Datadog (DDOG) Q2 Recap

Datadog announced Q2 earnings on August 6th. The results topped expectations for both revenue and earnings. Additionally, they raised Q3 and full year revenue estimates, but not by as wide a margin as in past quarterly reports. The stock dropped by almost 20% the next day, after a nice run-up in advance. During the earnings call, the leadership team discussed a few adverse effects as large customers optimized their usage in reaction to the macro environment. On the flip side, customer adds and DBNER continued at a high rate, providing confidence that Datadog can sustain growth after the COVID-19 headwinds clear. In this post, I review Datadog’s Q2 earnings and dig into the many product updates that occurred over the last several months. I also revisit the competitive landscape and examine how Datadog’s extension into new categories further expands their addressable market. For more background on Datadog, readers can review my prior quarterly recaps and initial analysis.

Headline Financial Results

• Q2 2020 Revenue was $140.0M, up 68% year/year. This compares to the consensus estimate for $135.4M, representing growth of about 62.7%, and the company’s prior guidance of $134-136M. Q1 Revenue growth was 87%.
• Q2 Non-GAAP EPS was $0.05 vs. $0.01 expected, representing a beat of $0.04. This compares to ($0.07) in Q2 2019. The company’s original estimate from Q1 was for EPS in the range of $0.00 to $0.01. Q1 2020 EPS was $0.06.
• Q2 Non-GAAP operating income was $15.3M, representing an operating margin of 10.9%. This compares to an operating loss of $5.5M in Q2 2019, representing an operating margin of -6.6%. Q1 operating margin was 12.3%.
• Q2 FCF was $18.6M, representing a FCF margin of 13.3%. FCF margin was -6.6% in the year ago period and 14.7% in Q1 2020.
• Q3 Revenue estimate of $143-145M, representing growth of 50.1% year/year at the midpoint. This compares to the consensus revenue estimate of $140.2M, or 46.2%. The raise was about 4% this quarter, versus a raise on next quarter revenue of about 10% as part of the Q1 report.
• Q3 Non-GAAP EPS estimate of $0.00 – $0.01. This compares to ($0.01) expected.
• Q3 estimated Non-GAAP operating income of ($1M) – $1M, for an operating margin of 0% at the midpoint.
• FY 2020 Revenue range of $566-572M, representing growth of 56.8% over FY 2019 at the midpoint. This compares to consensus revenue estimate of $563.6M or 55.3% growth. The company’s prior guidance from Q1 was $555-565M. Datadog raised full year revenue guidance by about 2% this quarter, versus a 7% raise of full year revenue estimates in the Q1 earnings report.
• FY 2020 EPS estimate of $0.11 – $0.13 versus consensus estimate of $0.05 and the company’s prior guidance from Q1 of $0.02 – $0.06.
• FY 2020 estimated Non-GAAP operating income of $28M – $34M, for an operating margin of 5.4% at the midpoint. Operating income estimates were raised from $0M – $10M in Q1.
• Ended the quarter with cash, cash equivalents and marketable securities of $1.5B. This includes $641M of proceeds from a convertible note issued during the quarter.

Other Performance Indicators

• Q2 Billings were $160.1M and up 62% year-over-year, relatively in line with revenue growth.
• RPO at Q2 end was $287M, up 53% year-over-year. Datadog did not see a material change in billings durations in Q2.
• Q2 Non-GAAP gross margin of 80% versus 75% in Q2 2019 and 80% in Q1 2020. Year-over-year improvement of gross margin was driven by efficiencies in cloud hosting.
• Breaking down Q2 Non-GAAP expenses by category, we see year/year reductions in relative percentage of revenue across all three spending categories. Management commented on the call that they are increasing spend in these areas, but revenue outperformance is greater. This reflects an operating model with significant leverage.
  • R&D = 27% (versus 30% in Q2 2019)
  • S&M = 33% (42% in Q2 2019)
  • G&A = 9% (9% in Q2 2019)

Customer Activity

• At end of Q2, Datadog had 1,015 customers with ARR of $100k or more. This represents an increase of 71% from 594 at end of Q2 2019. At end of Q1, Datadog had 960 customers with greater than $100k of ARR, generating sequential growth of 5.7%. In Q1 2019, Datadog had 508 of these large customers, for annualized growth of 89%.
• Customers with ARR greater than $100k made up 76% of total ARR in Q2, up from 75% in Q1 and 72% in Q2 2019.
• At end of Q2, Datadog had 12,100 total customers. This compares to 8,800 at end of Q2 2019, for annualized growth of 37.5%. Datadog ended Q1 2020 with 11,500 total customers, representing sequential customer growth of 600 customers or 5.2%. This compares to Q1 customer growth metrics of 40.2% annually and 8.7% sequentially. In Q1, Datadog management highlighted the large addition of 1,000 new customers.
• Dollar-based net retention rate was over 130%, consistent with the past 12 quarters.
• As of June 30, 2020, approximately 68% of customers were using more than one product, up from approximately 40% a year earlier. These metrics indicate strong momentum in the uptake of newer platform products.
• Approximately 75% of new logos landed with two more products. Over 15% of customers are now using four or more products. That percentage was 0 in the year ago period. 
• Management highlighted the significant international opportunity remaining in the 10-K. Revenue, as determined by the billing address of customers, from regions outside of North America was 24% for the six months ended June 30, 2020, compared to 25% for the six months ended June 30, 2019.

