Datadog (NASDAQ: DDOG) shares plunge 14% as major AI customer cuts usage despite 35% revenue growth

Datadog, Inc. (NASDAQ: DDOG) reported third-quarter 2026 revenue of $1.197 billion, increasing 35% year on year as enterprise customers expanded their use of cloud monitoring, application observability and artificial intelligence-related services. The November 4 results showed non-GAAP operating income reaching $334 million and free cash flow totalling $251 million, while the number of customers generating at least $100,000 in annual recurring revenue increased approximately 20%. However, management disclosed continued reductions in usage by its largest customer, an unnamed AI company, and forecast December-quarter revenue between $1.19 billion and $1.20 billion. Datadog shares fell 14.42% to $188.72 during the November 4 session, reflecting concern that unusually large AI workloads can create significant revenue volatility even when adoption across the wider customer base remains strong.

Why did Datadog’s shares collapse when third-quarter revenue and adjusted earnings remained strong?

Datadog reported quarterly revenue of $1.197 billion, compared with approximately $885.7 million during the corresponding period last year. The increase of about $311.4 million represented 35% annual growth, demonstrating continued demand for its cloud observability and security capabilities. Revenue also exceeded widely reported market expectations, indicating that the negative share-price response was not simply the result of a headline sales disappointment.

The company’s non-GAAP diluted earnings per share reached $0.81, exceeding the expectations reported by financial analysts. GAAP diluted earnings per share were $0.32, reflecting the impact of expenses excluded from adjusted results. The combination of revenue growth and profitability demonstrates that Datadog’s business continues generating substantial operating value.

However, management’s commentary on its largest customer shifted investor attention away from the aggregate results. The company said that the customer had reduced usage during the third quarter and was expected to move towards its minimum contractual commitment during the fourth. The disclosure raised questions about the amount of future revenue associated with that relationship.

Datadog’s revenue is influenced partly by customer consumption of its monitoring and analytical services. A large customer can expand expenditure rapidly as workloads increase, but usage may also decline when operating requirements change. This creates a different revenue profile from contracts where spending is entirely fixed across the agreement period.

The company’s November 4 closing price of $188.72 represented a decline of $31.80 from the preceding session. Trading volume exceeded 13 million shares, reflecting substantial investor activity during the earnings-related sell-off. The movement coincided with the weaker consumption outlook, although broader technology-market conditions also influenced valuations.

How much risk does Datadog face when one AI customer reduces cloud observability consumption?

Datadog’s largest customer is an AI company whose spending increased substantially as its computing workloads expanded. Management previously identified reduced consumption from this account during the second quarter, making the latest disclosure a continuation of an established concern rather than an entirely new event. The important update is that spending is expected to decline further towards the minimum contractual level.

The company has not publicly named the customer in the latest earnings release. Identifying it through speculation would not improve the analysis because the relevant financial issue is the change in consumption rather than an unverified association with a particular AI developer. Datadog’s business remains sensitive to the spending decisions of unusually large accounts.

The minimum contractual commitment provides some revenue visibility, but it does not necessarily preserve the higher consumption revenue previously generated. If actual usage falls towards the minimum, the customer may remain contractually committed while contributing less to incremental growth. This distinction explains why a customer relationship can remain commercially intact even as its revenue contribution declines.

Management indicated that revenue growth excluding the largest customer accelerated during the quarter. This suggests that the broader customer base is expanding at a stronger rate than the consolidated figures alone reveal. However, investors still need to assess whether that wider growth can fully compensate for the decline in a disproportionately large account.

Datadog’s AI customer population has also been expanding. Management described a growing and increasingly diversified group of AI-related customers, including organisations generating substantial annual expenditure. These relationships may reduce dependence on individual accounts over time, although newer customers must reach sufficient commercial scale before they can offset a major reduction elsewhere.

The company’s revenue model therefore contains both an opportunity and a risk. Customers operating complex AI infrastructure may consume substantial observability services, creating attractive growth potential. However, those customers can also adjust usage as they improve efficiency, modify their computing architecture or change operational requirements.

Why does Datadog’s fourth-quarter guidance imply almost no sequential revenue growth?

Datadog expects December-quarter revenue between $1.19 billion and $1.20 billion, establishing a midpoint of $1.195 billion. This is slightly below the September quarter’s reported $1.197 billion, implying broadly flat sequential revenue. The outlook contrasts with the strong annual growth recorded in the latest quarter.

Management’s guidance reflects its current assumptions regarding customer consumption, including the largest AI account. However, the forecast does not establish that all other customers are reducing expenditure. A sufficiently large decline in one account can offset substantial growth across numerous smaller relationships.

The company expects fourth-quarter non-GAAP operating income between $280 million and $290 million. The midpoint of $285 million is below the third-quarter result of $334 million, indicating expectations for a sequential reduction in adjusted operating profit. That outlook adds another dimension to the market’s concerns beyond the revenue forecast alone.

For the full year, Datadog expects revenue between $4.515 billion and $4.525 billion. Management also forecasts non-GAAP operating income between $1.095 billion and $1.105 billion, with adjusted diluted earnings per share between $2.73 and $2.75.

The annual outlook provides evidence that the company still expects substantial growth, but the December-quarter forecast creates a near-term test. Investors will want to see whether the largest customer’s usage stabilises at the anticipated level and whether the rest of the business continues expanding sufficiently to support a stronger revenue trajectory.

