Scandinavia by the end of the year. These network expansions provide customers access to fast, reliable bandwidth owned by Dropbox and dedicated exclusively to their use, ultimately improving performance and reliability. Switching gears to our ecosystem. We continue to forge partnerships that position Dropbox at the center of our users' workflows. We're firm believers in the power of integrations, one of our most important differentiators is that we're a uniquely open and interoperable platform.
We help our users stitch together their content and work across all types of devices, operating systems, and applications. In October, we announced a strategic partnership with Zoom to expand remote collaboration for teams and individuals around the world. Zoom is a leader in modern enterprise video communications, offering an easy-to-use and reliable cloud communications platform for audio and video conferencing, chat, and webinars across all devices and meeting spaces. Together, Dropbox and Zoom will develop a series of cross-platform features that facilitate real-time communication around shared Dropbox content, making teamwork more effective. Within Dropbox, users will have the option to initiate or join a Zoom meeting while viewing and working on shared content. During a Zoom meeting, users will be able to share content such as documents, slides, and images from Dropbox and display them on screen.
The access content can be saved back to Dropbox. The partnership will also promote cross-customer discovery and adoption of each other's products and is an example of how we're exploring new ways to drive growth through additional go-to-market channels. We also deepened our partnership with Salesforce. In addition to launching updated apps for Salesforce's Sales Cloud and Service Cloud, in Q3, we launched the Dropbox app for Quip. The Dropbox live app allows Quip users to better collaborate around their Dropbox content. Users can insert a Dropbox folder directly within Quip and navigate to view, open, and download any of the files inside. At Dreamforce, we also demoed our latest pilot integration with Salesforce's Marketing Cloud. For marketing teams, managing digital assets tends to be cumbersome, especially when working with external collaborators like creative agencies.
Our solution will let users create branded, personalized folders within Marketing Cloud that are accessible by both internal teams and external partners. With these two-way workflows, third-party contributors can add assets to a central Dropbox folder that's available from within Salesforce. As I mentioned earlier, I'm really proud of the progress we made over the last quarter. We launched new product experiences and features like Nautilus and Paper Timelines, and we're leading the way in building an open ecosystem that supports a more seamless work environment through our launch of Dropbox Extensions and deep integrations with companies like Zoom and Salesforce. We have a huge opportunity to improve the experience of using technology at work, and I'm excited about the future we're building. Now I'll turn it over to Ajay, our CFO, to walk through our financial results.
Thank you, Drew. Our Q3 results continue to demonstrate our strong execution and focus on delivering top-line growth and free cash flow generation. Total revenue for the quarter was up 26% year-over-year to $360 million, driven by an increase in total paying users and ARPU expansion. We ended Q3 with 12.3 million paying users, with the majority of new users coming through our self-serve channels. ARPU was $118.60 in Q3, up 6% from $112.05 a year ago. The year-over-year ARPU expansion was primarily driven by strong adoption of our premium professional and advanced plans by new paying users. We also saw some tailwinds from teams opting to remain on our advanced plan upon the expiration of their grandfathering period. Our continued growth in ARPU reflects our strategy to methodically convert our highest value users to drive sustainable monetization and retention.
In addition to driving higher value conversions, we're also focused on expanding our existing teams. Dropbox is unique in that we couple a highly efficient viral self-serve engine with a targeted outbound sales team. This go-to-market approach gives us the ability to serve customers of all sizes and is an effective mechanism to land and expand paid deployments within organizations over time. In Q3, we had a number of wins across a range of verticals, including technology, advertising, retail, and manufacturing. For example, we're excited to share that Flint Group expanded its Dropbox deployment last quarter. Flint Group specializes in the manufacturing of printing supplies and equipment and has over 140 facilities worldwide. The company needed a tool to share files with partners, customers, and suppliers. Without any internal marketing or training, Flint Group's Dropbox deployment has grown from about 100 users to over 1,500 users in 18 months.
We're also pleased to share that global IT distributor, Ingram Micro, expanded its Dropbox Business team deployment in Q3. Ingram Micro's cloud division has established several high-value workflows on Dropbox, including external sharing with partners, as well as internal collaboration and planning. With their most recent purchase, Ingram Micro's IT team will begin to leverage premium Dropbox Business features like domain verification and account capture to gain insights into Dropbox usage by employees and ensure that work and personal files are appropriately segregated. The company also chose Dropbox because of our ability to integrate with an array of best-in-class security providers. Before I move on to the P&L, I want to note that unless otherwise indicated, all income statement measures that follow are non-GAAP and exclude stock-based compensation.
