Datadog

Datadog priced at $27, told investors its growth rate would slow — and then compounded into a nearly $80 billion S&P 500 component up almost 800% from the IPO.

Datadog came public on September 19, 2019 as one of the cleanest high-growth SaaS stories of that cohort: ~$198 million of 2018 revenue growing ~80% in the first half of 2019, mid-70s gross margins, a tiny operating loss, and only ~$141 million of preferred raised before the offering. The S-1 was unusually blunt — it named AWS, GCP and Azure as competitors while renting their infrastructure, disclosed a 2016 breach in which customer credentials were exfiltrated, and told readers flatly not to extrapolate the growth rate. Shares popped 33.9% on debut to close at $36.15 and never revisited the offer price; the all-time low of $31.57 came less than a month later on October 13, 2019.

What followed was the bull case running further than the bull case. Revenue roughly doubled again in 2019 and 2020, crossed $1 billion in 2021 and reached about $2.7 billion in 2024, with the $100k+ ARR customer count climbing from ~590 at filing to several thousand and the $1M+ cohort from 42 into the hundreds. Gross margin expanded rather than eroded, into the ~80% range, contradicting both the cost-of-revenue red flag and the fear that hyperscaler marketplace distribution would compress economics — marketplaces became a major channel instead. The company turned GAAP-profitable in 2023, and in July 2025 was added to the S&P 500, dissolving the S-1's warning that dual-class stock made the shares index-ineligible. Both co-founders remained in place throughout.

The risks that did bite were the ones management flagged most honestly. Growth decelerated exactly as promised — from ~83% in 2019 to roughly 27% in 2023 — as customers optimized cloud spend in 2022-23, and dollar-based net retention slid from above 130% toward the mid-110s, dragging the stock from a late-2021 high near $200 down to the low $60s before it recovered. A multi-region outage in March 2023 knocked the monitoring vendor itself offline for roughly a day and produced around $5 million in service credits, a direct hit on the 'service-level commitments' risk factor, though immaterial financially. A newer concentration risk not contemplated in the S-1 emerged in 2024-25 as AI-native customers, led by a single very large one, grew into a low-double-digit share of ARR and injected fresh volatility into guidance.

As of August 28, 2026 the stock trades at $236.98, up 777.7% from the $27 IPO price and just off a July 2026 peak of $267.97. Meanwhile the competitors the S-1 listed by name were consolidated away: New Relic was taken private in 2023 and Splunk was acquired by Cisco in 2024.

Datadog is a fast-growing SaaS monitoring and observability platform that unifies infrastructure monitoring, application performance monitoring and log management for cloud and hybrid IT environments, sold bottom-up to developers via a free tier and self-service installation and expanded upward into enterprises. Revenue nearly doubled to $198.1 million in 2018 and grew about 80% year-over-year in the first half of 2019, with gross margins in the mid-70s and operating losses kept small relative to peers, and the company is raising IPO capital to fund product development, sales expansion, international growth and possible acquisitions under a dual-class structure controlled by insiders.

The bull case is that monitoring is being rewritten by the shift to dynamic cloud infrastructure, and Datadog's single platform — spanning infrastructure, APM and logs with more than 350 out-of-the-box integrations — replaces a stack of legacy point tools. A frictionless free tier and self-service install seeds adoption inside engineering teams, then a field sales force converts and expands those accounts: as of June 30, 2019 roughly 590 customers already paid $100,000+ in ARR and 42 paid $1 million or more. If dollar-based net retention stays high and international markets (24% of ARR at the end of 2018) scale off a newly built presence in Dublin, Paris, London, Singapore, Tokyo, Seoul and Sydney, revenue compounds on top of ~75% gross margins with an expense base that has already flirted with breakeven — the company reported net income of $0.5 million in the first half of 2018 — implying a credible path to durable profitability at scale.

