數據狗 (DDOG) 2019 Q4 法說會逐字稿

Good afternoon. My name is Jerome, and I'll be your conference operator today. At this time, I would like to welcome everyone to the Datadog Q4 2019 Earnings Call. (Operator Instructions). Thank you. Mr. A.J. Ljubich, you may begin your conference.

Thank you, Jerome. Good afternoon, and thank you for joining us today to review Datadog's fourth quarter 2019 financial results, which we announced in our press release issued after the close of market today. Joining me on the call today are Olivier Pomel, Datadog's Co-Founder and CEO; and David Obstler, Datadog's CFO.

During this call, we will make statements related to our business that are forward looking under federal securities laws and are made pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995, including statements related to our future financial performance, including our outlook for the first quarter and for the full year 2020, our strategy, benefits of our products, potential contribution of customers with ARR of $100,000 or greater, R&D and go-to-market investments, expected capital expenditures and the size of our market opportunity. The words anticipate, believe, continue, estimate, expect, intend, will and similar expressions are intended to identify forward-looking statements or similar indications of future expectations.

These statements reflect our views only as of today and not as of any subsequent date. These statements are subject to a variety of risks and uncertainties that could cause actual results to differ materially from expectations. For a discussion of the material risks and other important factors that could affect our actual results, please refer to our quarterly report on Form 10-Q for the quarterly period ended September 30, 2019, filed with the SEC on November 13, 2020 (sic) [2019] which is available on the Investor Relations section of our website. A replay of this call will also be available there for a limited time.

Additional information will be made available in our annual report on Form 10-K for the year ended December 31, 2019, and other filings and reports that we may file from time to time with the SEC. Additional non-GAAP financial measures will be discussed on this conference call. Please refer to the tables in our earnings release, which you can find on the Investor Relations portion of our website for a reconciliation of these measures to their most directly comparable GAAP financial measure. With that, I'd like to turn the call over to Olivier.

Thank you, A.J., and thank you all for joining us today to review our Q4 and full year 2019 results. Q4 was a great finish to what has been a milestone year for us. Results were driven by broad-based strength across customer segments and geographies and supported by the continued traction of our integrated platform.

To summarize Q4, revenue was $114 million, a close to 85% increase year-over-year and above the high end of our guidance. We ended the year with 858 customers with annual run rate or ARR of $100,000 or more, which is an increase of 89% year-over-year.

The majority or about 60% of our customers are now using 2 or more of our products as of the end of the year. As in past quarters, our dollar-based net retention rate was over 130% as customers increase their usage and adapted our newer products. We also continued to be capital efficient, with a free cash flow of $11 million and a [contractual life] that is still around a year or less.

For the full year, we generated revenue of $363 million, an 83% increase year-over-year, which was above the high end of our guidance. And free cash flow was approximately breakeven or around $800,000 for the year. To step back a little bit, our results clearly indicate that the market wants a unified observability platform for dev and ops teams, which is becoming more important than ever but also more difficult to accomplish in a world that is replatforming to cloud and external architectures.

Our performance in 2019 shows Datadog is a winner in this converging world. And our success is driven by our origins as an integration platform that collects many disparate sources of data across teams and silos, by our ubiquity seeing us deployed everywhere and used by everyone and by the efficient adoption we get from landing with our infrastructure monitoring product.

As a result, and today more than ever, Datadog is the monitoring and analytics platform for dev, ops and business users in the cloud age. We provide clarity and actionable insights into applications and IT infrastructure, all in real time, and we do so to enable our customers to deliver greater innovation and consistently provide an exceptional user experience.

With that outlook in mind, let's review our Q4 performance. First of all, we are very pleased with our results. Strength was broad-based, driven by both robust new logo additions as well as continued growth of existing customers. We added approximately 1,000 net new customers in Q4, which is a record for the company and almost twice the number we added in the same quarter a year ago. For the year, approximately 40% of our growth came from new customers, with our new logo ARR demonstrating very strong growth while net retention of existing customers performed just as well. And our platform strategy is clearly resonating, including strong uptake of our newer products.

From an R&D perspective, we accelerated our pace of innovation with multiple exciting developments in Q4. After the launch of Synthetics earlier in the year, we extended our suite of user experience monitoring products with the launch of Real User Monitoring or RUM to monitor the journey of actual users within an application. And continuing on our vision for full-stack observability, we launched Network Performance Monitoring or NPM to monitor network traffic flows across public and private clouds as well as on-premise environments.

We more recently complemented our network monitoring offering with the beta of our SNMP integration, which connects to physical network devices and extends visibility to customers with meaningful on-premise networks. We are pleased with the initial uptake of NPM and RUM, which demonstrates our opportunity to drive additional value for our customers and create future revenue drivers for our business.

In Q4, we also announced Security Monitoring as a first step to apply the power of our platform beyond observability use cases. We envision a future where silos continue to break down beyond dev and ops and extend to security teams, as it becomes clear that securing applications in the cloud world needs to involve all 3. We also believe that by harnessing the massive amounts of data already collected across metrics, traces and logs, we can help our customers better operationalize IT security.

