The AI Era Has a Complexity Problem. Datadog Is the Answer.
As Agentic AI adoption creates huge demand for IT Observability, Datadog is a top AI pick for the next decade or two, but it comes with a price tag
Datadog is one of the few SaaS stocks – if not the only large-cap – that has outperformed the broader market YTD (+63%) and is near all-time highs. And, honestly, for good reason, given this is one of the few SaaS companies where there is no doubt that AI adoption is a huge long-term tailwind, directly making it increasingly one of the most essential platforms in the IT stack.
Additionally, the company continues to deliver sensational numbers and accelerating growth, which can be immediately attributed to AI adoption and increasingly complex IT systems, an evolution still in its very early stages.
In other words, current enthusiasm toward the company – in sharp contrast with the rest of the sector – is fully justified. So, for those unfamiliar, what is Datadog?
Datadog acts as the observability and security backbone for over 33,000 companies, offering deep, real-time visibility across infrastructure, applications, logs, user experience, CI/CD pipelines, and security posture, all from a single, integrated console.
Put simply, Datadog helps companies keep their entire digital operations running smoothly — from websites and apps to AI agents — by providing a single, unified view of what’s happening inside all systems.
The Datadog platform is built for today’s modern, complex, cloud-native environments. Its strength lies in its ability to offer deep visibility across the entire technology stack, from backend systems and cloud services to user-facing applications, AI agents, and new coding, all in a single unified platform, offering a wide array of use cases, including infrastructure monitoring, application performance monitoring (APM), log management, incident management, and cloud security.
Datadog supports customers from the moment the first bit of code is written to the first GPUs deploying it and the customers accessing it – the entire IT lifetime runs through Datadog, ensuring as few issues as possible that can result in downtime, security breaches, etc.
You see, when digital systems break or slow down, such as a shopping website crashing, a mobile app freezing, or a bug in a newly published bit of coding, it can be challenging to identify the source of the problem, especially if different teams are each examining only one aspect of the puzzle. Digital complexity is at an all-time high, and it’s only getting worse.
Datadog solves this by pulling data from all these systems, including things like performance metrics, error logs, and user activity, and displays it on a single, easy-to-understand dashboard. It helps developers and IT teams quickly identify issues, understand their causes, and resolve them before they impact customers.
According to recent data, Datadog implementation can reduce software production incidents by 10x, downtime by 3.5x, and impacted customer transactions by 20x.
So, ultimately, the value proposition is straightforward: as companies migrate to the cloud and adopt AI, making their systems more distributed, fragmented, and complex, the cost of not knowing what’s going on becomes enormous. Downtime leads to lost revenue. Performance issues hurt user engagement. Security blind spots pose significant financial and reputational risks (think CrowdStrike). Datadog helps organizations stay ahead of it all, before problems impact customers or the bottom line.
And the AI era is making this value proposition not just more relevant, it’s making it existential, as the proliferation of AI agents is introducing an entirely new layer of operational complexity. Unlike traditional software, which follows predictable logic, agentic systems are inherently non-deterministic; they make decisions, invoke tools, call other agents, and can spiral into unexpected behaviors with no human in the loop to catch it.
As companies race to integrate AI into their products and workflows, most lack visibility into how their AI systems behave, what their agents are doing, and whether they deliver real business value.
Datadog is squarely positioned to solve this. Its AI Agent Monitoring capability maps each agent’s full decision path — inputs, tool invocations, calls to other agents, and outputs — in an interactive graph, allowing engineers to drill down into latency spikes, incorrect tool calls, and unexpected behaviors, and correlate them with quality, security, and cost metrics. This is observability applied to AI, and it’s a natural extension of everything Datadog has built.
Overall, as cloud adoption, software complexity, and AI workloads surge, the need for real-time observability is becoming existential. Datadog is not only riding this wave, but it’s the best-in-class option.
You simply can’t be better positioned for current developments in IT and computing than Datadog, whose TAM is expanding exponentially.
In early May, Datadog released its latest financial results – Q1 2026 – and the numbers were straight-up sensational, sending shares soaring over 30% in the following trading session to a new all-time high, driven by impressive business momentum and a strong beat. On top of that, Datadog held its 2026 investor day back in February, which provided loads of additional insight.