With hindsight: Every pillar of the bull narrative held: land-and-expand from a free tier worked, the enterprise cohort scaled from ~590 $100k+ customers into the thousands, international expanded off the seven-city base, and the ~75% gross margin structure improved to ~80% while the company crossed into sustained GAAP profitability by 2023 — far beyond the 'credible path' the S-1 described. Revenue grew from $198 million in 2018 to roughly $2.7 billion in 2024, and the stock is up 777.7% from the IPO price with S&P 500 inclusion in 2025.

• Revenue nearly doubled from $100.8 million in 2017 to $198.1 million in 2018, and grew from $85.4 million to $153.3 million in the first half of 2019.

  “Our revenue was $100.8 million, $198.1 million, $85.4 million and $153.3 million for the years ended December 31, 2017 and 2018 and the six months ended June 30, 2018 and 2019, respectively.” source ↗

• The company has built a meaningful enterprise cohort, with ~590 customers at $100,000+ ARR and 42 at $1 million+ ARR as of June 30, 2019.

  “As of June 30, 2019, we had approximately 590 customers with an ARR of $100,000 or more, and 42 customers with an ARR of $1.0 million or more.” source ↗

• More than 350 out-of-the-box integrations support a self-service deployment model, which management frames as core to the platform's adoption.

  “We currently offer more than 350 out-of-the-box integrations to assist customers in deploying Datadog, and we need to continuously modify and enhance our products to adapt to changes and innovation in existing and new technologies to maintain and grow our integrations.” source ↗

• International already contributes 24% of ARR with sales presence established across seven cities outside North America, framing further geographic expansion as the growth lever.

  “Customers outside North America generated 24% of ARR as of December 31, 2018. Beyond North America, we now have sales presence internationally, including in Dublin, Paris, London, Singapore, Tokyo, Seoul and Sydney.” source ↗

• The platform is positioned as a unified replacement for multiple traditional monitoring product categories rather than a single-category point tool.

  “Our unified platform combines functionality from numerous traditional product categories, and hence we compete in each of these categories with home-grown and open-source technologies, as well as a number of different vendors.” source ↗

• Growth strategy depends on a free trial/free tier motion converting bottom-up usage into paid subscriptions.

  “To encourage awareness, usage, familiarity and adoption of our platform and products, we offer free trials and a free tier of our platform. These strategies may not be successful in leading customers to purchase our products.” source ↗

1. Competes directly with the cloud providers it runs on

   didn't happen

   competition · structural

   Datadog names AWS, Google Cloud and Microsoft Azure native monitoring as competitors, while simultaneously depending on those same providers to host substantially all of its infrastructure and increasingly to distribute its product.

   “With respect to cloud monitoring, we compete with native solutions from cloud providers such as Amazon.com, Inc. (Amazon Web Services, or AWS), Alphabet Inc. (Google Cloud Platform, or GCP) and Microsoft Corporation (Microsoft Azure).” source ↗

   What happened: Native tools like CloudWatch and Azure Monitor did not displace Datadog; the hyperscalers became distribution partners, with cloud marketplaces growing into a material bookings channel while revenue compounded from $198M (2018) to roughly $2.7B (2024).

2. Broad multi-front competition against much larger vendors

   didn't happen

   competition · structural

   Because the platform spans several product categories, it competes at once with IBM, Microsoft, Micro Focus, BMC, CA, Cisco, New Relic, Dynatrace, Splunk and Elastic, plus open-source and home-grown tools, and it faces active price competition.

   “In addition to product and technology competition, we face pricing competition. Some of our competitors offer their solutions at a lower price, which has resulted in, and may continue to result in, pricing pressures.” source ↗

   What happened: Rather than losing to larger vendors, Datadog outgrew the field — New Relic was taken private in 2023 and Splunk was acquired by Cisco in 2024, while Datadog's gross margin expanded from the mid-70s toward ~80%, the opposite of sustained price compression.

3. Business model hinges on net expansion and renewals it cannot predict

   partly came true

   growth · structural

   Subscriptions are primarily monthly or annual with no renewal obligation, and usage-based expansion drives revenue; management states it cannot accurately forecast dollar-based net retention, which could decline for many reasons including customer consolidation.