A reconciliation of GAAP to non-GAAP results may be found in our earnings release, which was furnished with our Form 8-K filed today with the SEC and in the supplemental investor materials posted on our investor relations website. Gross margin for the quarter was 76%, an increase of seven percentage points compared to the third quarter of 2017. The increase in gross margin was primarily driven by unit cost efficiency gains with our infrastructure hardware, including lower depreciation as a share of revenue. We now expect gross margins to be approximately 75% for fiscal 2018, as lower depreciation expense on storage infrastructure will offset spend on network expansion as we grow our global footprint. Moving to operating expenses, third quarter R&D expense was $105 million, or 29% of revenue, compared to 26% in Q3 a year ago.
The increase as a percentage of revenue was primarily driven by higher headcount and investments in new product development and testing. S&M expense was $87 million in the third quarter, or 24% of revenue, compared to 23% in Q3 a year ago. The higher spend was primarily driven by incremental marketing expenses, including our presence at Dreamforce. G&A expense was $35 million, or 10% of revenue, and two points lower than our G&A expense as a percentage of revenue in the prior year. The decrease was a result of lower non-income-based taxes and outside services spend. Taken together, we earned $46 million in operating profit in the third quarter. This translates to a 13% operating margin, which is a five percentage point improvement from Q3 of 2017. Our strong operating margin performance was driven by the gross margin expansion and higher revenue in the quarter.
Net income for the quarter was $45 million, up from $24 million a year ago. Diluted EPS was $0.11 per share, up from $0.07 per share in Q3 2017, based on 420 million diluted weighted average shares outstanding as of Q3. Moving on to cash balance and cash flow, we ended Q3 with cash and short-term investments of more than $1 billion. Cash flow from operations was $128 million in the quarter. Capital expenditures were $8 million, yielding free cash flow of $120 million, or 33% of revenue. CapEx in Q3 included $3 million of spend on our new headquarters. Excluding the spend, free cash flow would have been $123 million. We did not receive any tenant improvement allowance reimbursements from our landlord during the quarter.
For fiscal 2018, we expect total CapEx to be $60 million-$65 million, including $30 million-$35 million related to the build-out of our new headquarters. We do not expect to collect any offsetting tenant improvement allowance reimbursements in the fourth quarter. In Q3, we had $29 million of additions to our capital lease lines for data center equipment. We continue to expect additions to capital lease lines to be high single digits as a percentage of revenue this year. Turning to our guidance, for the fourth quarter of 2018, we expect revenue to be in the range of $367 million-$370 million, non-GAAP operating margin to be in the range of 9%-10%, and diluted weighted average shares outstanding to be in the range of 417 million-422 million, based on our trailing 30-day average share price.
For the full year of 2018, we are raising our revenue guidance, which was previously $1.366 billion-$1.372 billion, to $1.383 billion-$1.386 billion. We are also raising our non-GAAP operating margin guidance, which was previously 9.5%-10.5%, to 11.5%-12%. We are raising our free cash flow guidance, which was previously $340 million-$350 million, to $350 million-$360 million. This figure includes one-time spend related to the build-out of our new corporate headquarters. Excluding the spend, free cash flow would be $380 million-$395 million. Finally, we expect 2018 fully diluted weighted average shares outstanding to be in the range of 408 million-413 million, based on our trailing 30-day average share price. I'll now turn it back to Drew for closing remarks.
Thank you, Ajay. In closing, we had another great quarter. We're using machine intelligence and adding innovative new features to achieve our mission of designing a more enlightened way of working. We're building deep partnerships with best-of-breed players, giving our users more choice without adding more noise to their work. We're doing all this while generating strong free cash flow. On behalf of our management team, I'd like to take a moment to thank our customers, partners, and the entire Dropbox team. With that, I'd like to invite Yamini, our chief customer officer, to join Ajay and me for Q&A. As a reminder, Yamini oversees our customer and partner-focused functions, including sales, marketing, and business development. Operator?
Thank you. Ladies and gentlemen, at this time, we'll begin conducting a question and answer session. To ask a question, you may press star then one on your touch-tone telephone. To withdraw your question, please press the pound key. Please limit yourselves to one question and one follow-up question. Our first question comes from the line of Richard Davis from Canaccord. Your question, please.