On the earnings call, Datadog leadership highlighted several customer wins:

• A few notable new logo wins in travel and education, including two global hotel chains, an amusement park chain, a large U.S. university and a European airline. These wins show that even in the face of challenging times for these customers, investing in their migration to digital experiences is a top priority.
• A seven-figure upsell with a large FinTech company. This customer has been able to move from multiple disparate monitoring tools to using a single platform for all three pillars of observability with Datadog. This allowed them to refocus engineering resources on building new features. The customer expects more than $1M in savings from consolidating their monitoring and logging vendors onto Datadog. 
• Another seven-figure expansion came from a European automotive company, which is modernizing their infrastructure and migrating to Microsoft Azure. Through the adoption of Datadog infrastructure monitoring, APM and NPM, their teams are now collaborating on a shared platform and are moving to an increasingly agile development model.
• A large entertainment platform committed to over $10M in spend for the year. This company has made the decision to increase investment in observability and broader use of Datadog both with new products and by scaling up on existing products.
• A high six-figure customer land with a leading asset manager, which is now using Datadog for infrastructure monitoring, APM and logs, as well as Synthetics and early adoption of Security. 
• A six-figure upsell to a seven-figure ARR with a social networking platform that has seen tremendous growth during the pandemic. At record levels of scale, they use Datadog to quickly drill down into any failed request and easily identify contributing systems for cause. This company is now using all three observability pillars, as well as Synthetics, RUM and NPM, and has standardized their monitoring on Datadog.

Analyst Reactions

Following Datadog’s Q2 earnings results, 6 analysts provided updated coverage ratings. Of these updated ratings, 5 analysts raised their price targets. Two analysts rated the stock at a Buy equivalent and four gave a Hold rating. The average price target for these updates is $88.50, representing a 17% increase from the closing price after earnings of nearly $75 on August 7th.

Date	Analyst	Rating	Price Target
8/7	Needham	Buy	Raised from $105 to $106
8/7	Jeffries	Hold	Raised from $59 to $95
8/7	DA Davidson	Neutral	Raised from $52 to $80
8/7	Credit Suisse	Neutral	Raised from $55 to $80
8/7	RBC	Sector Perform	Lowered from $85 to $80
8/7	JP Morgan	Overweight	Raised from $65 to $90

After the earnings results, Jefferies set one of the highest price targets, raising from $59 to $95. Analyst Brent Thill provided this commentary.

Jefferies analyst Brent Thill raised the firm’s price target on Datadog to $95 from $59 and keeps a Hold rating on the shares. He notes the company’s 68% revenue growth represents a deceleration, but still topped the Street consensus and the company’s guidance. He sees the quarter as a “temporary setback,” but believes the stock is fully valued at current levels, Thill added.

TheFly.com, August 7, 2020

Product Development Activity

As we have come to expect, Datadog made significant progress on product development over the last several months. Various improvements were announced through the quarter and highlighted on the Q2 earnings call. Additionally, following earnings, Datadog held its annual user conference, Dash. This kicked off with a number of announcements for new product offerings and extensions to existing ones.

Along with their Q2 earnings results, Datadog announced that they had acquired Undefined Labs, which provides testing and observability capabilities for developer workflows before pushing to production. As Datadog’s products have traditionally been focused on the production environment, this acquisition extends their visibility into pre-production development and test cycles. Undefined Labs’ Scope tool is integrated into existing CI/CD platforms, like CircleCI and Jenkins, to enable developers to automatically take advantage of monitoring and testing capabilities within their existing workflows. Scope allows developers to execute unit tests, measure performance, tie test failures back to source code and view summarized results in a consolidated dashboard.

This acquisition aligns Datadog more closely with developers during the design and test phases of software development, as opposed to the DevOps teams responsible for managing the production environment. Undefined Labs tools are leveraged by developers when they are creating the first pass of their code and testing it locally on their own machines. This is often before the developer even merges their code with the main branch in their team’s central repository. Because of this, Datadog will have access to all code changes and application behaviors from initial code creation to the finalized form for deployment. This will yield a higher level of insight into the full development process for an engineering team and could provide a launching pad for other developer productivity offerings in the future. This tie-in between progressive code changes through the development and test processes to the final form of the code in production will represent a unique data input for the future Datadog platform. Combined with production code performance through profiling tools will yield valuable insights that Datadog could share with developers early in the code design process.

The plan for absorbing the Undefined Labs product set is to sunset their existing products and rebuild them on the Datadog platform, so that full integration is realized off the bat. The engineering team is actively engaged in this work now. This approach is smart in my opinion. Rather than trying to maintain a near term revenue bump by keeping Undefined Labs products in the market, Datadog is looking at the bigger picture opportunity to extend the Datadog platform into the full lifecycle of development. This is an important consideration for investors, as developer products for pre-production workflows have traditionally been the domain of Atlassian (BitBucket, Bamboo), Microsoft (GitHub) and GitLab. There is even a little overlap with aspects of JFrog’s platform (upcoming IPO). However, Datadog’s move is primarily focused on extending performance monitoring, test execution and compliance into the pre-production environment. They have made it clear they don’t intend to roll a code repository or CI/CD toolset. Connecting pre-production with production represents a smart strategy by Datadog and provides significant value for engineering teams, as it combines DevOps with developer activities. I could see future product extensions along this vector, which would logically be easier from the full Datadog platform. Investors should pay attention to future developments in this area, as it represents a significant advantage over other observability vendors.