The distinction between customer consumption and contracted commitments is particularly important for cloud software companies. A large enterprise contract may establish a minimum spending obligation while allowing additional usage-based expenditure. Revenue growth can therefore accelerate or slow according to customer activity without a corresponding change in the number of customers served.

Can Datadog’s expanding base of large enterprise customers offset the pressure from its biggest AI account?

Datadog reported approximately 4,860 customers generating at least $100,000 in annual recurring revenue as of September 30. That represented an increase of about 800 customers, or nearly 20%, from approximately 4,060 a year earlier. The growth indicates that adoption is broadening among organisations making substantial financial commitments to the platform.

Large customers are commercially important because they may purchase several products covering applications, infrastructure, security and operational monitoring. Expanding product adoption can increase customer spending while strengthening relationships. However, higher customer counts do not reveal how expenditure is distributed across the entire base.

The company is seeking to become a unified platform for managing increasingly complex digital systems. Its products cover infrastructure monitoring, application performance, logs, cloud security and other operational requirements. Customers deploying AI applications may need visibility into both traditional software services and specialised model workloads.

This creates opportunities for additional product adoption. An enterprise using Datadog for conventional application monitoring may expand into GPU observability, AI model monitoring or automated incident management. Such expansion could increase revenue per customer without relying exclusively on entirely new customer relationships.

However, product breadth also introduces commercial challenges. Customers may evaluate overlapping features supplied by cloud infrastructure providers or competing observability platforms. Datadog must demonstrate that its integrated offering produces useful operational outcomes at a cost customers are willing to sustain.

The growth in large accounts provides evidence of continuing market demand, but the next important financial test is consumption growth across those relationships. A wider customer base would become more valuable if expanding usage consistently offset volatility in individual major accounts.

How does Datadog Scout 1.0 change the economics of AI-assisted incident investigation?

Datadog introduced Scout 1.0, its first specialised AI model developed by its internal research team. The model is designed to investigate production incidents and currently powers Bits Investigation in preview. The company is positioning the technology as a way to reduce the effort required to analyse operational problems across complex computing systems.

Incident investigation often involves examining logs, traces, metrics and other information to determine why an application or infrastructure component has malfunctioned. AI models can potentially assist by identifying patterns and connecting information from several monitoring sources. However, useful results depend on the quality of the underlying data and the model’s ability to distinguish relevant signals.

Datadog has said Scout can achieve performance comparable with certain frontier models for selected investigation tasks at a fraction of the inference cost. This is a company performance claim and should not be treated as independently established across every operational environment. The commercial benefit will depend on actual accuracy, processing expenditure and the number of investigations completed.

The strategy could reduce Datadog’s dependence on externally supplied models for some analytical functions. A specialised model may require less computing capacity than a larger general-purpose system when applied to a narrowly defined task. That could support more frequent automated investigations without proportionally higher inference costs.

However, developing and maintaining proprietary models introduces research expenditure and operational responsibilities. Datadog must ensure that the model remains effective as software architectures and threat patterns evolve. Customer adoption will also depend on confidence in the quality and explainability of the resulting findings.

The current preview status is important. The launch establishes technical progress, but it does not demonstrate universal commercial availability or independently quantified customer savings. The next milestone will be broader adoption accompanied by measurable reductions in investigation time and operating costs.

Can Datadog sustain 28% adjusted operating margins and strong free cash flow despite customer volatility?

Datadog reported GAAP operating income of $75 million, equivalent to a 6% operating margin. Non-GAAP operating income reached $334 million, representing a 28% margin. The difference reflects exclusions from adjusted reporting and should remain visible when evaluating the company’s profitability.

Operating cash flow reached approximately $290 million, while free cash flow totalled $251 million. The latter represented approximately 21% of quarterly revenue, indicating substantial cash generation despite the volatility associated with large customer consumption.

The company also reported approximately $5.2 billion in cash, cash equivalents and marketable securities at the end of September. This provides resources for research, acquisitions and ongoing operations without requiring the business to depend entirely on immediate external financing.

However, the November results demonstrate that profitability and revenue growth do not eliminate valuation sensitivity. Datadog’s reported 52-week trading range extends from approximately $98 to $301, while its market capitalisation was about $67.8 billion at the latest close. Investors continue assigning substantial value to its future growth potential.

The immediate financial question concerns the transition from an unusually large customer-driven growth contribution towards a more diversified revenue base. Broader enterprise adoption and strong cash conversion are encouraging, but management must demonstrate that the anticipated decline in the largest account does not create a prolonged slowdown.

The next measurable evidence will be fourth-quarter revenue relative to the $1.195 billion midpoint, stabilisation of the major AI customer’s usage and continued growth among other large enterprise accounts. Sustained profitability alongside renewed sequential revenue growth would provide stronger validation of Datadog’s long-term commercial position.

What should investors know about Datadog’s Q3 2026 results and AI customer concentration?

• Datadog reported third-quarter revenue of approximately $1.197 billion.
• Revenue increased 35% year on year.
• GAAP operating income reached $75 million.
• Non-GAAP operating income totalled $334 million.
• Operating cash flow reached $290 million, with free cash flow of $251 million.
• The number of customers generating at least $100,000 annually increased to approximately 4,860.
• Management expects lower consumption from its largest AI customer during the fourth quarter.
• December-quarter revenue guidance stands between $1.19 billion and $1.20 billion.
• Datadog introduced Scout 1.0 for AI-assisted incident investigation.
• Shares fell 14.42% on November 4, making broader revenue growth a central investor concern.