Hey, thanks very much. Kind of when you think about it, if you have something with legs, and you kind of touched on this, you're a platform. If you have a platform, you get a bunch of partners to extend your technology. Again, you kind of touched on that, but could you just talk broadly about your strategy there? You've added a bunch over time, and I just feel like, beyond Zoom and things like that, how do you kind of think about this in terms of creating a much bigger business? Thanks.
Sure. Thanks, Richard. Our platform is a valuable asset, and our ecosystem is an area where we're going to continue to make a lot of investments. The starting point for us is the content that lives in Dropbox, and we see that a lot of the workflows in a business revolve around content. When you think about CRM or our partnership with Salesforce, salespeople are sending a presentation to a prospect, they're getting a signed contract back, and all that content lives in Dropbox. Similar with communication, a Zoom meeting, someone's going to be presenting some content from Dropbox. The transcript from the meeting could then live in Dropbox. When we think about the content we have in Dropbox, we have an exabyte of it, that is a lot of fuel for engagement.
Our integrations drive engagement, that platform, that ecosystem, drives engagement, which drives retention and stickiness to the platform, ultimately lifetime value and monetization. We see that these integrations help us deepen our relationship with customers. In the future, there'll be opportunities for us to drive distribution of our partners' apps, vice versa. It contributes to the business in a number of ways, and I think it's also a really important part of our approach, which is different from the typical Office suite.
Perfect. Thank you so much.
Thanks.
Thank you. Our next question comes from the line of Alex Zukin from Piper Jaffray. Your question, please.
Hey, guys. Congrats on another strong quarter. I wanted to unpack the ARPU growth a little bit. It saw another acceleration, and it sounds like it's being driven more from new user growth than existing teams sticking with higher price points. I guess, is that because you're starting to see some of the benefits from the increased marketing and branding investments you made last year? How sustainable, I guess, is this ARPU trend over time, over the intermediate term? That'd be helpful.
Yeah. It's driven by both adoption of the business product generally and then adoption of our higher-tier SKUs. I would say that is a combination of both investments we've made in our product-driven conversion engine over the last few years, and it's also a lot of the improvements in the higher-end functionality that's in our premium SKUs paying off.
Sure. This is Ajay Vashee. I'll just add a little bit more color and reiterate some of what Drew just said. If you look at the three drivers of ARPU growth for us in the past quarter, the primary driver was just higher attach rates to those premium professional and advanced SKUs by new paying users, and that's a result of us bundling more and more value in those plans and including more and more products in those plans, as well as a function of us getting better at targeting those new paying users through marketing initiatives and other user acquisition activities. We saw a continued tailwind from the expiration of grandfathering for certain team subscribers.
As Drew mentioned, we are seeing a higher and higher mix shift towards teams licenses, and that's as we get better and better at promoting and managing our customer journey, we're driving higher attach rates to our team plans more generally than those have a higher ASP. Our strategy going forward will be to continue to drive revenue growth through a combination of paying user conversion and ARPU expansion. At any given quarter, one of these levers may outpace the other, depending on the initiatives that we're deploying. Looking ahead as it pertains to ARPU and the question that you asked, while we're certainly focused on continuing to deliver more value to our users and on increasing adoption of our premium plans, there's likely to be just some quarterly variation in the rate of expansion in ARPU.
That's how I think about it for the time being.
Got it. That's helpful. Maybe just a quick follow-up. If you look at the net adds, again, another solid number versus consensus, flat with last quarter. If we unpack it a little bit around gross adds versus historical trends and any high-level commentary on churn, particularly given the continued strong ARPU growth and grandfathering?
Sure. I can talk at a high level about net revenue retention and churn, both metrics that we provided earlier this year as part of our IPO process, not ones that we're publishing regularly. I think starting with just revenue retention at the highest level, given the higher team subscriber mix that we're driving over time, you can assume that we're improving our net revenue retention. That's something we're certainly seeing with the business. Then as it pertains to churn, a component there, overall, across the business, those rates, again, remaining very stable over the past few quarters.
Got it. Great. Thanks again, guys. Great job.
Thanks.
Thank you. Our next question comes from the line of Justin Post from Merrill Lynch. Your question, please.
Great. Two questions. First, bigger picture, how do you see your go-to-market strategy changing with the CEO transition? Anything new you'll be focusing on that seems pretty interesting? Then, secondly, just on the guidance for 4Q, sequentially, I'm calculating about 3% at the high end. Any changes in October or just kind of the usual, we'll see how the quarter goes from here and conservatism? Thank you.