In 2019, Datadog became the first to unify the 3 pillars of observability with the launch of our logs product. In 2019, we believe we have proven ourselves to be a most comprehensive observability platform, and this is evidenced by our customer adoption. As of the end of the year, the majority or about 60% of our customers are using 2 or more products, which is an increase from about 25% a year ago. This means that today, over 6,000 of our customers are using multiple products, which is more than some point solution vendors have today.

We also know that penetration is relatively even across enterprise, mid-market and SMB segments. Additionally, as of the end of the year, approximately 25% of our customers are using all 3 pillars of observability, combining infrastructure, NPM and logs, which is up from 5% a year ago. This is especially impressive considering that our third pillar, logs has been on the market for less than 2 years. And this makes it clear that customers are finding value by adopting our platform in full and benefit from correlations and workflows that can cross boundaries between teams and systems.

Finally, one of the greatest surprises to us this year has been the success of newer products and initial land deals. In 2019, approximately 65% of our new logo deals had 2 or more products, up from only about 25% in 2018. This demonstrates the pent-up demand for our integrated platform and our ability to add value from the very start of a customer relationship. To summarize, we believe we have a very significant opportunity to further expand our product portfolio and increase our wallet share with customers. We are a product-driven company and investing in innovation is a core part of our business strategy.

Now let's move on to the go-to-market, starting with a look at our approach. We have been expanding coverage in both commercial and enterprise channels to capture the opportunity across company sizes. While continuing rapid growth in North America, we are also expanding in new and existing territories internationally. Additionally, we have been expanding in new markets, such as building a government-focused team. And finally, we are investing in the partner channel as announced in January with the launch of the Datadog Partner Network. We are excited about the opportunity with channel partners and plan for continued investment.

Turning to our Q4 sales. We saw strong new logo additions as well as expansion from existing customers. As of the end of the fourth quarter, we had approximately 10,500 customers, up from 7,700 a year ago. As mentioned earlier, we added approximately 1,000 customers in Q4, which is a record for the company. We ended Q4 with 858 customers with an ARR of $100,000 or more, up 89% from a year ago and an increase of more than 130 in Q4. Given that more than 70% of our ARR is generated from customers over $100,000, we expect this cohort of customers to be a large driver of future growth. We also ended the year with 50 customers with ARR of $1 million or more, which is up from 29 a year ago and only 12 2 years ago.

Now let's review some of our key wins in the quarter. First, we had a multimillion-dollar upsell with a global financial information services company. This customer started with Datadog in 2017 and has since increased spend by a factor of 20, adopting all 3 pillars. With Datadog, this customer has unified disparate teams on a single view and reduced tool sprawl caused by point solutions.

Second, we had another 7-figure upsell to a large e-commerce retailer. This customer first came to Datadog in 2017 after experiencing large amounts of false/positive alerts right before Black Friday. Today, they have increased their spend by a factor of 15 in adaptive infrastructure, APM, Synthetics and more recently, Network Performance Monitoring.

Third, we had an exciting land this quarter from a large U.S.-based global airline. As the company embarked on a cloud-first initiative, it realized that its lack of cloud-scale monitoring was causing significant risk. Datadog has enabled an improved time-to-market and a move towards agility. And as a 6-figure initial deal, we see ample room for growth over time.

Fourth, one of Europe's largest national railways had a 7-figure upsell. After using over 20 different monitoring tools, this organization has turned its eyes on Datadog as a single source of truth across metrics, traces and logs in this cloud environment. This deal was also won through a public procurement process, which demonstrates our potential in the public sector.

We can't go over every deal we made but there were many other exciting deals in the 6- or 7-figure range, such as: a traditional U.S. bank; a 100-year-old global beverage company; an Asia-based public sector organization; a manufacturer of industrial and transportation equipment; a Fortune 500 insurance company and a professional services firm with over 50,000 employees. We believe these examples demonstrate that the massive IT replatforming driven by cloud migration is happening globally and through organizations that span all industries. Datadog today is a strategic partner to a diverse set of customers across both cloud-native and traditional enterprises.

As a conclusion, we are in the early stages of what we think is a tremendous market opportunity, which we believe we are well positioned to capture. We have been performing at a very high level and our focus is on doubling down on what has made us successful today. So in 2020, we plan to continue to invest in hiring great engineers and delivering innovation to our existing and new customers. We also remain committed to investing in our go-to-market, expanding our sales capacity globally across all geographies as well as investing in new opportunities such as the partner channel and public sector.

With that, I would like to turn the call over to our Chief Financial Officer, David Obstler. David?

Thanks, Olivier. As mentioned, we are very pleased with our fourth quarter results, which capped off an exceptional year for Datadog. I will now review Q4 results in detail. Revenue was $113.6 million, up 85% year-over-year. As Olivier mentioned, the quarter's strength was broad-based, driven by new and existing customers, success across segments and geos as well as driven by strong platform adoption.