On August 19th, Datadog announced that it had achieved FedRAMP moderate impact “in process” status for SaaS. Datadog is now fully available in the FedRAMP marketplace. This achievement enables Datadog to address U.S. federal government departments and agencies, who can use their SaaS platform for observability of their cloud applications. This follows Datadog’s recent announcement of low impact status.

Here is a list of other major product accomplishments during the quarter:

• Launched the general availability of Private Locations for Synthetic Monitoring. Datadog’s traditional monitoring products are focused on public-facing internet applications. Yet, many companies have large scale internal applications that serve a controlled audience. These could be for employees or partners, where access is managed on a private network. The availability and performance of these applications is just as important as for public web sites. Datadog’s new offering allows companies to set up synthetic monitoring for these internal applications with the same capabilities as the public facing product. Customers can easily configure usability tests and alert DevOps personnel when a performance issue is encountered.
• General availability of the Datadog mobile app. This provides access to all Datadog monitoring dashboards through a mobile app available on both iOS and Android. It is integrated with popular on-call notification tools, like PagerDuty and OpsGenie, so that clicking on the alert notification takes the user directly to the associated dashboard displaying the issue. The user can then examine the details of the alert, zoom in and out on graphs and easily transition to other views, all with standard mobile-friendly gestures. Once the user has assessed the issue, they can silence the alert, share their findings with others or open a new channel for collaboration. This is very useful for on-call personnel, allowing them to confidently respond to alerts without immediately having to pull out their laptop or return home.
• Support for streaming log data directly from Amazon Kinesis Data Firehose to Datadog. Amazon Kinesis Data Firehose receives logs from services such as Amazon CloudWatch, API Gateway, Lambda and EC2, and then routes this data to third-party tools and systems. Datadog is now able to receive this log data directly to enable analysis within the Datadog toolset. Customers can tail the live data, apply standard filters, conduct automatic parsing and enrichment, generate metrics from it and correlate the data to active monitors for other infrastructure.
• The preview release of the Datadog IoT agent to deliver visibility into Internet of Things devices. This provides a lightweight version of the standard Datadog agent that can collect over 100 health metrics and application logs. The agent can be installed on the most popular IoT device operating systems, like x86 and ARM v7/v8 processors. Once installed, the agent will collect metrics and logs and stream them back to the Datadog console for analysis. Operators can add tags to various device types, so that they can aggregate data across different IoT deployments (like device type, location, etc.). As with any datadog metrics, operators can set up custom dashboards to create a single view with graphs of all relevant metrics and log events, supporting standard drill down for more granular analysis. This provides a useful complete overview of an entire IoT infrastructure, spanning the devices, management services (like Azure IoT Hub) and back-end applications.
• Datadog achieved AWS Lambda Ready designation. This designation demonstrates that Datadog has completed a deep integration with AWS Lambda. This extends Datadog’s monitoring reach into serverless. Serverless compute provides the benefit to developers of removing the need to provision stand-alone servers or containers to run their code. Serverless applications are generally organized around Functions, which can run independently and focus on a specific set of operations. With this designation, Datadog monitors can return important signals to developers about the performance of their serverless functions on AWS, including execution times, errors and frequency of execution.

In addition to these enhancements launched during the quarter, Datadog announced several new product offerings at their annual Dash conference on August 11. These represented significant additions, and a few will generate incremental revenue streams:

• Incident Management. This new product streamlines on-call response workflows for DevOps teams by unifying alerting data, documentation, and collaboration into a centralized view. Incident response benefits from being viewed alongside the actual observability data associated with the service issue in Datadog’s interface. To facilitate on-call communications, Datadog’s incident management solution includes integration with their mobile app and a dedicated issue response Slack channel. The feature makes it easy to declare an incident as well, right from the Datadog interface. This product is being launched as a beta.
• Marketplace. This extends the existing Datadog Partner Network to provide a central marketplace where partners can list and sell their custom integrations built on top of the Datadog platform. Datadog users can view these offerings within the Integrations tab of the Datadog application interface. If they find an integration of interest, they can instantly activate it. Datadog handles all billing, so the process for the user is seamless. Benefits to Datadog include increased usage of the platform and a share of the partner’s revenue.
• Continuous Profiler. This product captures resource utilization (CPU, memory, etc.) at a very granular level in the production environment, down to the line of code. This provides very useful feedback for developers. First, if an availability issue arises, the developer can use this tool to determine if a particular function or behavior within the code is causing the issue by consuming too many resources. Second, DevOps can use this tool to monitor cloud resource utilization in general, as a means to identify expensive operations and forward those observations to developers for optimization. Continuous profiler is being launched as a full product with pricing. This represents an incremental revenue stream for Datadog. It also provides a strong complement to the new code-level insights during the entire development cycle to be gained from the Undefined Labs acquisition.
• Compliance Monitoring. This new service monitors production environments for misconfigurations that could cause service or compliance issues. Cloud resources, like security groups, load balancers and storage buckets, all have configurations that require extensive customization for the specific architecture of the user’s environment. These resources can be misconfigured, or worse, left in default settings that introduce vulnerability risks. These risks can manifest as security incidents or put a company out of compliance with standards like PCI (credit card processing). Datadog’s compliance monitoring service continuously examines production environments for these issues and notifies DevOps personnel when one is introduced. The service is being rolled out as a beta program.
• Error Tracking. This feature is being tucked into the Real User Monitoring product. It monitors for JavaScript errors within browsers and groups those into logical issues for investigation by developers. The feature intelligently analyzes each error to indicate when it started, which code is associated with it and most importantly, what may be causing it. Since error tracking leverages the existing integration with RUM, customers don’t need to install an additional agent or SDK to utilize the capability.