This is Drew. I'll start with the first. For others on the call, our Chief Operating Officer, Dennis Woodside, transitioned out of company last quarter. Maybe this is a good opportunity to introduce Yamini in a second, who is our Chief Customer Officer. We're super excited about Yamini and Ling and the bench that Dennis built. The short answer is no, at a high level, we don't see changes to our strategy. We think we have an incredibly scalable and efficient model. I'll let Yamini speak to it as well.
Thanks a lot, Drew. Our model is really starting with viral adoption and then driving to self-serve land and expand, then going into an outbound motion that is very targeted. This is a super efficient motion. We've been able to use data science and proprietary models to be able to really figure out where there are opportunities for expanding and leveraging our outbound sales teams on that targeted opportunities. This has been super efficient and scaling well, and we are really proud of the success we have achieved there. We'll continue with that strategy.
To answer the part of your question on our Q4 guidance, this is Ajay. Our philosophy there is consistent with our philosophy over the past couple of quarters. At the highest level, I'll say we continue to have a very healthy and a steady and predictable business. While we remain very focused on investing in growth, we don't incorporate benefits from new initiatives until we observe enough data to make an estimate as to their impact. We believe we have a number of levers to continue to grow revenue over time, and we'll incorporate those initiatives into our guidance as we execute on them.
Great. Thank you. Appreciate it.
Thank you. Our next question comes from the line of John DiFucci from Jefferies. Your question, please.
Yes. Thank you. I'm going to ask a follow-up to Alex's question about ARPU, because obviously that's something that's really important to our models. I'm not sure what you can tell us, but on those new paying customers, and that's actually great to hear, but can you tell us roughly what % of new customers are choosing the more advanced versions? Anything you can tell us along those lines, just because that, as Ajay said, was the primary driver of that.
Sure. This is Ajay. There's not a whole lot of detail that we're disclosing there. I can say a couple of things. If you look starting with the year-over-year ARPU growth, if you look at the $650 and change that we drove in year-over-year growth, again, the majority of that was driven by adoption of our premium SKUs by new paying users as we get better and better at landing users in the plan that's best suited for them. That's one thing I'll say. The secondary impacts and tailwinds to ARPU there were the continued grandfathering expiration, although we've gotten through the majority of those grandfathered users at this point in time. We're seeing this mix shift from individuals towards teams over time, and that's also a tailwind to ARPU growth.
As it pertains to the specific attach rates, what I can say is that we're getting better and better at driving higher attach rates, and so those rates are increasing pretty consistently. It is a structural tailwind that we now see to ARPU over time.
Okay. Cool. That's helpful. The attach is getting better. The majority, that was last year. I'm sorry, last quarter also, right, Ajay? Well, you said about 50%, I guess it was close to the majority. I guess one quick follow-up, and this is for Ajay, too. We had expected to see the CapEx jump this quarter. It sounds like it's going to jump next quarter to really come in full force for the headquarters build-out. Should we continue? Our model assumes CapEx stays at an elevated level into next year and beyond. Is that still how we should be thinking about that?
Sure. CapEx for the year, our current expectations are $60 million-$65 million. Then as it pertains to some of the one-time expenditures for our headquarters, still expecting about $30 million-$35 million this year. You're correct in that the majority of that spend is going to be in Q4, and we do plan to execute on that spend in Q4. That really starts to kick off the construction cycle around that new space, and that'll continue into 2019 and beyond. We're actually managing the build-out of our new facility in phases, we're currently focused on the first phase ahead of initial occupancy next summer. We're going to move the company over next summer. We'll have more detail to share on where we see CapEx trending, and some ranges there when we provide guidance on 2019 on our call in Q1.
Okay, great. Okay, thanks a lot, guys.
Thank you. Our next question comes from the line of Heather Bellini from Goldman Sachs. Your question, please.
Yeah. I had a question just following up on what everyone's asking related to ARPU, where you're seeing just really, really good year-over-year growth. I know you're not guiding to 2019, for next calendar year, we're all going to be setting our forecasts. I'm just trying to get a sense of how much of a tailwind would you say there is to grandfathering so that when we're kind of setting our models up for the next 12 months, we could think through, because obviously the comp's going to start getting tougher pretty soon. Is there any thought process you could give us to that?
Sure. As of the end of Q3, we had gotten through, this is Ajay, we had completed the renewal process for about two-thirds of our grandfathered team subscribers.
Okay.
That remaining third will be managed pretty ratably across the next two quarters, so across Q4 and Q1. In March of 2019, the whole grandfathering expiration process is going to end. I would just reiterate that the primary driver of that year-over-year growth in ARPU expansion and the majority of the driver of that 655 is just organically higher attach rates to our premium plans, and those attach rates are increasing. We do expect that general trend to continue going forward into next year.