Taken together, I think it’s due time for an update on this amazing business. In today’s analysis, I will go over the Q1 results and investor day insights in detail to update my Datadog thesis and financial framework.
Without further ado, let’s delve in and see how the business is doing and at what price it’s worth picking up some shares!
This is a paid-exclusive Deep Dive. Want full access to this 7,000-word analysis? Consider upgrading to the paid tier for $9.50 a month or $95 a year! This gets you:
6–8 Research Reports per month, 2x more than free subs do, which includes 3-4 Deep Dives like this one! (that comes down to just $1 per analysis)
Full visibility into my personal portfolio, including allocations, transactions, and conviction levels
An overview of target prices and ratings, updated weekly
InvestInsights Premium is designed to help you find best-in-class compounders, identify deep-value opportunities, avoid costly mistakes, and stay focused on fundamentals when markets get noisy, with full transparency into how I make decisions over time.
Financial & Performance Review
Jumping straight into the numbers and financial details, Datadog reported total Q1 revenue of $1.01 billion, marking the first quarter over $1 billion and bringing ARR to over $4 billion. The print beat consensus by a sizeable $50 million and came in above the high end of guidance. Not a bad start.
But the headline number isn’t even the most impressive part. Revenue grew 32% year-over-year, accelerating from 29% last quarter and 25% in the year-ago quarter — the fastest growth rate in three years, achieved on a revenue base that has more than doubled since then. Let that sink in: Datadog is seeing such strong, accelerating demand that it can deliver its fastest growth rate in three years on 2x the revenue base.
Sequential momentum tells the same story. Revenue rose 6% quarter-over-quarter, the strongest sequential growth since 2022. The $53 million QoQ add was the highest ever for a first quarter, and the company posted an all-time record for sequential ARR growth, accelerating each month throughout the quarter, so demand and momentum only seem to be getting stronger.
This exceptional business momentum is driven by record sequential growth in existing customer usage, a strong influx of new customers, and a broad-based acceleration across both AI and non-AI customers.
Datadog is simply seeing record demand, and that trend continues to strengthen month after month – you can’t ask for more proof that AI adoption and the subsequent compounding of IT complexity are a huge tailwind for Datadog, proving much stronger than anticipated.
Impressive is an understatement.
Breaking down the growth drivers, customer additions were remarkably strong in Q1. Datadog ended the quarter with 33,200 customers, up 9% YoY or nearly 3,000 YoY additions.
However, real growth doesn’t come from the number of additions but from the size of the contracts these new customers commit to from the get-go. You see, Datadog reported a new logo, annualized bookings at a new all-time high by a significant margin, more than doubling YoY. This is driven by new customers coming in with more modules, driving new logo average land size, or simply the average new customer contract size, at a record high, more than doubling YoY.
That is huge, significantly accelerating growth. Besides, it means customers are more convinced of the necessity, importance, and ROI of the Datadog platform and are more comfortable committing to larger, longer initial contracts. Really, that is a very bullish signal, especially as this momentum accelerates and the IT complexity revolution driving it is in its early stages.
Meanwhile, in terms of customer additions, Datadog has a huge growth runway. Fortune 500 penetration has been rising sharply from 30% in 2021 to 50% by the end of 2025, but that still leaves plenty of room to grow, especially given that Datadog currently captures just 7% of its total customer TAM. And with observability becoming practically a necessity amid AI adoption, specifically Agentic AI, and Datadog being the best platform to deliver the best ROI, I am quite bullish on the medium-term prospects. I think it can maintain this high single-digit customer growth rate, if not accelerate it into the double digits, in the coming years, considering the rapidly growing demand.
And as each of those customers comes in with significantly larger initial contracts, that can drive exceptional growth.
But that isn’t even Datadog’s primary growth driver.
As highlighted below, in 2025, only 25% of added ARR came from new customers, with nearly 75% coming from its existing customer base through cross-selling new products (adoption of new modules) and growth in usage of existing products.