Overall, I think these additions represent thoughtful extensions to Datadog’s product suite. Continuous Profiler was launched as a new stand-alone product with a separate revenue stream. As beta offerings, both Incident Management and Compliance Monitoring should eventually transition into paid products as well. The Marketplace is a standard offering that other SaaS vendors have adopted with success and it is great to see Datadog tapping into this potential. I think all of these product additions will help further bolster Datadog’s growth in the future and continue to expand its reach. They are also indicative of Datadog’s continued rapid product development cadence.

One separate consideration around these announcements is for investors in PagerDuty (PD). Datadog’s new Incident Management product appears to have overlap with aspects of PagerDuty’s Incident Response offering. While Datadog’s Incident Management product description includes reference to integration with other on-call notification services like PagerDuty and OpsGenie, that service integration only refers to PagerDuty’s core on-call management and notifications product. As PagerDuty investors may know, the company started in on-call notifications, and layered additional workflow and visibility products around this over time to grow their reach. Incident Response has been a source of incremental revenue for PagerDuty beyond basic on-call management. Datadog’s move into this space is a logical extension for them, as most of the incidents for a customer organization would be associated with an event surfaced by one of Datadog’s observability tools. Given that Datadog already has the context for the incident in its interface, including incident management in the same toolset provides greater efficiency for the DevOps team that responds. Looking at the diagram below of PagerDuty’s platform flows from their web site, one could inject the Datadog logo in place of PagerDuty’s and it would still appear accurate.

While PagerDuty has faced competition before from other vectors, like Atlassian’s OpsGenie and ServiceNow, I would posit that Datadog’s move into this space is potentially more disruptive, as it is based on the foundation of observability, rather than tools for workflow and ticketing management. Coupled with Datadog’s other moves into developer workflow, I think we will likely see more expansion by Datadog into the broader ecosystem of tooling around managing service uptime and response.

Competitive Market Activity

Datadog’s evolution through product offerings and target markets has been fascinating to watch. At the core is a commitment to enable collaboration and drive efficiencies for development, operations and now security teams. Datadog has a product-oriented culture that values engineering prowess and rapid delivery iterations. This has enabled them to quickly respond to customer feedback and pursue new opportunities.

As part of my review of Q1 2020 earnings results for Datadog and the original investment analysis, I conducted a very detailed comparison of product offerings with competitors. Interested investors can reference those blog posts for additional detail.

Datadog started with infrastructure monitoring in 2012, which was largely left to open source solutions at the time. Other commercial offerings in the monitoring space focused on application performance (New Relic) and log analysis (Splunk, Elastic). As part of the migration to the cloud, micro-services and virtualized containers began proliferating. This made monitoring exponentially more complex, as the surface area of discrete software components to track increased rapidly. This sprawl applied to infrastructure monitoring as much as APM and logging.

Datadog had the benefit of starting as these trends were accelerating, allowing them to design extensibility, flexibility and enormous scale into their architecture from the beginning. They tackled ephemeral cloud instances, containerization and micro-services upfront, which laid the groundwork for their future expansion into other areas. Competitive offerings remained tied to a model based on physical servers for a while, before pivoting towards support for more modern deployments.

The other advantage of starting with infrastructure monitoring for Datadog, versus APM and logging, was that their agents needed to be deployed on just about every device in the data center or cloud instance. Engineering organizations will generally monitor every component of infrastructure for basic availability, while not every infrastructure component needs an APM agent or log analysis. This is an important consideration. At the time, New Relic would generally be deployed only on application servers, as that is where the traces were generated. Similarly, due to pricing by log volume, only application and other high value logs would be forwarded to Splunk for ingestion.

From this foundation in infrastructure monitoring, Datadog rapidly expanded their reach into other aspects of application monitoring. They released an APM solution in 2017 and added log analysis in 2018. Datadog’s approach to log analysis was revolutionary, as they introduced the concept of “Logging without Limits“. This refers to the idea that all log data can be ingested and stored by the Datadog solution, with just a portion designated for persistence and detailed analysis.

After addressing the “three pillars of observability“, Datadog continued expanding. In 2019, they released solutions for monitoring the actual user experience, including synthetics and real user monitoring (RUM). Synthetics enables simulation of user interactions with a web application at the UI level, to check that full user paths are functioning as expected. RUM allows for real user interactions with a company’s software application to be captured. These might be actions that have business value, like clicking on a play button, conducting a product search or adding an item to a shopping cart.

All of these solutions run on top of a single, unified Datadog platform. The base layer is data ingestion. Early on, Datadog built a flexible data model that could represent a broad set of data types. This data model is used to store inputs from all kinds of sources, like logs, performance metrics, user activity and application traces. It is necessarily extensible. The data structure is also efficient to minimize storage space. It is designed to be very fast and scalable, as it needs to accommodate enormous data flows, yet be able to support real-time log tailing and dashboard updates.

The common application components of the platform allow for sharing of functionality across the different data types, whether metrics, logs or traces. These components include search, visualization, analysis, alerting and collaboration. On top of all data flows, Datadog inserts a machine learning layer. This enables the system to identify common ranges and patterns for monitored data, so that abnormal behavior can be quickly flagged for operators.