To provide some more context, in Q4, we saw strong new logo additions, with contributions across sales channels and regions. This was a record quarter for new logo ARR, demonstrating very strong growth year-over-year. Additionally, we saw strong continued expansion of existing customers. In the fourth quarter, our dollar-based net retention was above 130% for the tenth consecutive quarter. Our robust retention rate is primarily driven by the increased usage of existing products as well as cross-selling to newer products. We continue to see robust and frictionless expansion as customers continue their cloud migrations. The cross-selling of newer products is a more recent driver of our net retention rate.

Next, we would note a strength across segments with robust growth across enterprise, mid-market and SMB segments. This demonstrates the power of our simple but not simplistic platform, which is powerful enough for the largest of organizations and also easy to deploy and use for the smallest. I'm happy to report that the average ARR of our enterprise customers at the end of 2019 was about $230,000, an increase from $160,000 at the end of 2018. And average ARR of our mid-market customers at the end of 2019 was about $170,000, an increase from $110,000 at the end of 2018. We believe there remains ample room for continued penetration of each of these segments.

Lastly, we had a very strong quarter internationally in Q4, with both EMEA and APAC performing well. Q4 saw record international ARR adds and strong momentum with international growth outpacing that of the aggregate business. I note that many of these teams in the region -- these regions are still ramping.

Turning to calculated billings, defined as revenue plus the sequential change in deferred revenue, which was $130.3 million, up 77% year-over-year. Adjusting for the timing of a large invoice that changed its timing that we discussed last quarter, calculated billings growth would have been 85% in Q4, in line with revenue growth. As discussed previously, we strongly discourage the use of billings as a gauge of momentum as it does not accurately portray the growth of our business. Revenue is a better indicator of growth, given the mix of monthly and annual billing terms amongst our customers and the tight relationship between revenue and ARR.

Now let's review the income statement in more detail. As a reminder, unless otherwise noted, all metrics are non-GAAP. We have provided a reconciliation of GAAP to non-GAAP financials in our earnings release. Gross profit in the quarter was $88.4 million, representing a gross margin of 78%. This compares to a gross margin of 76% last quarter and 75% in the year ago period. Improvement in gross margin was driven by efficient use of our cloud hosting spend. As we have discussed previously, we are focused primarily on product development and innovation as well as the build-out of cloud data centers in newer geographies. Therefore, our gross margins may fluctuate within certain ranges as we prioritize product innovation over resource optimization.

R&D expense was $31.6 million or 28% of revenue, consistent with the year ago period. We continue to benefit from product-led adoption and have made extensive investments in our platform. This is evidenced by the launch of Network and RUM products in Q4 as well as the announcement of Security Monitoring. We continue to see a meaningful opportunity to innovate and expand our platform and, therefore, plan to continue to make investments in R&D.

Sales and marketing expense was $39.3 million or 35% of revenue, down from 46% in the year ago period. The change in Q4 was pronounced partly due to the outperformance on the revenue line. Furthermore, we note the greatest leverage came from marketing expenses, which grew at a lower rate year-over-year versus sales expenses, which grew closer to the rate of revenue. We plan to continue to invest aggressively to expand our go-to-market globally.

G&A expense was $10.4 million or 9% of revenues, slightly higher than 8% a year ago, given some public company-related expenses. Operating income was $7 million or a 6% operating margin compared to an operating loss of $4.3 million and a negative 7% margin in the year ago period. Net income in the quarter was $10.1 million or $0.03 per share on 327 million weighted average diluted shares outstanding. Profitability outperformance was driven primarily by the revenue outperformance.

We have a highly efficient business model and have experienced a high return on our investments in sales and marketing and in R&D. While we have operated around breakeven to slightly profitable and outperformed on profitability in Q4, we see ample opportunities to continue to invest in the large market opportunity ahead of us.

Turning to the balance sheet and cash flow. We ended the quarter with $778 million in cash, cash equivalents, restricted cash and marketable securities. Cash flow from operations was a positive $17.4 million in the quarter and a positive $24.2 million for the full year. After taking into consideration capital expenditures and capitalized software, free cash flow was positive $10.9 million in the quarter and approximately breakeven or around $800,000 for the full year.

I would now like to turn to our outlook for the first quarter and the full year 2020. Beginning with the first quarter, we expect revenue to be in the range of $117 million to $119 million, which represents a year-over-year growth of 68.5% at the midpoint. Non-GAAP loss from operations is expected to be in the range of negative $5 million to negative $7 million. Non-GAAP net loss per share is expected to be negative $0.01 to negative $0.02 per share based on approximately 296 million weighted average shares outstanding.

For the full year 2020, revenue is expected to be in the range of $535 million to $545 million, which represents 49% year-over-year growth at its midpoint. Non-GAAP loss from operations is expected to be in the range of negative $20 million to negative $30 million. Non-GAAP net loss per share is expected to be in the range of negative $0.03 per share to negative $0.07 per share based on approximately 302 million weighted average shares outstanding.