   “We cannot accurately predict our dollar-based net retention rate given the diversity of our customer base, in terms of size, industry and geography.” source ↗

   What happened: Dollar-based net retention ran above 130% through 2021, then fell to roughly the mid-110s during the 2022-23 cloud cost-optimization cycle, slowing growth and cutting the stock roughly 70% from its late-2021 high before it recovered.

4. History of losses and rising cost base

   didn't happen

   profitability · serious

   Losses in every period since inception, an accumulated deficit of $120.3 million, and explicit plans to keep spending heavily on R&D, sales and international expansion mean profitability is not assured.

   “We generated net (loss) income of $(2.6) million, $(10.8) million, $0.5 million and $(13.4) million for the years ended December 31, 2017 and 2018 and the six months ended June 30, 2018 and 2019, respectively. As of June 30, 2019, we had an accumulated deficit of $120.3 million.” source ↗

   What happened: Datadog reached sustained GAAP profitability in 2023 and grew net income substantially in 2024, retiring the accumulated-deficit concern flagged at IPO.

5. Management concedes the growth rate will decline

   came true

   growth · serious

   The filing states outright that revenue growth is expected to decelerate as the business matures, and lists a long chain of conditions that must be satisfied to sustain growth at all.

   “You should not rely on the revenue growth of any prior quarterly or annual period as an indication of our future performance. Even if our revenue continues to increase, we expect that our revenue growth rate will decline in the future as a result of a variety of factors, including the maturation of our business.” source ↗

   What happened: Growth decelerated exactly as disclosed — from ~83% in 2019 and 66% in 2021 to about 27% in 2023 and the mid-20s in 2024 — as customers optimized cloud spending.

6. Total dependence on outsourced third-party cloud hosting

   partly came true

   platform dependence · structural

   Substantially all infrastructure is outsourced to third-party hosting services while Datadog gives customers uptime commitments; capacity limits, termination or prolonged outages would directly impair onboarding, expansion and reputation.

   “We outsource substantially all of the infrastructure relating to our cloud solution to third-party hosting services. Customers of our cloud-based products need to be able to access our platform at any time, without interruption or degradation of performance, and we provide them with service-level commitments with respect to uptime.” source ↗

   What happened: A March 2023 multi-region incident took Datadog's own platform down for roughly a day, but no hosting-capacity, termination or pricing event of the kind warned about impaired the business, and gross margins improved despite growing infrastructure spend.

7. Disclosed 2016 breach of customer credentials

   didn't happen

   tech & security · serious

   An unidentified third party exfiltrated data from Datadog infrastructure including a database of customer platform and integration credentials, forcing a password reset; the company handles sensitive customer telemetry and admits remediation may not be successful.

   “For example, in July 2016 an unidentified third party gained unauthorized access to, and exfiltrated data from, certain of our infrastructure resources, including a database that stored our customers’ credentials for our platform and for third-party integrations.” source ↗

   What happened: No comparable publicly documented breach of Datadog customer credentials has occurred since the IPO, and no securities-relevant enforcement or litigation tied to the 2016 incident became material.

8. Cloud marketplace distribution could erode margins and customer relationships

   didn't happen

   platform dependence · serious

   Selling through hyperscaler marketplaces may grow, but the company warns it reduces direct commercial relationships with customers and compresses profit margins on those sales.

   “An increasing number of sales through cloud provider marketplaces could reduce both the number of customers with whom we have direct commercial relationships as well as our profit margins on sales made through such marketplaces.” source ↗

   What happened: Marketplace sales grew into a significant channel, yet gross margin rose from the mid-70s at IPO to roughly 80%, and the enterprise customer base expanded rather than becoming intermediated away.

9. Service-level credits and outages

   came true

   operations · serious

   Subscription agreements contain uptime and response-time commitments; failures can trigger service credits, terminations with refunds of prepaid amounts, and reputational damage for a monitoring vendor whose value proposition is reliability.