Just to clarify, and the premium uplift that you're seeing, does that include people who are choosing to pay for the advanced SKU?
The-
The people who are grandfathered in that then are deciding to pay for it, are you assuming that you're going to get, or I guess is there a way to think about the % of people that are converting to the advanced SKU that were grandfathered?
Sure. We aren't sharing specific numbers, but I can say that the majority of team subscribers that are coming up for renewal are proactively electing to remain in that advanced SKU. They're just getting a whole lot of value out of the features that we've included in that plan.
Okay. That's really helpful. Thank you.
Thank you. Our next question comes from the line of Mark Murphy from JPMorgan. Your question please.
Thank you, Ajay. My congrats. Just looking at the strength of gross margins and free cash flow, I'm curious, as it relates, can you talk about how far along you are in the rollout of this high-density SMR technology? Just to the extent that it allows you to nearly double the storage volumes, does that in turn reduce your CapEx requirements that would be related to the build-out of data centers for capacity planning?
Sure. That's a great question. This is Ajay. I'm happy to take it. I would say that the strong revenue and gross margin that we delivered in Q3 were the primary drivers of our operating margin performance and our free cash flow performance. If you look to gross margin specifically, then to get to your question around SMR and CapEx, gross margin increased for a couple of reasons. One, just higher unit cost efficiency gains with our infra-hardware. Two, lower depreciation as a share of revenue. A third reason is also that we are beginning to realize efficiencies from our rollout of SMR technology. Just as a reminder to those on the call, SMR is something we talked about last quarter. It's a high-density disk drive technology that increases our storage capacity without increasing the physical dimensions of a disk.
It's something we're really excited about. We're the first major tech company to be deploying SMR at scale, and it's the kind of thing we can really benefit from
By owning our own infrastructure. If you think about our long-term roadmap for gross margins and our CapEx profile longer term, certainly investments in SMR and other efficiencies are a part of our plan to achieve our target gross margins and operating margins over time, and they do reduce the CapEx requirements for us to store core user data.
Okay. As a follow-up, Drew, I recall that in the past you have mentioned search as a concept you're working on, and how to make search more immersive and engaging. I know you touched on that very briefly tonight, but I'm just wondering if you could shed a little more light on your vision in that area and perhaps what it would mean for the user experience.
Sure. We're focused on a couple dimensions. One, I spoke to a little bit earlier, where we have our homegrown search infrastructure that allows us to deliver personalized search, at pretty massive scale, because it's a little bit different than searching the internet broadly. Every user has access to different sets of data. In the future, when we think about organizing and storing really any kind of content in Dropbox, in many cases, that data can live in a bunch of different places. We want to help our customers deal with the fragmentation they're experiencing because their content is scattered in all these different places. We think about search both of the Dropbox corpus and then more broadly, how do we help people bring all their content into one place and really organize it for them?
In addition to this, the core search of text, we also made it so that you can search images for text using our OCR technology. That really came about because we observed our customers with these super inefficient workflows around flipping through documents, trying to manually find information because it turned out that it was a photo of a document, normal search turned up no results. Part of it is search when someone types something into a search box, but even better would be if people don't have to search at all. We find our customers are also inundated with huge volumes of information, and we see a big opportunity to use machine intelligence to cut through the clutter and show our users what's relevant, using signals in Dropbox and using signals from other places like your calendar.
You can see that today in the product as we've shifted the core experience to being about activity, and less about a list of files. That kind of news feed at work is something that we're always evolving, and we think is a really great way to help people see what's relevant for them right now and to be able to predict that, machine learning is a really important part of that. Finally, one of the unique elements of our platform is that we use data science and machine learning to help drive engagement, and drive monetization. Helping try to predict, okay, what are you really trying to do in Dropbox? If you're sharing something, use that opportunity to show you some of our higher tier functionality like Dropbox Showcase, which is a much richer sharing experience.
Also help match you to the paid plans, whether individual or business, that makes sense for you. Zooming way out over a many-year timeline, I think it's going to transform the experience, and our goal is to have a much more calmer and focused working environment where we do the heavy lifting of trying to figure out what's relevant, so that you're not inundated with information.
Got it. Thank you.
I'd just say last point on that is our scale and the amount of data we have in Dropbox really plays to our strengths because if you think about it, the content in Dropbox or data is really the fuel for machine learning, and the fact that we have an exabyte of content and the scale of our adoption allows us to more rapidly build a better experience.
