Author: Rita
Datadog delivered a stellar performance for the second quarter of 2026: revenue grew 35.6% year-over-year, accelerating by 340 basis points sequentially, marking the strongest growth rate since 2022. The company's quarterly revenue exceeded guidance by $46 million, and it raised its full-year guidance midpoint by more than triple.
Behind these "above-expectation" numbers, Bernstein sees a growth peak.
The report's title directly presents the judgment: Q2 Growth Peaks, H2 Mild Deceleration. Bernstein expects growth to slow by about 200 basis points in the second half of the year, dropping to around 33%. The largest AI customer, OpenAI, reduced its consumption level upon renewal at the beginning of Q3, and AWS web traffic metrics also hint at a potential "air pocket" in Q4. Growth stock investors expecting continuous acceleration may face a test in the coming quarters.
The Stronger the Q2 Data, the Clearer the Deceleration Signal
In Q2 2026, Datadog's revenue was $1.121 billion, representing a 35.6% year-over-year increase.
This accelerated by 340 basis points sequentially, marking the strongest growth rate since 2022. The company's overall beat amounted to $46 million, and the full-year guidance midpoint was raised by more than triple. Non-AI business revenue grew over 20% year-over-year, and the contribution from AI-native customers continued to expand, jointly driving the acceleration.
Bernstein points out that this "strength" is precisely the peak signal. The growth rate implied by the company's guidance for the second half is already slowing. Even if management continues to "beat and raise" at the recent peak level in subsequent quarters, the growth rate can only stabilize at the current level and cannot move higher. Bernstein expects the growth rate to fall back to around 33% in the second half, approximately 200 basis points lower than Q2.
The inflection point in growth has appeared. The market needs to answer the question: Is this a mild deceleration, or the beginning of a trend slowdown?
Largest AI Customer Reduces Consumption, NRR to Experience a One-Time Dip
The most direct signal for slowing growth comes from the largest AI customer.
Bernstein believes this customer is OpenAI. OpenAI renewed a nine-figure contract at the beginning of Q3, but at a consumption level lower than the previous period. Bernstein lowered the Annual Contract Value (ACV) for this customer to approximately $115 million, slightly up from the roughly $100 million in its previous model, but actual usage declined at the start of Q3. Management confirmed during the earnings call that this customer has been "fully de-risked" in the Q4 guidance. The Q4 guidance includes only the minimum contracted consumption amount for this customer, assuming no over-consumption.
This implies that Net Revenue Retention (NRR) will experience a one-time dip in Q3. Bernstein calculates that NRR has been improving since bottoming in Q2 2023, but OpenAI's reduction will interrupt this trend.
The second-largest AI customer (believed by Bernstein to be Anthropic) is expected to grow 10-15% quarter-over-quarter in Q3 and Q4, with no major recent model releases, indicating growth is in a steady range. Growth rates for other AI customers are well above 100%, but they have a smaller base and are not yet sufficient to offset the gap created by the reduction from the top customer.
AWS Web Metrics Hint at Q4 "Air Pocket"
Bernstein uses AWS SSO web traffic data as a leading indicator for non-AI customer demand.
Web traffic at the end of Q2 was relatively weak, providing a lower starting point for Q3. If weekly trends in Q3 follow the pattern of 2025, year-over-year growth will face pressure. However, Bernstein also notes that if the sequential trend in Q3 is strong enough, this pressure might only be a one-quarter "air pocket," with growth returning to recent trends in Q1 2027.
Bernstein explicitly characterizes this impact as a one-time issue in the report and expects net new ARR growth to rebound to low-to-mid double digits in 2027-2028.
Potential Q4 weakness has data support, but Bernstein believes it is not structural.
Valuation and Crowded Positioning are the Real Risks
Bernstein maintains a Market-Perform rating on Datadog with a target price of $237, implying approximately 16% downside from the current stock price of $283.
The report emphasizes that the rating is unrelated to business fundamentals. Bernstein states in the report that the "Market-Perform" rating is "not a call on the business strength," highlighting that current valuation and deep crowding are the real issues. Bernstein worries that some investors expect growth to continue accelerating, and when growth stabilizes or even declines in the coming quarters, the market will be disappointed, and crowded long positions will unwind.
The current stock price implies a 2026 adjusted P/E of 105x and 82x for 2027, with an EV/Sales multiple of about 17.5x. For a company where growth is about to slow from 35% to 33%, these numbers leave no room for any loosening of expectations.
The long-term logic of observability and security platform convergence, AI-driven security analytics, and the appeal of BYOC architecture to large-scale customers remains valid. Bernstein mentions in the report that Datadog is using the Bring Your Own Cloud (BYOC) architecture to support customers with excessive log volumes, high costs, or security sensitivities, providing analysis and control-plane capabilities without centralizing data in the Datadog SaaS environment. Management sees this as a significant opportunity to win large-scale deployments. The proportion of training workloads is also increasing, with new model training use cases among AI customers in recent quarters, alongside inference and production applications. However, these long-term logics are already fully, if not excessively, priced into the current valuation.
Datadog is a good company, but the stock price has already priced in the good news for the next 12-18 months.
Disclaimer: This article is a compilation and interpretation of a third-party analyst research report (Bernstein, August 6, 2026) combined with publicly available market information. The ratings, target prices, earnings forecasts, and related judgments cited in this article are the views of that analyst, representing only the stance of their affiliated institution. They do not represent the views of Chaoxiang Research and do not constitute any investment advice. The market carries risks; decisions should be made independently. This article should not be used as a basis for trading any securities.