A few things to take into consideration in this guidance. First, while we saw strong efficiencies in our gross margin in the fourth quarter, we continue to focus on product development and diversification of our cloud hosting vendors and regions. Therefore, our gross margins may fluctuate within a certain range as we prioritize innovation over resource optimization. Second, we have some important corporate events in Q1, including our sales kickoffs. Next, our intention remains to invest much of our recent outperformance, including aggressive hiring targets in R&D and sales and marketing.

Then some notes below the operating loss line. We expect approximately $10 million to $13 million of interest income for the full year based on our cash and marketable security investments. Second, we do not expect to be a federal taxpayer but have a tax provision related to our international entities. For the full year, we expect a tax provision of $1 million to $2 million. Note that our share count forecast for Q1 and 2020 reflects our basic share count since we are forecasting a loss in each period.

To summarize, we are very pleased with the business performance in the fourth quarter as well as the full year. We have growth at scale that few can match and have demonstrated efficiency in our model. We are making continued investments for growth for the foreseeable future. We believe we are at the very early stages of a multibillion-dollar market opportunity, and we feel very good about our ability to build a large and successful company over time.

With that, we will now open the call for questions. Operator, let's begin the Q&A.

(Operator Instructions) Your first question comes from the line of Sanjit Singh of Morgan Stanley.

Congrats on the really exceptional 2019, really strong performance across the board. I wanted to follow up on a point, Olivier, that you made in your script about 25% of customers adopting all 3 pillars. Could you give a better sense of sort of the market momentum in terms of going towards these unified platforms like Datadog? Where are you in that journey in terms of the different customer segments, whether it's mid-markets, on which it seems like they're moving pretty aggressively but also sort of the enterprise? And who are you sort of displacing as customers consolidate and standardize more on the Datadog platform?

Yes. So the momentum is the same actually in each and every one of the segments. Whether you talk about SMB, mid-market or enterprise, we see the same momentum. And the -- in all cases, by and large, we're going to enter through net new deployments in cloud environments. So we typically don't displace anything from day 1. What we're going to displace is going to be majoritarily (sic) [majority] homegrown open source systems. That's when we land these multiple customers. Over time, we'll displace -- for some of the customers that have spent more time in cloud environment and already have another pillar with a separate vendor, we'll displace that. But that's not the majority of the deployments we have.

Understood. And as my follow-up, it's sort of related to some of the momentum you're seeing across products. So if I look at the product portfolio in, let's say, 3 buckets: core infrastructure monitoring; two, the products that you've released in 2017, 2018, that would be APM and logging; and then sort of the 2019 cohort, where we're talking about Network Performance Monitoring, RUM, the Real User Monitoring and Synthetics, could you sort of give us a sense about the customer traction across those 3 buckets? They're obviously going to vary, but I just want to get a sense of what's picking up sort of faster than expected and where momentum is still sustaining.

Yes. So what I can tell you is if we single that infrastructure monitoring, which was our initial product, in its own right, it would be a super high-growth SaaS company and could [hold its] own. The APM and logs products are both in hyper growth and have significant scale. We're not disclosing the exact scale, but I can tell you that they are growing faster than that, that grows at a similar scale, the infrastructure product initially.

And then for the other products, the only 1 that's really been in the market for some time is Synthetics. The others, we either just started charging for them or we haven't even yet started charging for them. For example, we haven't started charging for RUM yet, but we started charging for NPM in December. So these deals are very new. We have some exciting stories about them. I mean we mentioned 1 customer in the script that adopted -- 1 large customer that adopted NPM, and we have more stories like that.

But we see, so far, less scale from the newer products, obviously, because they're more recent, but we see very broad adoption. Synthetics has been on the market. They started charging for -- during the summer. Is -- has a large user base and keeps growing. And so we are very hopeful -- we can't tell which of these products are going to grow faster in the end or have the biggest footprint. They're still very early for many of them. But in general, we are very satisfied with the adoption we see. And we believe that there's going to be a lot of long-term growth, thanks to all of these products.

Your next question comes from the line of Chris Merwin of Goldman Sachs.

So I just wanted to ask about the traction you're seeing, in particular, with enterprise accounts. Can you talk about which of your non-infrastructure products, in particular, are resonating with that segment? And what are some of the reasons that you're hearing from those customers about why you're winning relative to some of the competition?

So in enterprise, look, we're going to land, in the majority of cases, with 2 or more of the products, right? The breakdown is the same as we had in the other categories. And it's going to be typically infrastructure and logs or infrastructure and APM. There's no -- not a very notable difference in the attach rates there. Though what we'll see is that the -- there's a bit more friction in adopting APM at scale. So typically, the dollars -- even though they're going to use multiple products, dollar from day 1 are going to be higher for customers that adopt logs than they are for customers that adopt APM. But both are going to grow over time so there might be a longer fuse on the growth in APM, for example.