In other words, Datadog derives the majority of its growth from its existing customer base. For reference, revenue per customer has grown at a 17% CAGR since 2020, driving incremental growth within the existing base. This is the power of Datadog’s land-and-expand strategy, aided by impressive product innovation, with Datadog going from 9 modules in 2020 to 26 today.
By now, the Datadog platform covers the entire IT stack, from GPU monitoring and data observability to real-user monitoring, software delivery, and security. And newer modules are seeing rapid adoption, as customers are happy to shift a larger piece of their IT stack into Datadog. For example
Log management revenue is up 7x from 2020, nearing $1 billion in ARR.
Cloud SIEM has grown 18x since 2020.
Datadog security grows strongly, now covering 8,500 customers and 25% of the Fortune 500.
Multi-product adoption is strong and growing rapidly. Today, 84% of customers use 2 or more products, up from 72% in 2020. 56% use 4+ products, up strongly from just 22% in 2020 and up 6 percentage points from 2024. 35% use 6+ products, up from 28% one year ago, and 20% use 8+ products, up from 13% one year ago and 1% in 2021.
This adoption of more modules, which is rising sharply as Datadog’s platform delivers significant ROI for customers, is driving rapid growth in average customer value. Below is a customer example of exactly this dynamic.
Highlighting this dynamic further, Datadog customers spending more than $100k annually have grown at a 28% CAGR, and customers spending $1 million at a 43% CAGR between 2020 and 2025, reaching 603.
In Q1, Datadog reported 4,550 customers with ARR over $100k, up a huge 21% YoY, indicating strong acceleration from recent quarters and the fastest growth in nearly three years. This clearly indicates that Datadog’s most valuable customers are getting more valuable, faster. The bigger the customer, the deeper the platform embeds itself, and the more spend naturally follows. It’s the land-and-expand flywheel operating exactly as designed, and right now it’s spinning faster than it has in years.
In other words, Datadog is accelerating its value per customer, despite a high base, as usage growth and module adoption are stronger than ever amid current technology shifts.
Really, all these operational and demand metrics are pointing in the right direction and strengthening well ahead of expectations – that is a really bullish signal once more.
Unsurprisingly, a key driver of the acceleration we are currently seeing is AI adoption and the complexity it entails. As already noted, AI only increases the need for Datadog’s solutions.
Datadog splits AI into two buckets: AI for Datadog and Datadog for AI.
AI for Datadog includes products and capabilities that make the Datadog platform better and more useful – functionalities within the platform that leverage AI.
For example, Datadog launched its MCP server for general availability, giving developers access to live production data to debug their applications directly in their AI coding agent or IDE. In plain terms, when a developer is writing or fixing code using an AI coding assistant — like Claude Code, Cursor, or GitHub Copilot — they normally have to switch between their coding environment and Datadog’s dashboard to understand what’s actually going wrong in production. Datadog’s MCP server eliminates that context switch. It’s a subtle but strategically important move, as it embeds Datadog deeper into the daily workflow of every developer using an AI coding agent, which is rapidly becoming most of them.
Additionally, Datadog launched Bits AI Security Agent a while back, which, to quote management, “autonomously triages Datadog Cloud SIEM signals, conducts in-depth investigations of potential threats, and delivers actionable recommendations.”
Bits AI is Datadog’s own AI assistant built directly into the platform. Instead of a human engineer having to stare at dashboards, piece together logs, and manually figure out what’s wrong when something breaks, Bits AI does that investigative work automatically. You can see it as Datadog’s Agentic solution, further reducing the need for engineers to patch up issues.
According to management, Bits AI can reduce investigations that could take hours to as little as 30 seconds, and do so fully automatically. Pretty awesome.
Adoption of these AI-for-Datadog solutions drives incremental revenue.
Then there is Datadog for AI, which includes Datadog capabilities that deliver end-to-end observability and security across the AI stack. So, these aren’t platform features but systems designed to manage a company’s AI activity. This includes GPU monitoring to understand GPU fleet utilization, workload efficiency, thermal and power behavior, and interconnect performance to drive GPU ROI and reliability.
A good way of framing Datadog’s position in relation to my ServiceNow post from last week is this: Where ServiceNow is positioning itself as the air traffic controller, deciding which planes fly, on what routes, under what rules, Datadog is the radar system. It tells you where every plane actually is, whether the engines are healthy, and flags when something’s about to go wrong.