After establishing their leadership in observability in 2019, Datadog has expanded into other areas. These have included network monitoring (NPM), security monitoring (SIEM), and most recently continuous profiler, incident management and monitoring injected into the development/test pipeline. These are all enabled by Datadog’s flexible and scalable data processing platform.

The obvious driver for these expansions is to continue adding more revenue streams to support future growth. Datadog currently has 9 different products that are monetized. To this, will be added compliance monitoring, incident management and some portion of revenue from the Marketplace.

Beyond revenue streams, Datadog is approaching the point where we can make the competitive advantage case for a single vendor “platform” approach to software delivery monitoring. If Datadog can check all the boxes around various aspects of monitoring and observability, for both production and development environments, then they can press the single vendor value proposition with customers. This certainly provides tangible benefits, like a single bill, simplified deployment of one monitoring agent, one set of documentation, one set of dashboards, etc. We saw evidence for the benefit of this approach with at least one large customer in Q2, as highlighted on the earnings call with the 7-figure upsell for a large FinTech company that wanted to consolidate various monitoring vendor relationships into one with Datadog.

It is this race to provide a broad menu of monitoring services that I think will represent Datadog’s competitive advantage in the future. They are leveraging their platform and engineering-led culture of rapid product development to continue to lap their traditional competitors in observability. Specifically, by offering security monitoring, Datadog is capitalizing on the emerging theme of DevSecOps. Datadog first leveraged the DevOps movement, which essentially encouraged closer collaboration between developers (responsible for writing the code) and operators (responsible for system uptime). With DevSecOps, modern engineering organizations are taking this one step further, recognizing that security can no longer be a separate function from development and operations. These three disciplines need to be considered in parallel throughout the application design, build and deployment processes.

By extending their core DevOps toolset to security, Datadog is enabling all three teams to monitor application performance and security from one set of dashboards and data streams. By leaving out security, competitors are forcing engineering organizations to contract separately for security monitoring, which will be increasingly difficult to justify, when vendors like Datadog (plus Splunk and Elastic) offer a full set of monitoring tools.

Specifically, Dynatrace and New Relic do not provide solutions for security monitoring currently. Splunk offers SIEM, in addition to observability, and therefore can compete with Datadog on this broader playing field. Elastic goes a step further and adds endpoint protection to their security offering, which also includes SIEM. At a recent analyst conference, the Elastic CEO asked the rhetorical question “if we are going to monitor, why not also protect”? While I won’t examine in-depth the merits of adding endpoint protection to observability, I think it underscores the general point about including security monitoring in an observability solution.

While discussing Elastic, it’s worth examining briefly how their observability solutions relate to those offered by Datadog. Elastic is fundamentally a programmable data processing platform, upon which various pre-packaged solutions have been built. These solutions are targeted at the enterprise search, observability and security markets. Due to the open source nature of their code, Elastic customers can extend the platform to address any use case that requires ingesting, analyzing and displaying large data sets. This has resulted in a number of unique customer applications, which go beyond standard monitoring of Internet-delivered applications. Customers are applying this notion of “observability” to different domains, where monitoring a system for expected performance is needed in a generic sense. Examples of this extended observability include:

• Sky for monitoring OTT video delivery
• Volvo for tracking service issues for a fleet of 1M connected vehicles
• John Deere for a service that allows farmers to monitor and optimize the performance of their agriculture operations
• Walmart to monitor gift card usage for fraud
• Cox to track video on demand delivery through their own cable network
• Verizon for wireless network service monitoring and outage tracking

In this regard, comparing Elastic to Datadog solutions for traditional observability use cases (which are largely web site delivery based) is a little like comparing apples to oranges. Datadog observability products are well-suited for customers with traditional Internet-delivered experiences that want a plug-and-play solution that immediately generates value with little configuration. Elastic products, on the other hand, allow for customization to address non-standard system “observability” use cases (like cellular networks, video delivery, IoT device fleets, manufacturing lines, financial transactions). As such, they require development resources and more configuration. This difference is important for investors to consider and puts Elastic in a slightly different context than other observability vendors. I cover Elastic separately and will provide an update on their latest activity soon.

Incident management is a whole new motion for Datadog, which has elements of workflow and collaboration. As mentioned in the product updates section above, I think Datadog is actually better positioned to handle incident response than competitors like PagerDuty and ServiceNow, given their grounding in observability and their purpose-built data processing platform. I think it is much easier to add incident management (and other PagerDuty features) on top of Datadog’s monitoring platform, than for an incident response company to address observability. It will be interesting to see how Datadog expands this offering in the future and if they further encroach on the incident response market.

Finally, Datadog’s recent extension into monitoring application performance in the development pipeline with the Undefined Labs acquisition pushes their reach into pre-production monitoring. This also distinguishes their offering from the other observability vendors and provides another argument for customers to form a “platform” relationship with Datadog. Tying application performance back to their new visibility into actual code changes will deliver a unique perspective going forward that will likely further enhance the intelligence of the Datadog platform.

In order to further examine the platform coverage argument, the matrix below details the various product offerings and the current state by each competitor. Datadog currently has the fewest gaps in coverage and is putting real distance between themselves and the traditional APM vendors of Dynatrace and New Relic.