Great. And then just a quick follow-up on pricing. You obviously came up with a new pricing mechanism for logs that seems to be resonating in the market. And now, I guess, a lot of competitors are trying to replicate the broader products that you've put together. So when you talk to customers, is there a lot of sensitivity to price or are they just very much more focused on the functionality and, therefore, have less sensitivity around how you're charging for the products?

Well, the real sensitivity for customers is not so much on the price itself, but as you know, there's 2 things: one is the value and two is the controls they have over it. So the way we make all of our products -- all of our pricing work is that we want to give all the controls to the customers so they can align what they pay with the value they get for it. So that's what we've done with what you mentioned on the logs side and where we also have versions of that for APM and infrastructure monitoring products. But we want to make sure customers have the controls and don't end up being prisoners of a pricing agreement that they don't like at scale.

Your next question comes from the line of Raimo Lenschow of Barclays.

Congrats from me as well. Olivier, just a quick one on security. Like that obviously brings you into a field that you haven't played in before, and there are some decent vendors out there. What's the investment needed in terms of sales support there? Is that kind of a whole specialist sales force that you need because you start talking to a CISO, et cetera? Or how do I have to think about that?

Yes. So right now, the bulk of the investment we're making is in the R&D and product side so that's where we're starting. We're still very early in security. We're not charging for anything yet there. So we're going to -- it's going to take some time before we see a full ramifications on our go-to-market organization. But I would say in the short term, the signals we get from customers and from the early products we have are that: one, there's a great deal -- a great amount of open space when it comes to cloud environments -- new cloud environments that need to be secured. So customers have a need, they have a gap there and it's net new. And it's -- and the reality is also amenable to being adopted bottom up, which is the rest of what we've done, which is exactly the rest of what we've done.

So initially, the go-to-market is very well aligned with what we've done in the past and we see signs of that being a fit. In the longer term, there might be more changes required and more customization to the sales team. But in general, anyway, we're making changes in customization to the sales team as we have more and more channels to go-to-market, and we cater to customers that are in different geographies, different sizes and that might need different ways to address them.

Okay. Perfect. And then one on the networking side. So historically, the networking monitoring guys were always a little bit different because you had -- when you had to be on the different switches, et cetera. Is that changing with kind of some open source projects, et cetera? Or how is it possible to do it now because it used to be like a separate game?

Well, I mean, there is -- so first of all, we're not going to network monitoring with the objective and ripping and replacing everything that people already use on-prem today. Our starting point is customers that have a significant footprint in public and private clouds and then want to extend that visibility to on-prem network devices. So this already limits the -- I would say the subset of devices we're starting with just by selecting these types of customers.

And then when it comes to integrating with all of this and it's sort of what we do already, right? I mean with our infrastructure monitoring products, integrating with a lot of different devices, battle testing these integrations in the real-world with a lot of small customers testing them, that's a core competency of the company. So that's not going to happen overnight obviously, but we -- that's definitely something we keep investing in.

Your next question comes from the line of Brad Zelnick of Crédit Suisse.

Congratulations on the very impressive results, especially the metrics you've shared around multiproduct adoption. My first question for you, Olivier, I wanted to touch on the competitive dynamic for logging specifically and who you're mainly competing against. Perhaps if you could touch on, what sort of success you're seeing in the enterprise market? And how you compete for business that spans both public and on-premise environment, specifically in logging? That would be very helpful.

So we see -- so today, we see great adoption from the logging product in large enterprises. It's, I would say, on an equal footing, give or take, with the APM product as the second product that's being adopted in this environment. And it's a higher dollar amount typically because there's less friction to submit a lot of data from day 1. It's also growing very fast, again because those environments are growing very fast themselves.

In terms of the competitive dynamics, in most cases, customers don't have anything in those cloud environments to start with. Even though they all have some other logging vendor on-prem, so the most large enterprises are going to have a Splunk. Everybody knows that. And the fact that they have a Splunk on-prem is not a [lost] cycle to them adopting a different platform altogether in cloud environments. So that's what we typically see. We don't try to go and replace the on-prem environment before we are well established in next-generation public and private cloud. That's our strategy.

And maybe just a follow-up to that. In reference to the competitor that you mentioned, have you seen any changes in the market now that they've had pricing changes in effect for a number of months? Does that, in any way, change the pricing dynamic for you and the competitive win rates that you're seeing?

There's no changes that we could feel on our end. The situation is exactly the same as it was in the last quarter we commented on.

Your next question comes from the line of Matt Hedberg of RBC Capital Markets.

This is actually Matt Swanson for Matt. Olivier, you put out a really interesting report a couple of weeks ago on the state of serverless. It really makes kind of a strong case for the growth in the technology. Could you talk a little bit more about how this trend impacts Datadog specifically and then maybe things you can do to invest to be positioned for it?