   “If we are unable to meet the stated service-level commitments, including failure to meet the uptime and response time requirements under our customer subscription agreements, we may be contractually obligated to provide these customers with service credits which could significantly affect our revenue” source ↗

   What happened: The March 8, 2023 global outage disrupted all Datadog regions for roughly a day and management quantified about $5 million of associated revenue impact from service credits — real, disclosed, but immaterial against a multi-billion-dollar revenue base.

10. Concentrated insider voting control via dual-class stock

    partly came true

    governance · serious

    Class B shares carry super-voting rights; the two co-founders alone held roughly 23% of voting power pre-IPO and existing holders will control the vote, and the structure makes the stock ineligible for major indices.

    “For example, as of June 30, 2019, Olivier Pomel and Alexis Lê-Quôc represented approximately 23% of the voting power of our outstanding capital stock, and if they retain a significant portion of their holdings of our Class B common stock for an extended period of time, they could control a significant portion of the voting power” source ↗

    What happened: The dual-class structure persisted with founders retaining outsized voting power, but the specific harm flagged — index ineligibility — reversed when Datadog was added to the S&P 500 in July 2025, and no governance dispute materially damaged shareholders.

11. Privacy regulation exposure (GDPR, ePrivacy, CCPA)

    didn't happen

    regulation · serious

    The platform stores and transmits personally identifiable information; GDPR fines can reach the greater of €20 million or 4% of global revenue, transfer mechanisms are of uncertain durability, and CCPA takes effect January 1, 2020.

    “In particular, under the GDPR, fines of up to 20 million euros or up to 4% of the annual global revenue of the noncompliant company, whichever is greater, could be imposed for violations of certain of the GDPR’s requirements.” source ↗

    What happened: No material GDPR, CCPA or ePrivacy penalty against Datadog has been publicly documented in the years since the IPO.

12. Dependence on co-founders and scarce engineering talent

    didn't happen

    key person · serious

    Growth depends on co-founders Pomel and Lê-Quôc plus the CFO and existing software engineers, all at-will, in a market where competition for SaaS engineers and sales talent is intense.

    “Our success and future growth depend largely upon the continued services of our executive officers, particularly Olivier Pomel, our co-founder and Chief Executive Officer, Alexis Lê-Quôc, our co-founder, President and Chief Technology Officer, and David Obstler, our Chief Financial Officer” source ↗

    What happened: Olivier Pomel remained CEO and Alexis Lê-Quôc remained CTO throughout the post-IPO period, and headcount scaled from roughly 1,000 at filing to well over 5,000 without a disclosed talent-driven disruption.

13. Thin patent portfolio and unavailable trademark in the EU

    unclear

    legal · serious

    The company acknowledges it lacks a large patent portfolio to deter infringement claims, and that it has been unable to register the Datadog trademark in certain jurisdictions including the EU as it expands internationally.

    “For example, as we have expanded internationally, we have been unable to register and obtain the right to use the Datadog trademark in certain jurisdictions, including in the EU, and as we continue to expand, we may face similar issues in other jurisdictions.” source ↗

    What happened: No major IP litigation or trademark dispute involving Datadog has been publicly reported as material, and the company does not disclose outcomes on the EU trademark issue in a way visible from market data.

14. Enterprise upmarket push lengthens cycles and raises upfront cost

    didn't happen

    growth · serious

    Targeting large enterprises brings longer, less predictable sales cycles, custom configuration and pricing negotiations, and substantial upfront investment with no guarantee of broad deployment.

    “large enterprise customers often begin to deploy our products on a limited basis, but nevertheless demand configuration, integration services and pricing negotiations, which increase our upfront investment in the sales effort with no guarantee that these customers will deploy our products widely enough across their organization to justify our substantial upfront investment.” source ↗

    What happened: The upmarket motion succeeded: customers paying $100,000+ in ARR grew from ~590 at filing to several thousand, and $1M+ ARR customers from 42 to the hundreds, while operating leverage improved enough to reach GAAP profitability.

15. International operations concentrated in France and Europe

    didn't happen

    operations · boilerplate

    About 31% of full-time employees sit outside the U.S., half of those in France, exposing the company to more onerous European labor law, Brexit uncertainty and FX-driven expense volatility while revenue is billed in dollars.