Over the last quarter, Datadog competitors made a few updates to their product suites, but none matched the pace of Datadog’s addition of whole new product lines. Dynatrace extended their integrations with services from the cloud vendors. They announced support for for all AWS services that publish metrics to Cloudwatch in August and Azure services that feed Azure Monitor in July. They also added support for Kubernetes environments, the ARM Platform and native mobile apps. In February, they improved support for business analytics processing by extending monitoring to business-level KPIs, like revenue trends, customer engagement and churn. Datadog already addresses most of these integrations. Dynatrace differentiates their offerings through an emphasis on AI to provide more insight into issues and on analytics to tie production events back to business metric performance.

Splunk similarly focused on extensions to existing products over the course of the quarter. They made enhancements to their IT Service Intelligence platform. They also released a new version of their Machine Learning Toolkit, which enables users to build and run machine learning models against data processed by Splunk. They updated their Data Stream Processor to add new data sources and enhanced support for machine learning predictive features. Finally, Splunk Mobile can be integrated with more mobile device management (MDM) providers.

Perhaps the most disruptive product announcement in the quarter came from New Relic. On July 31, they announced their “reimagined” New Relic One platform. They simplified their observability platform into three primary components. The Telemetry Data Platform allows users to collect large volumes of operational data in any form, whether metrics, logs, events or traces. This data is then visualized and acted upon in the Full Stack Observability product. This includes monitoring and analysis tools for APM, infrastructure, serverless, digital experience and logs. Finally, machine learning capabilities find insights in data and reduce alert noise through the Applied Intelligence product.

While these solutions have parallels with the offerings of the competitors discussed above and aren’t unique in that regard, New Relic introduced a new pricing model along with it. As part of this, they introduced a perpetual free offering, which has the following generous limits and no time period:

• Data ingestion up to 100GB every month
• Unlimited basic users that can run queries and view dashboards
• One standard user with full access to all features on the platform
• 100M Proactive Detection app transactions and 1,000 Incident Intelligence events every month as part of Applied Intelligence

Beyond the limits on data ingestion for the free offering, each additional 1GB of data costs $0.25. Each additional user is $99/month. New Relic does offer Pro and Enterprise packages for larger customers that have broader requirements for support and enhanced capabilities. What is different about this pricing model, as compared to other competitors, is the per user cost for access to all observability features, versus charging on a per host or product basis (like Datadog, Dynatrace and Splunk).

This new pricing model could result in some commoditization of observability product offerings across the market. It was released in late July, so it may take a few quarters to see if there is impact. While competing on price isn’t always a great strategy, it could win some segments of customer business or cause a similar move from another competitor. On Datadog’s Q2 earnings call, an analyst referenced New Relic’s new pricing model, highlighting the shift to per user and offering a free tier. The Datadog CEO responded by emphasizing the flexibility their pricing model provides and the value realized for the customer by delivering a better product.

So I think we’re very careful about pricing in that the way we do it is, first of all, we encourage having the maximum deployment with our customers. Meaning, that’s why we don’t charge by the user. We want absolutely everybody to use it.

And then we want to make sure that customers have the levers to align what they pay to Datadog with the value they get. And in our case, that means having differentiated pricing for very different parts of our platform because not every single gigabyte of data that’s being sent to us has the same value to the customer and represents the same amount of processing and other things online. So we try to align that. Right now, we’re happy with our pricing.

We believe in our ability to win in the market. And if you win the market and give them more and more value, you have pricing power in the long term. And that’s where we are today, where we want to be in the future. I think when your only tool is to play on pricing, like it’s usually bad news, it means you can’t actually win based on the quality of your products alone.

Datadog CEO, Q2 Earnings Call

I agree that the breadth and quality of the Datadog product will prevent much of the disruption from a less expensive pricing model. New Relic will likely cut into business from smaller customers, given that their perpetual free tier would be beneficial to bootstrapped companies just getting started with building their business. This will be something to monitor going forward for the observability market. Datadog leadership even acknowledged on a subsequent analyst call that they will be watching for signs of impact on their business.

Finally, Sumo Logic has filed for an IPO and should be going public shortly. They focus on log analytics for cloud-based applications to deliver operational, security and business intelligence. Their solutions enable application performance monitoring, SIEM and business metrics tracking. These largely overlap with use cases for Datadog and the other observability competitors discussed above. Sumo Logic’s revenue grew by about 38% in the last 6 months, but with pretty high negative operating and FCF margins. Last year revenue (ending Jan 2020) was $155M total, growing at 50% year/year. So, they are substantially smaller than the other companies examined here. I won’t perform a deep dive analysis at this point, but will include Sumo Logic in future updates after they are public.

With this foundation, let’s compare the financial performance of these companies from the most recent quarterly reports. All relevant metrics are Non-GAAP.

Metric	DDOG	DT	NEWR	SPLK	ESTC
Total Rev	$140M	$155M	$163M	$492M	$129M
Rev Growth	68%	27%	15%	-5%	44%
ARR Growth	NA	37%	14%	50%	NA
Gross Margin	80%	85%	81%	78%	77%
Op Margin	11%	33%	5%	-13%	-3%
R&D	27%	14%	20%	26%	29%
S&M	33%	31%	44%	54%	37%
G&A	9%	11%	13%	11%	14%
FCF Margin	13%	24%	14%	-37%	17%
DBNER	>130%	>120%	100%	NA	>130%
Market Cap	$24B	$11B	$3B	$30B	$9B
P/S Ratio	44.2	18.7	5.3	12.5	17.5

As you can see, Datadog continues to experience the fastest growth. During the most recent quarterly earnings, all companies referenced seeing some impact from the COVID-19 situation and appear to be impacted at roughly equal levels. Datadog’s leadership in growth doesn’t vary from the comparison I produced in my Q1 review of Datadog. However, their lead in revenue growth is shrinking. In Q1, Datadog’s annualized revenue growth of 87% was about 60% higher than the average of Dynatrace and New Relic. This quarter, it is 47% higher.