Yes. So to answer the last part of your question first, so we're definitely investing heavily in supporting serverless. I mean, to us, this is the next stage in a continuum of evolution for the way infrastructure is being packaged and run. To step back a little bit, when you look at what infrastructure was like 15 years ago, we had physical hardware, then we moved on to VMware-type VMs that were very static, didn't go anywhere. Then you had cloud instances that were smaller VMs that could go up and down every second. Then you had containers, they were smaller VMs, who got smaller cloud instances that could go up and down every second -- every millisecond, sorry. And then you had these -- the serverless that are basically -- serverless functions that are basically smaller containers that get -- that go up and down every microsecond.

So all of that is basically 1 big continuum that goes from static -- a few things that are moving very statically to lots and lots and lots of tiny components that keep being -- coming up and down and being recombined in different ways. So the way we see it, there's really a shift that's actually happening in terms of the value to provide that goes from the run time towards the observability and the understanding of that run time.

So we think in the end, this creates more value for us. Our job in the end is to solve the complexity for our customers, make sure that the people understand what the machines are doing and can keep changing what the machines are doing and manage it and manage the experience of it. And the more complexity, the better it is for us. So we think it's a -- it's definitely a development that tilts the market in our favor.

That's really helpful. And then David, it kind of feels like this is another quarter that we'd describe as accidentally profitable. But when we're thinking about investments into 2020, can you just talk about some of those levers you have during quarters or during the year to ramp up investments if you continue to see outperformance on the top line?

Yes. I think that's what we said last time, which is a lot of the outperformance has been going to the bottom line. It has to do with hiring plans. We're planning based on a little more of a conservative case. And then we've been outperforming. The levers to pull are -- to hire more quickly, most of the hiring is in R&D and in sales. And then the second would be to change the marketing dollars to the extent we can do some more marketing or trade shows. Those are the levers we can essentially pull in, in a quarter.

Your next question comes from the line of Sterling Auty of JPMorgan.

So wanted to dig in on the 1,000 net adds in the quarter, a big uptick. And what I'm wondering is, are you seeing an inflection in terms of the gross new customers that are coming in, an improvement in the overall renewal rate of customers or both that's driving that metric?

It's a bit of both. So the renewal rates are generally stable, I would say, trending up softly over time. They're already pretty high so there's not a lot of room for them to go. And we see just in more customers go through the door. As we scale the organization in general, we've seen more customers go through the door.

Yes, I would say the new logo performance in the second half of the year was stronger than the first half. It's related to our development of the sales team and the ramping and it's -- as we said, it's broad-based. For instance, EMEA, which we were developing performed very well in the second half of the year in new logos. And that's because of the ramping and the sort of the build-out that we had been doing last year and the year before.

Got it. And then one follow-up. When you look at a new enterprise logo that comes in and buys all 3 pillars, where are you finding that primary point of contact is in the organization for that type of customer? Are you all the way to the CIO level in an enterprise account? Or are you finding that you're still finding success either in a division or geography and then expanding upwards from there?

Well, typically, it's going to be a cloud initiative, and it's going to be small because it's cloud and it's new. So there's going to be a relatively small team that is driving that. And that team is going to -- because it's new and because it's the future, that team is going to be connected to the leadership team in general. So sometimes, the CIO; sometimes, some other people there. And that team is going to have purview of the whole environment to start with, which is why it's relatively straightforward for us to enter with multiple products at once.

Your next question comes from the line of Brent Thill of Jefferies.

This is Parthiv on for Brent. You talked about strong net new ARR adds on the international front. Just any commentary on whether -- on how either purchasing patterns, sales cycles, competition sort of compare from what you're seeing here in North America and Europe?

The main comment is that we really see the cloud migration happening in the rest of the world. It's something that is starting to happen at scale. This is combined with the fact that we installed sales teams and started ramping them in various parts of the world, in EMEA 2 years ago and in APAC last year. The combination of that is what drives the success we're seeing there. I will say it's still early, those teams are still ramping. There's still obviously things that will take more time to work well reliably but we've seen some great signs. And in Q4, we had very strong performance in EMEA and APAC. Great logos and great successes there, both on the mid-market side and on the enterprise side. So we're very, very optimistic in the future, but it's still early.

Got it. Okay. And on the landing ARR side, are you seeing anything meaningful with the cohorts in the fourth quarter relative to prior quarters, either greater infrastructure usage on landing or attach of multiple products when the initial deal is signed?

We see a bigger attach of more products.

Which is causing a larger land slowly but surely, but -- and it's just been progressive throughout the year as the attach rate of additional products has gotten larger.

Your next question comes from the line of Brad Reback of Stifel.

Great. 2 quick ones for me. I think during the fourth quarter, you began the FedRAMP process. Any update on where you are there? And how should we think about potential upside from that?

So not yet. We'll dutifully file a press release once we have the outcome of it. There's going to be upside at some point, I would say, not in the very short term. We still have to fully build out the teams there, and there's number of -- there are more things we need to do but we're investing.

Got it. Okay. And David, on your last comment there a moment ago about larger lands, is there any risk in the short term that with bigger lands, you start to see less net dollar expansion because of the bigger starting point?