    “As of June 30, 2019, approximately 31% of our full-time employees were located outside of the United States, 50% of whom were located in France.” source ↗

    What happened: European operations expanded without a disclosed labor-law, Brexit or FX event that materially affected results; international revenue grew as a share of the business roughly in line with the plan described at IPO.

How the S-1 reads

The document is unusually blunt for a high-growth SaaS IPO: it names its competitors individually (including the three hyperscalers whose infrastructure it rents), states outright that its growth rate will decline, and discloses a specific 2016 breach in which customer credentials were exfiltrated rather than hiding behind generic cyber language. At the same time the bull-case metrics are thinly sprinkled through the risk factors — customer counts at $100k and $1M ARR, 350+ integrations, 24% international ARR — while the dollar-based net retention figure that underpins the whole model is discussed only as an unpredictable variable, and all dual-class voting percentages are left blank at this stage. Governance choices lean firmly toward insiders: super-voting Class B, staggered board, 66 2/3% supermajorities, no written-consent actions, and both Delaware and federal exclusive-forum provisions (with an unusual admission that the Chancery Court has already found the federal Securities Act forum clause unenforceable). Two details reward close reading: a ~$5 million non-recurring tax-liability release that reduced H1 2019 operating expenses, and a lock-up that can free 20% of insider shares roughly 90 days after pricing if the stock is up 33%.

• “You should not rely on our revenue for any prior quarterly or annual periods as any indication of our future revenue or revenue growth.” source ↗

  The company tells investors flatly not to extrapolate its growth rate.

• “In addition, we may increasingly choose to allow these third-party hosting providers to offer our solutions directly through their customer marketplaces.” source ↗

  Datadog acknowledges the awkward economics of selling through the cloud providers' own marketplaces.

• “Historically, we have experienced seasonality in new customer bookings, as we typically we enter into a higher percentage of subscription agreements with new customers and renewals with existing customers in the fourth quarter of the year.” source ↗

  Fourth-quarter concentration in bookings is expected to intensify as the company chases enterprises.

• “Market opportunity estimates and growth forecasts included in this prospectus, including those we have generated ourselves, are subject to significant uncertainty and are based on assumptions and estimates that may not prove to be accurate, including the risks described herein.” source ↗

  Management discloses that its own market-size estimates may be wrong.

• “Further, our anticipated proceeds from this offering increase the likelihood that we will devote resources to exploring larger and more complex acquisitions and investments than we have previously attempted.” source ↗

  IPO proceeds are expected to push the company toward larger, riskier M&A than it has attempted before.

• “We believe our culture has been a key contributor to our success to date and that the critical nature of the platform that we provide promotes a sense of greater purpose and fulfillment in our employees.” source ↗

  The company frames culture retention as a competitive asset at risk from going public.

What this one teaches

• A management team that explicitly tells you growth will decelerate is often the most credible kind: Datadog's growth did fall from ~83% to ~27%, exactly as the risk factor said, and the stock still returned nearly 800% because the deceleration happened off a base that grew thirteenfold.
• 'We compete with our own suppliers' reads as a structural risk but can be a distribution opportunity — hyperscaler marketplaces became a growth channel for Datadog while gross margins expanded, the reverse of what the risk factor and the margin red flag implied.
• Watch which disclosed variable actually drives the stock. Dollar-based net retention was left out of the headline metrics as 'unpredictable,' and its slide from 130%+ to the mid-110s in 2022-23 was precisely what caused the deepest drawdown.
• Governance risks flagged as permanent can expire with the rules: the S-1 warned dual-class stock made Datadog index-ineligible, and index-provider policy changes plus S&P 500 inclusion in 2025 quietly erased that concern.

The paper trail

1. 2019-08-23 S-1 filing index ↗ document ↗
2. 2019-09-09 S-1/A filing index ↗ document ↗
3. 2019-09-17 S-1/A filing index ↗ document ↗
4. 2019-09-19 424B4 filing index ↗ document ↗

Filed as Datadog, Inc.. All documents are public domain, served by SEC EDGAR.