At the Oppenheimer Technology Conference on August 12, Datadog’s CFO was asked about the impact of competition. Interestingly, this was the first time I have heard Datadog leadership actually acknowledge seeing competitors in new customer deals. While he still maintained that they don’t see that much competition, they do encounter Dynatrace in the “legacy market”. He also acknowledged that they are watching for effects from New Relic’s pricing changes and that it is too early to tell if that will impact Datadog competitively for some customers.

My take is that Datadog will continue to win an oversized share of customers over these competitors, but will not dominate them all. The CFO’s mention of Dynatrace in legacy markets likely refers to their “2,400+ Blue Chip Enterprise Customers“. These represent more traditional businesses in the Fortune 500, like Ford, P&G, Hertz, Comcast, Costco and Allstate. As these companies continue their digital transformations, their software application installations will likely grow and drive incremental revenue for Dynatrace. Similarly, New Relic’s new pricing model and free tier may appeal to SMB’s. Splunk has a broad platform offering covering all of observability, as well as a mature SIEM offering. They already claim 92 of the Fortune 100 as customers.

While Datadog excels with the digital innovators, observability solutions in this category are likely becoming saturated, with fewer greenfield deals in the future. That’s not to say that Datadog’s growth will go to zero. Rather, I think it will drift to levels inline with the actual growth in usage by their customers. This is likely driving Datadog’s expansion into adjacent categories, which will fuel incremental revenue growth outside of their core observability solutions. Given their existing strong market position and continued product expansion, I can foresee Datadog settling into predictable 35-45% annual revenue growth for several years.

Future Opportunities for Datadog

As discussed, Datadog has a history of rapid product development and is expanding into new areas. They seem intent to continue to pursue a disruptive expansion into adjacent categories to continue to fuel their growth. The fact that they just announced several new product offerings at Dash, while competitors made incremental enhancements or tweaked pricing structures is significant.

On the Oppenheimer call, we received some insight into future growth opportunities. Datadog’s primary strategy is to continue to expand their platform of product offerings and then engage in cross-sell activities to encourage customers to keep adding product subscriptions. This tactic has worked well to date, as evidenced by their DBNER over 130%, with 68% of customers using more than one product and 75% of new logos starting with two or more products. While Datadog still initiates most engagements with infrastructure monitoring, that is quickly followed by other observability solutions. The CFO even rank ordered some of the newer offering by revenue contribution on the call – Synthetics, RUM, Network and Security. This order mirrors the launch date for these products.

Newer product announcements, like continuous profiling, compliance monitoring and incident management should follow the same path of gradual adoption by customers, providing a long-tail of sustained growth. The extension into the pre-production environment with the acquisition of Undefined Labs represents a further expansion opportunity. This would apply the same product capabilities in performance, logs, security and synthetics monitoring to the development and test environments. I don’t think this will represent a new, stand-alone product, but rather provides another deployment target that could be monetized in similar ways to production. On the Oppenheimer call, the CFO stated that Datadog doesn’t intend to take over code repository or build functions (like GitHub or GitLab). This makes sense to me as these are dramatically different product motions. However, there is still plenty of functionality to address in the observability and testing of applications before they are promoted to production.

The security opportunity for Datadog could be large. They launched security monitoring earlier this year, which focuses on monitoring for penetration attempts. In August, they added compliance monitoring, which also falls under security. The CFO stated on the analyst call that Datadog does not have or intend to address endpoint or enterprise security (referred by him as “inside the firewall”). When asked about the potential future opportunity for security, the CFO thought that the TAM for Datadog security offerings could be as large as that for observability.

Finally, as discussed above, Datadog is making moves into incident response and management. I am not sure how far this encroachment will go. With the release of their mobile app for DevOps personnel, Datadog did make incident response easier, by tying application performance context to the alert received. Similarly, providing tools to manage an incident and document root cause after resolution are natural extensions to Datadog’s existing observability motion. It will be interesting to see how far Datadog pushes this and encroaches on the domain of PagerDuty. At minimum, I think Datadog’s expansion is a counter to Splunk’s presence in on-call management through its VictorOps capability.

As a caveat, though, Datadog could risk doing too much. Feedback on the Q2 earnings call and Oppenheimer conference was that Datadog’s pricing matrix is getting complex. The Datadog leadership team feels this is still an advantage, in that it provides their customers with transparent flexibility to cater their product consumption to their unique needs. Also, if Datadog supports too many products, it is possible that competitors beat them on depth for point solutions. However, Datadog leadership is thoughtful around how they select new product offerings relative to leverage on their platform. Because of the design of their platform as fundamentally processing and displaying data sets, many of these new product offerings are simply layers on top of core functionality.

Datadog Take-aways

Q2 was a tough quarter for Datadog. On the surface, performance metrics are strong. Revenue growth of 68% year/year is best in class. However, this is down from 87% in Q1. Like many software companies, COVID-19 is having an impact on their business. If we look at how competitors and other software companies in the infrastructure space performed, we see the following changes in revenue growth from Q1 to Q2:

• Dynatrace: Revenue growth decreased from 30% in Q1 to 27% in Q2.
• New Relic: Revenue growth decreased from 21% to 15%.
• Elastic: Revenue growth decreased from 53% to 44%.
• MongoDB: Revenue growth decreased from 46% to 39%.
• PagerDuty: Revenue growth decreased from 33% to 26%.