We said that's one of the reasons. We said that we wanted to be prudent in our guidance long term, but we haven't seen that today. We've seen, as Oli mentioned, that net retention be very stable and strong in the second half of the year.

Our next question comes from the line of Michael Turits of Raymond James.

This is actually Robert Majek. I believe you mentioned that you haven't started charging yet for some of the new products. Should we expect a high level of pent-up demand next quarter? And has that been fully built into your Q1 guide?

So in general, we had a -- this summer, we had a lot of pent-up demand from Synthetics because we actually had customers using it for 5 months before we started charging for it. I think in this case, we'll start charging quicker. So there probably would be a bit less pent-up demand and the products might have also a bit more friction than Synthetics, which we found to be a very low friction product so very easy for tons and tons of customers to get started with it.

At this point, our expectations for these products are baked into guidance. But as we multiply the products, obviously, some of them are going to be bigger hits than others. So in the long run, we're optimistic that we'll get a lot of upside. But in the short term, they're going to have a relatively small impact.

Your next question comes from the line of Jack Andrews of Needham.

I was just wondering, I mean, just given the proliferation of new products that you're introducing, how would you describe your sales force, the maturity and their ability to understand and appropriately sell these products?

Yes, that's a great question. So that's also a big area of investment for us in enabling the sales force and making sure that everybody knows everything we can sell and how now. In most situations though, when we land a customer, which is where the bulk of the -- I would say, the hard part of the sales job is taking place, we land with the marquee infrastructure monitoring, which is very, very, very -- which is a great fit for the pain that customers are feeling from day 1 when they deploy on the cloud. And then logs and APM, that customers also understand very well and fully expect now to be part of the same platform. So there's not a lot of challenge there in terms of training the sales force and making sure we can present everything in a way that makes sense to the customer.

Now as we grow with customers over time, and as we put more products in front of them and make sure that they know they can also use Real User Monitoring or Network Performance Monitoring or Synthetics and others, I think there's definitely more investment we need to do there to make sure that we can fully capitalize on that. So that's 1 of the levers for growth we have and 1 of the things we're investing in.

Yes. And as we mentioned, I mean, we're pretty much selling a platform. So you can tell that from the attach rates. And so in that case, the salesperson is able to sell that platform, that's what the client wants. We also augment the salesperson with sales engineers who are expert across the platform and in products. So they're not out there alone. They're being helped by some technical presales.

But at the end of the day, the adoption for all of these products is mostly frictionless. So we see, for example, we see adoption dynamics that are fairly similar between large enterprises that are higher touch with enterprise sales and sales engineers and SMB mid-market that are going to be lower touch and handled by customer success teams. So these -- so that customers can find their way and adopt these new products. There is still probably some upside if we can figure out the best way to enable the customer success teams and the sales teams to go and proactively present new products to customers.

Great. I appreciate the color. And just as a follow-up to that, could you expand a little bit more on the new partner network? I mean which channel are you most excited about? And who should we be keeping an eye on in terms of partners who could really move the needle for you?

Well, the -- it's very broad coverage and it depends in different parts of the world. In the U.S., it's going to be mostly SIs, either local SIs or global SIs. In other parts of the world, you're going to find -- there's going to be more importance to VARs in Europe. In Asia, you'll have a lot more channel, for example. So it really depends on the part of the world. The program as it is designed has broad appeal and then we have teams on the ground that are going to leverage this program to go and recruit very specific types of channels and resellers.

Our next question comes from the line of Bhavan Suri of William Blair.

Let me echo my congratulations. I guess to start with something much more strategic here. When you think about the platform as a whole, there is an underlying sort of data elements, right? So we're capturing all the data. And then there's a next step, which is, "Hey, what are we going to do?" When we think about DevOps and one of the challenges with DevOps is integrating dev and ops or CI/CD together. And as you think about the data element and capturing all the data, one could equate in a relational database there's always -- there's fields, there's all that management. I guess I'd love to understand how you guys think about approaching a data opportunity to really delve into more of the ops side and the dev side, right? So as I develop software, trying to do continuous integration and continuous deployment, how do you guys think about sort of how that plays out? And how your product road map flows into that over, say, the next 3 to 5 years?

So our product actually -- we're in the business of solving the problem of complexity for our customers. And the problem of complexity doesn't start or stop at the infrastructure or at the logs or what's happening when you trace an application in production. It actually covers everything wall to wall between the source code the engineers are writing, the output of the application for the business and all the various remediation tasks that need to take place to keep it going in production.

So as time goes by, we extend the footprint of that platform to really cover as much as possible of that space. So for example, we added -- we announced a profiling products that is very interesting because it's another step towards bridging the gap between the code developers write and what's actually running in production. And we had a fairly differentiated approach to it. We're actually very excited about it, even though it's also another product we aren't charging for yet. So we do see the value of having as broad coverage as possible end to end between the -- all the way between the developers and what the machines are doing.