Adoption here is especially strong as companies start to integrate more and more AI, especially as this turns Agentic, growing the need for visibility and reliability.
Currently, Datadog already counts 6,500 customers using AI integrations in one way or another, which is about 20% of the customer base, up from 9% in early 2025 and 18% in early 2026, so growing rapidly, and this is accelerating. Bits AI investigations have doubled between December and March; the number of spans sent to the LLM observability product nearly tripled sequentially; the number of Datadog MCP server tool calls quadrupled sequentially; and the number of Bits Assistant messages increased by a factor of 12 in that period.
So, fairly sensational growth. And the long-term potential is huge.
Ultimately, every organization deploying agents in production needs to monitor them. Datadog, with its unified platform and decade of telemetry expertise, is the natural place to do it. Monitoring non-deterministic, distributed agentic systems at the telemetry level, including tracing every tool call, every latency spike, every model decision, is engineering-heavy work that Datadog has spent a decade building the infrastructure to do, making it a natural choice, especially when companies use multiple different AI providers, making Datadog by far the best neutral tool.
This is exactly why Datadog has already seen a significant influx of AI native customers, which are driving a large portion of current growth. For reference, AI-native customers are companies whose core business is built around AI. Think OpenAI, Anthropic, Mistral (a confirmed customer), Perplexity, and the wave of startups building AI-powered products. They’re not traditional enterprises adopting AI on the side; AI is their product.
For Datadog, they matter for two reasons. First, they are extraordinarily infrastructure-heavy. Training models, running inference at scale, and managing GPU clusters generate massive volumes of telemetry data, which translates directly into Datadog usage and, therefore, revenue under its consumption-based model. A single large AI-native customer can spend at a scale that would take a traditional enterprise years to reach. Second, they grow fast. As an AI-native startup scales its user base and inference volume, its Datadog bill scales automatically with it, without any additional sales effort.
As of 2025, AI-native customers account for 11% of total revenue, up from 5% in 2024, and are growing rapidly in the triple digits, significantly outpacing the rest of the business. 22 of these customers already spend more than $1 million annually, and 5 of these more than $10 million.
This customer base is extremely promising, and Datadog is seeing a good inflow of new customers in recent quarters, while usage has exploded. This will remain a key growth driver in the coming years.
However, it’s not just AI-native customers that are driving growth – Datadog is not entirely reliant on this cohort – as non-AI customer revenue growth also continues to accelerate, at mid-twenties in Q1, up from 23% in Q4 and 19% one year ago. That is a very impressive acceleration, driven by strong cloud migration momentum and early-stage AI adoption, which is driving greater module adoption and accelerating usage growth.
Meanwhile, company-wide retention is strong. Gross retention in Q1 was in the mid-to-high teens, likely around 97%, which is where it has consistently been. Worth pointing out is that Datadog has historically had a large SMB customer base, which generally sees lower retention simply because SMBs fail at a much higher rate than enterprises. This has a dilutive impact on Datadog’s overall renewal rate, which makes the 97% retention the more impressive and, above all, stable. For reference, Datadog’s enterprise renewal rate is above 98%.
This perfectly reflects the mission-critical nature of the Datadog platform.
Moreover, Datadog’s net retention rate in Q1 was in the low-120s, up from about 120% in Q4, and hugely impressive. In practical terms, this means that even if Datadog added zero new customers in the past year, its existing customer base alone would have grown revenue by more than 20%, simply through expanded usage, deeper module adoption, and larger contracts. It is the clearest measure of how deeply the platform embeds itself over time, and the fact that it is reaccelerating proves just how strong Datadog’s current business momentum is amid current technology shifts.
Finally, billings in Q1 were $1.03 billion, up 37% YoY, and RPO hit $3.48 billion, up 51% YoY. Notably, RPO duration increased YoY, with more customers opting for multi-year contracts, further confirming that Datadog has nothing to fear from AI and that its platform is becoming even more critical as a result.
Honestly, this was a sublime report.
Let’s then move to the P&L.