Thank you.
Thank you. Our next question comes from the line of Sarah Hindlian from Macquarie. Your question, please.
All right. Thank you so much. I appreciate it, Drew and Ajay. Drew, I want to follow up on what Richard was asking a little bit earlier in the call, because we've always thought the unique value of Dropbox is the sheer amount of content you have, and that it gets really interesting when you start to pull in workflow and communications tools into the platform. So we're clearly seeing the partnerships with Zoom and Salesforce, and I want to dig into what those may mean for the next act for the company. What do you think that next act or new product or adjacency you see in the future is now? Is it in that type of workflow or comms arena, or is it search? I guess if it's through partnerships, it would be really helpful if you could talk me through how those monetize.
Then I have a follow-up.
Sure. First, we watch our users' workflows, and we see a ton of opportunity to take friction out of the experience. The collaborative world today is very different from 10 years ago when you just think about the sheer volume of different tools and different companies. We find that our customers are kind of stuck duct taping all this stuff together. We see a huge opportunity for Dropbox to help tie together all those different ecosystems and tools. As you can see, over the last couple of years, we've been evolving the core Dropbox experience from being kind of a repository for your files to, in the future, being an intelligent workspace for any kind of content.
Integrations are a critical part of that because. Again, when you think about these other workflows like video conferencing or messaging, often what those people are talking about is content that lives in Dropbox. When you think about that workspace, turning Dropbox into a place where all that work happens, but people still have the freedom to use these best-of-breed tools without having to toggle back and forth constantly or cut and paste or all the other friction points, we see a huge need there. We'll continue investing in those integrations both to improve the experience, and you asked about how do we drive monetization. I'd say that happens indirectly because those integrations drive engagement, that engagement drives retention and lifetime value. Any way you look at it, we see that. Teams that have a third-party integration, they convert and retain at higher rates.
In the future, there's an opportunity to drive monetization directly to the extent that we provide distribution to our platform partners, and open up our platform to the SaaS companies that are looking for a bigger audience. We see a big opportunity for the platform to both deepen our engagement with our customers, do more for our users, and provide a great experience.
Thank you, Drew. That's super helpful. Just a quick follow-up then. Maybe this is for Yamini. Just with the go-to-market changes that you're taking on with Dennis leaving, what are some of the things you think you can do in particular to drive increases in net paid conversion? I think really in that area of paid conversion, user growth on net adds, can you get to growth, and how do you do that?
Yeah. Thanks a lot, Sarah. In terms of go-to-market, we are actually not changing any of the go-to-market strategy. I think we have a very effective strategy for landing in small teams and then driving the adoption into larger and larger teams. There is strength in that model, and we'll continue to kind of invest in that model. In terms of paid conversions, as Ajay mentioned, stepping back, we introduced a teams SKU last year in Q1 called Advanced. That in addition to the larger enterprise SKU, and the value that we drive there is what is driving paid conversions in our business customers. That's how we would continue to drive growth within the outbound segment as well.
This is Ajay.
Thank you very much.
I can add a bit more, Sarah, to your question around growth initiatives more broadly, and Yamini gave some great color on everything that we're focused on to continue to fuel that targeted outbound growth. At a high level, we're always working on growth initiatives to increase the size of our paying user base. A few examples where we're always investing in that product-driven conversion engine. We productize conversion at scale, and we're operating one of the largest self-serve conversion engines in the world. That is through investments that we're making continuously in things like machine learning and data science. Two, Drew talked a bit about the Dropbox platform. Expanding that platform is really important to us because those high-value integrations make our platform more attractive to new users, then they drive business value for us.
Three, I would say we're also exploring new go-to-market channels to acquire paid users, and our Zoom partnership is a great example of this. Still very early, but there is a joint go-to-market component of that deal where as we deliver more and more value to Zoom users, they're able to discover our products and subscribe to our products over time. Finally, we're always working on new product experiences to bring to market, and that helps extend our value to more and more users over time as well.
Thank you to all of you. I really appreciate the time.
Thank you. Our next question comes from the line of Mark Mahaney from RBC Capital. Your question please.
Hi, guys. It's Zachary Schwartzman. I'm for Mark. I have some follow-up questions on the continued growth in ARPU and net adds. The first is a question for Drew and Yamini. Can you talk a bit about some of the recent improvements to Dropbox self-serve acquisition funnel model you mentioned, specifically using data science techniques? Were any of these new levers that were turned on over the last two quarters? Ajay, in terms of larger enterprise contracts, does the company still feel it has a strong pipeline of lower hanging, under-penetrated, or under-monetized leads? Has this dynamic changed since the IPO? Thank you.