I think that's helpful. I'd love to take just one so to just figure out, what bridges that gap? You guys obviously are a player. There are other players in the market. But at some point, the gap still exists between that idea of development and integration and implementation. And lots of people are doing real-time CI/CD, but it feels like given the data and the model you have, that gap should be easier for you to bridge than anyone else. We'll be just happy for you to respond to that or not.

Yes, that's the idea. I mean that's why we think -- that's why we're so optimistic about the future. That's why we keep investing. That's why we think there's a very large opportunity because when you think of the -- what makes us valuable is that because we started from the infrastructure we deployed everywhere, not just in the small subset of machines or environments we deployed everywhere. We need to see it to tell you if it's working, therefore, we deployed everywhere.

And then because of the way we designed the product to bridge the gap between different teams and be very easy to adopt, we're used by every single engineer every day. And that combination of deployed everywhere, used by everyone, is what gives us the subset of contact with the customer so that we can actually keep filling the gaps and add stuff in the middle and cover more and more of the problem space. So if you see this, as you mentioned, there's tons and tons to do. I mean it's a very large problem space, which means there's a lot of value we can provide.

Got it. Got it. I mean I have a quick follow-up there. Just more tactical now on the vertical side. Obviously, verticalization matters. It matters probably less in your space, I'd love to understand how you guys think about verticalization and how you approach the vertical markets.

Yes, I think we're not there yet in terms of verticalization. Right now, the opportunity as far as we're concerned is very horizontal. I think maybe when we're further along in the cloud migration, that's something that might make more sense. Maybe also when we expand into different product categories, that may be a little more specific to what the -- our customers are doing with the data. That's also something that might make sense. But today, with our current set of products, we're not really verticalizing.

Your next question comes from the line of Pat Walravens of JMP Securities.

And congratulations. So Olivier and David, I'd love to hear your perspective, each of you, on the same question, which is, I mean, 85% growth and working on so many things, which has come up in all the questions on this call. I mean, Olivier, what's sort of your #1 focus right now? And David, same question for you in terms of where you're spending your focus and your time.

All right. So I think we have 2 focuses. Yes. So focus number one, hiring enough of the right salespeople, enabling them everywhere in the world. And that's the best predictor of success 1 year from now. And focus number two, hiring enough of the best engineers and training them and retaining them, which is the best predictor of success 2 years from now. And that's -- if we can get all that done, we'll be fine.

Yes. And my focus is on providing scalability and the infrastructure and the transparency of information to support that, whether that be in the hiring process, the information around sales or products, et cetera, and support that, so we can see what's going on and make good business decisions.

Your next question comes from the line of Yun Kim of Rosenblatt Securities.

So congrats on that very strong impressive results. Echoing Brad's earlier remarks and others, one of the most impressive trend within your results is that you are able to land initial deals with more than your core infrastructure product. Obviously, that's what you guys are known for. And then obviously, there's also a fairly high percentage of customers already adopting multiple products, which is obviously again, very impressive. Just trying to better understand, is there a concerted sales effort, perhaps additional incentives or the sales process to push for this multiproduct adoption, especially on the initial land deals? Or is it just simply customers asking for them from the start? Again, just trying to better understand what's driving this very impressive multiproduct adoption trend.

No, there's no incentives for it. The incentive is that customers want it more if we have more of the products. So it's easier to sell.

Yes, the platform, single pane of glass, observability platform is being led by the clients. As we talked about, we're looking at clients first in where we are product strategy and we're following that. And it's very much pulling us in...

But the compensation is completely [useful] to it.

For the sales teams, yes.

That's great. Yes. That's great. So it's definitely market driven, which is a validation of your product and technology out there. David, real quick, deferred contract cost was up quite a lot, which is obviously a good sign. I'm just curious if there was a -- if there is any kind of special commission payout for certain products or on higher payoffs on initial land deals just because it seems like that performed well in the quarter. Just trying to better understand the strong uptick in deferred contract costs beyond strong sales execution and perhaps some accelerators kicking in.

Yes, there's no special deals or anything different in the commission plan. It is purely the sales performance and those getting into accelerators. So it's all related to the sales performance.

I would add, if anybody is listening to this call and is in sales, we have high expenses because of accelerators with a large commission, so feel free to apply.

There are no further questions at this time. Please continue, Olivier.

All right. Thank you. So in closing, I'd like to repeat that we are very pleased with results for Q4 and for 2019 as a whole. We are achieving growth at scale few companies can match. We have proven the efficiency of our model, and we have demonstrated our platform traction. That being said, we still feel that we're just getting started. We have a tremendous opportunity ahead of us and have many ambitious goals to execute on in 2020 and beyond.

With that, I'd like to thank all Datadog customers for their trust and, of course, all Datadog employees for their hard work, dedication and success to what has been a fantastic year. Together, we have accomplished great things, but I believe the best is yet to come. Thank you.

Thank you.

This concludes today's conference call. You may now disconnect. Thank you.