On a proportional basis, Datadog’s revenue growth decrease was inline with competitors, but on an absolute basis, a 19% sequential decline appears large. As we look at estimates for the rest of the year, we see further deceleration of revenue growth. Q3 revenue is estimated at 50% year/year currently. Assuming a similar level of beat as Q2, they could deliver 56% growth. Looking at implied Q4 growth shows further deceleration. Given the current full year revenue estimate of $566M – $572M, if we take the midpoint of guidance for full year and Q3, it implies Q4 revenue growth of about 36%:

$569M – $144M – $140M – $131M = $154M (35.6% year/year growth)

Granted, I think we can expect a beat and raise for Q3 and Q4 actuals, but even then, we could exit 2020 with a quarterly revenue run rate in high 40% range. This would represent a notable reduction from 84% growth in Q4 2019 and 87% growth in Q1. The Q4 revenue growth rate for this year is obviously challenged by the large outperformance from Q4 2019. This also sets up a difficult comp for Q1 2021.

In the prepared remarks, the CEO did mention that in July they witnessed a “notable improvement in usage growth relative to Q2”. This implies that growth did start to pick back up again in Q3. The CEO further stated that they are being conservative with guidance through the rest of the year, as it is not clear if this uptick will persist. As such, the raise to Q3 revenue guidance was just 4% and full year guidance was raised by 1.5% from analyst estimates. For comparison, in the Q1 earnings report, Datadog raised next quarter’s revenue guidance (Q2 estimate) by 10% and full year guidance by 7%. If this uptick in usage continues through the year, it implies more upside to Q3 and Q4 growth estimates.

On the earnings call, the leadership team provided some explanation for the revenue slow-down. They attributed it to slower growth with existing customers. Many of their larger customers used this situation to optimize usage of Datadog, reducing monitoring in non-critical areas and cleaning up any excess. This does happen occasionally in normal circumstances, but management commented that many customers did this optimization in unison in Q2. Some customers, such as delivery and at-home media companies, did see elevated demand and actually meaningfully grew their Datadog usage. On the other hand, some COVID-impacted customers reduced usage. These customers, such as those in hospitality and travel, contribute less than 10% of ARR, and leadership called them a “mild detractor.”

As investors, I think we need to be prepared for an annualized revenue growth rate for Datadog in 2021 in the 40-50% range. While this is still strong, we have gotten accustomed to Datadog annualized revenue growth in the 60-80% range. High growth in 2020 makes the 2021 comps difficult. Also, the amount of greenfield opportunity in observability and security will fade as most large customers will have an existing commercial solution in place.

Profitability measures are strong with high operating (11%) and FCF (13%) margins. We can expect these to continue to improve gradually. On the earnings call, leadership did point out that gross margins are “running towards the top end of our long-term target”. So, we can’t assume too much upside there. However, R&D and S&M spend in particular are continuing to demonstrate operational leverage.

International represents a growth opportunity at 24% of Q2 revenue. Datadog is investing heavily in building out their go-to-market teams for Asia, EMEA and Latin America. This does represent an area to deliver accelerated growth, as these markets are largely under-penetrated and represent a relatively small percentage of Datadog’s revenue currently.

The other bright spot with Datadog is in continued customer growth. Although slowing from Q1, they still delivered a 71% increase in customers with ARR of $100k or more in Q2. Total customer growth also increased by a reasonable rate. DBNER in excess of 130% for the twelve quarter underscores the ongoing growth that we can expect from existing customers, once they are on the Datadog platform.

Personal Investment Plan

At this point, I feel I have enough visibility into Datadog’s operations to set a 5-year price target, wanting to wait until they had delivered four quarters of public company performance. I think they can deliver $585M of revenue for 2020, representing annualized growth of about 61%. Going into 2021, though, I think revenue growth will decrease, let’s assume to 45%. In 2022, they could hit 40% growth, and finally settle to 35% annually after that. This would mean $2.165B in revenue by end of 2024. A 35% revenue grower with high profitability margins could touch a 30x P/S ratio at the high end on a temporary basis. This assumes gross margins stay near 80% and operating margins reach 20%. FCF generation would be strong at this point. If DDOG reaches these targets, we could see their market cap hitting $65B by end of 2024 (of course, not accounting for share dilution). DDOG’s market cap is currently about $24.3B, which would lead to about a 2.7x increase. At $80 a share now, I think the stock could reach $215 within 5 years. Investors can repeat the calculations above using their own valuation multiple assumptions.

For my personal portfolio, I plan to open a position in DDOG at some point. I think Datadog represents a prudent long-term bet on the observability market, which will benefit from the secular tailwinds of digital transformation efforts for many years. Datadog should maintain their leading position in the market and is generating new revenue streams with product expansions. Currently, I don’t think DDOG’s valuation takes into account the anticipated revenue growth slowdown. If they outperform and deliver $600M in revenue for 2020, that yields a forward P/S ratio of 40 at today’s market cap of $24B. I will likely wait until later in the year and build a position when the P/S ratio drops into the 35-40 range (from 44 now).

NOTE: This article does not represent investment advice and is solely the author’s opinion for managing his own investment portfolio. Readers are expected to perform their own due diligence before making investment decisions. Please see the Disclaimer for more detail.