This is Drew. We think about our user growth and subscriber growth in terms of one customer journey. The way we've reached all these hundreds of thousands of paying business customers often starts with someone going to dropbox.com, getting the free version, bringing it to work, and then they might start collaborating on a project team with Dropbox, and then that project team becomes a self-serve deployment for the whole department, and then goes wall to wall with an enterprise or outbound deal. We see our job as helping move our customers along that journey as effectively as possible. Importantly, we want to drive sustainable growth. As Ajay has mentioned, we want to drive a good balance of healthy growth and profitability. In terms of what we've done, I would say there are teams working every day to continue to optimize that journey.
When I say optimize, I mean remove friction, help educate our users, figuring out how do we help our users, our customers, get all their information in Dropbox, how do we get your team properly set up? I would say there's a whole bunch of different levers across the business that we're always improving. At the end of the day, so much of it is driven by engagement and the quality of the product. When we look at these integrations, that gives users and teams more reasons to spend more time in Dropbox, get more value out of Dropbox. We see all these things working together as a system, and that's what we mean by a product-driven conversion engine.
Yep. As it pertains to the second part of your question, Zachary, this is Ajay. There's been a whole lot of consistency to our model over a multi-year period, certainly since we went public earlier this year, and that handoff between self-serve and outbound and that extension into larger organizations and enterprise has remained really consistent. I'll hand it over to Yamini to maybe provide a little bit more color on how that lead generation pairing works for us.
Yeah, absolutely. There is what we call a baton passing from our self-serve land and expand to our targeted outbound motion, and we continue to use data science models there as well. For example, we will look for signals on deployments that are expanding significantly. We call it the up-sell model, and we find that anybody that scores in the top decile of the up-sell model has a 6x chance of actually expanding with us. We use proprietary signals like that to really figure out where our users are engaging with the products, how they are actually driving the adoption, and that's where we queue up our team. That baton passing is working pretty effectively.
Great. Thank you. Congrats on the continued progress.
Thank you.
Thank you.
Thank you. Our next question comes from the line of Karl Keirstead from Deutsche Bank. Your question, please.
Thank you. One for Ajay, one for Drew. Ajay, just on cash flow, through the first nine months of 2018, you've put up a terrific 25-plus% revenue growth. Operating income growth has been over 150%, but operating cash flow growth has only been up 16%, so there's obviously some offset that might be holding the growth back, as solid as it is. I'm wondering if you could elaborate on what that is to just help us as we think through likely cash flow growth in 2019. For Drew, I just wouldn't mind asking you, Microsoft is talking a little bit more about bundling OneDrive into Office 365 SKUs, et cetera, and I'm just wondering, Dropbox has obviously powered through that very well. You can see that in the numbers, but I'm just curious what your thoughts are on how you sell against that bundling trend.
Thank you.
Sure. I can start with the first part of the question. This is Ajay. I would just talk maybe at a high level about the correlation between the P&L and cash flow for us, then I can talk a little bit about trends and where we see that going forward. At a high level, our operating income includes certain non-cash adjustments that don't always flow through to free cash flow. As an example, quarter to quarter, free cash flow can be impacted by factors like timing of spend as well as FX in a way that doesn't necessarily translate one-to-one to what's recognized on the P&L. Looking forward to Q4 and longer term, our strategy is certainly to expand free cash flow over time, but we also want to maintain the flexibility to spend on initiatives that we believe can help to drive growth.
Going forward, our guidance will continue to reflect our view of free cash flow trends, and we are raising our guidance this quarter just based on what we're seeing year to date. I'll pass it over to Drew to answer the second part of the question.
Yeah, for sure. Around Microsoft and Office 365 and OneDrive, as you alluded to, that's a dynamic that's OneDrive's been out for a long time and part of Office 365 for a long time. All of our growth has happened within that environment, and we see ourselves continuing to compete successfully there. I think that starts with, first, just the product experience and our focus on design. Users love the fact that Dropbox just works. You think about our cumulative investment in that experience, that exceeds just about everybody else, including Microsoft and Google, because just think of all the things that those companies are doing, where this is our singular focus. It starts with building the best experience, then that's led us, especially when you think about our viral adoption model, that's pretty different from the traditional top-down, IT-led distribution approach.
We often find that our customers have a ton of organic adoption of Dropbox within a company. The employees have voted with their feet that Dropbox works the best for them because the design, because the existing sharing networks they've established, and they don't want to switch. Finally, and this will be increasingly important, is our open ecosystem approach. When you think of all the investments we've been making in these integrations, and particularly with folks like Google, the fact that you'll be able to store really any kind of cloud content in Dropbox, and in particular Google Docs and Sheets and Slides. We find a lot of our customers are adopting tools from every ecosystem and the best-of-breed tools, and they're increasingly turning to us to be the best place to do that.
We see our open ecosystem and these integrations as a pretty powerful proprietary differentiator over time.
Got it. Okay. That's very helpful. Thank you both.
Yep.
Thank you. Our next question comes from the line of Pat Walravens from JMP. Your question, please.
Great. Thank you, and congratulations. Drew, shifting to sort of an internal focus for a minute, as I read through your employee reviews on Glassdoor, overall, the rating is really good. The word that comes up a lot on the where you could do better side is growing pains. I'm just wondering, what do you think your biggest growing pain is, and how do you address it?
Sure. Well, we've always been a company that's growing really quickly. In some ways, that's a good thing because I feel like if you're not pushing the boundaries or if you're not having some kind of growing pains, you're probably not growing fast enough. That doesn't necessarily mean that growing pains are good. You want to make sure that you're hiring the best people and that you're running the company efficiently and effectively, and that you have a deep bench of leaders that can help scale. For sure, I would say when you look at the maturity of the business, and I think the predictability, those are a lot of the things that I and the management team pay attention to, where we've made big strides, and even in the last couple of years.
I'm sure we'll continue to have growing pains, my focus and what I pay attention to and what I spend time on is making sure that we are always improving the company and resolving those growing pains faster than they appear.
All right. Thanks. If I can add one more, where are we on international, and what's the growth strategy there?
Thank you for that question. As you know, half of our revenue actually comes from international locations, so we're pretty evenly split between what we see as revenue growth here as well as outside. A lot of the international strategy is really dependent on cloud adoption in those different countries. Specifically, if you look at Western Europe, where cloud adoption is pretty significant, we see a lot more growth. Similarly, Australia, which is a cloud-first, mobile-first economy, we see a lot more adoption of Dropbox there. There are countries that are lagging in that adoption curve. We continue to invest in a lot of these markets, and we'll continue to see growth from international.
Thank you.
Thank you. Our final question for today comes from the line of Rishi Jaluria from D.A. Davidson. Your question, please.
Hey, guys. Thanks so much for taking my questions. Thanks for squeezing me in. One for Drew, a quick follow-up for Ajay. Drew, on Dropbox Paper, I know you've talked about how you've grown the features and functionality around that. Can you maybe give us a sense directionally for what portion of your users are using Paper now or have tried it?
Sure. We don't break out specific stats on it right now. We have shared that our users have created millions of Paper docs. I can share that what we pay attention to internally is engagement and levels of adoption, and how it helps drive monetization. Paper has been improving across all those dimensions. It contributes to the overall platform because we see that teams during trials that use Paper convert to a paid subscription rate at twice the rate as folks that don't. They retain better, and Paper also serves as a test lab to help shape our future investments or what we see as the future of work.
It plays a number of roles, and it's relatively early in its evolution, but we've had a lot of early success with it and we're excited by the progress, and companies like Pinterest, you might have seen Saturday Night Live was in our IPO roadshow video. Some pretty cool use cases, and we're really excited about its future.
Okay, great. That's helpful. Then Ajay, more of a housekeeping question, but since we've brought up international, I think about a third of your revenue is denominated in foreign currency. Just wanted to get a sense for, A, if there were any FX impacts within the quarter itself, and B, how we should think about the FX impact on our model from here on out. Thanks.
Sure. Quarter-to-quarter movements in FX rates have a smaller impact on top line for us, and that's just based on how our rev rec model works. The vast majority of our revenue is already on our books as deferred revenue heading into a given quarter at historical FX rates. For us, movements in FX rates have less of an impact in the near term versus the long term. With respect to 2019 and beyond, still a little too early for us to have a clear point of view. We'll have to see where rates move from now into the start of next year, but we'll certainly have more to share on that on our Q1 call.
Got it. Thank you so much.
All right. Well, thank you for joining us today. We really appreciate your support, and we're looking forward to speaking with you again next quarter.
That concludes our call. Thank you for joining us. You may now disconnect.