Datadog, Inc.
Combined average of Moat (AI Resilience), Growth, and Valuation scores.
Moat Score
Datadog is the unified observability platform across infrastructure, APM, logs, security, and AI/LLM workloads — embedded as the operational nervous system at 30,000+ enterprises with deep agent-based instrumentation that compounds switching costs as architectures grow more complex. Q2's +36% print and the Adaptive ML acquisition (RLOps for agentic post-training on Datadog's telemetry) deepen the AI surface without changing the core embedment thesis; Bits AI, AI Guard, and Bits Agent Builder are now generally available after DASH 2026.
Datadog's moat is built on Agent Embedding, Multi-Product Bundle Lock-In, and AI-Native Observability:
- Agent Embedding & Operational Embedding: Datadog's lightweight agent runs on every host, container, serverless function, and Kubernetes pod across customer infrastructure — over 850+ integrations span every cloud, OS, database, and SaaS. Once instrumented, every alert, dashboard, runbook, and on-call rotation references Datadog metrics. Ripping out Datadog requires re-instrumenting thousands of services and rebuilding institutional muscle memory across SRE teams — a multi-year program.
- Multi-Product Bundle: 8+ Products, Land-and-Expand: Customers using 8+ Datadog products represent a steadily growing share of the base, and $100k+ ARR accounts reached ~4,720 (+23% YoY). The cross-product correlation value — APM traces linked to logs, infrastructure metrics, security signals, and now LLM observability — cannot be replicated by single-product competitors (Splunk for logs, Grafana for metrics, New Relic for APM).
- AI-Native Observability Beachhead: AI workloads remain the growth accelerant: Q2 revenue re-accelerated to +36% YoY as customers build and deploy with AI on the Datadog platform. DASH 2026 capabilities are now GA — fully autonomous Bits AI (detect → investigate → remediate), AI Guard against prompt injection, Bits Agent Builder, and Bring Your Own Cloud — while the Adaptive ML acquisition adds RLOps and agentic LLM post-training on Datadog's real-world infra and security data. Generative AI is observability-hungry: prompt logs, token usage, model drift, hallucination rates, and GPU utilization all become billable telemetry.
- Compounding Data Volume from AI & Agents: AI workloads generate exponentially more telemetry than traditional apps — every LLM call produces traces, every agent run produces step-level spans, every model output requires evaluation logs. Datadog's consumption-based pricing captures this expansion natively. Q2's beat (+36% vs a +29–31% guide) confirmed the pull-through; the Q3 guide ($1.135–1.145B, only ~2% sequential) is the next falsifiable read on whether largest-customer conservatism and summer digestion mute that trajectory into H2.
Ten Moats Verdict
Datadog's durability is the agent plus the multi-product bundle: transaction embedding and bundling are strong; the other AI-resilient pillars (proprietary data, system of record, network effects, regulatory lock-in) are intact. AI workloads generate more telemetry than traditional apps, and consumption pricing captures that expansion — confirmed by Q2's +36% re-acceleration and a FY26 raise to ~+30%. Bits AI GA and Adaptive ML deepen attach on the same substrate without turning customer telemetry into a unique dataset. Primary risks are largest-customer concentration (visible in the soft Q3 sequential guide) and Splunk-Cisco bundle pressure; after the ~18% post-print gap those risks are partially in the price rather than fully ahead of it.
70.8 resilient · 74.5 vulnerable · 80/20 = 71.6 · = 72
Datadog dashboards, query language (DDQL), and notebook workflows require fluency that SRE teams build over years; Bits AI Chat and autonomous agents are partially abstracting this, but advanced incident analysis and custom monitor design still require platform expertise.
Customers encode thousands of monitors, SLO definitions, dashboards, runbooks, and incident workflows in Datadog — real switching cost, but it is customer-owned configuration rather than Datadog-owned logic competitors cannot replicate. Same bar as Snowflake SQL/dbt (intact): portable with effort, painful, not a vendor franchise.
Datadog operates on private customer telemetry, not public datasets.
SREs and platform engineers fluent in Datadog command premium salaries and remain in short supply; AI-assisted observability (Bits AI) is augmenting rather than replacing senior reliability engineers.
Datadog sells 20+ products (Infra, APM, Logs, RUM, Synthetics, Security, LLM Observability, GPU Monitoring, Bits AI suite, etc.) on a single agent and unified data model — 8+ product adoption drives outsized retention and expansion. Bits AI GA and Adaptive ML's RLOps layer add another attach surface on the same telemetry substrate; the cross-product correlation (traces ↔ logs ↔ metrics ↔ security signals) remains a structural advantage no single-domain competitor can match.
Datadog ingests trillions of telemetry events daily across 30,000+ customers, and Adaptive ML trains agents on that infra and security data. The flywheel is real at scale, but the underlying data is the customer's and can be dual-homed or exported — not a unique corpus like CrowdStrike's Threat Graph or S&P's benchmarks. Scale of ingestion is not uniqueness.
FedRAMP, HIPAA, SOC 2, ISO 27001, PCI DSS certifications support regulated industries; not as deep a lock-in as ServiceNow's federal moat but meaningful for healthcare and finance customers.
Indirect network effects via 850+ integrations: as more SaaS/cloud providers integrate, Datadog becomes more valuable to customers; partner ecosystem (consultancies, MSPs) deepens implementation density.
Every alert, every incident page, every postmortem, every SLO calculation, and every change deployment flows through Datadog at instrumented enterprises. The agent IS the operational nervous system — every code deploy, container start, and AI inference triggers Datadog telemetry by default. Bits AI autonomous remediation (now GA) tightens that loop from detect → fix without leaving the platform.
Datadog is the operational history for metrics, traces, logs, and incidents at instrumented cloud-native shops — sticky, but OpenTelemetry exists specifically to make that record portable. CrowdStrike rates the equivalent endpoint-telemetry SoR intact; identity, payments, and ServiceNow's CMDB are the strong bar.
Combined average of Moat (AI Resilience), Growth, and Valuation scores.
Moat Score
Datadog is the unified observability platform across infrastructure, APM, logs, security, and AI/LLM workloads — embedded as the operational nervous system at 30,000+ enterprises with deep agent-based instrumentation that compounds switching costs as architectures grow more complex. Q2's +36% print and the Adaptive ML acquisition (RLOps for agentic post-training on Datadog's telemetry) deepen the AI surface without changing the core embedment thesis; Bits AI, AI Guard, and Bits Agent Builder are now generally available after DASH 2026.
Growth Score
Q2 2026 revenue grew 36% YoY to $1.12B — accelerating from +32% in Q1 and crushing the $1.07–1.08B (+29–31%) guide — with non-GAAP EPS of $0.65 (vs $0.58 consensus), non-GAAP operating margin of 23%, and free cash flow of $279M (~25% margin). FY2026 guidance was raised to $4.45–4.47B (~+30% YoY) and non-GAAP EPS to $2.50–2.54, from $4.3–4.34B / $2.36–2.44. Q3 was guided to $1.135–1.145B — a beat versus Street but only ~2% sequential off the Q2 print, which the market read as digestion into a fully priced multiple and gapped the stock ~18%. $100k+ ARR customers reached ~4,720 (+23% YoY). Bits AI, AI Guard, and Bits Agent Builder are now GA, and Adaptive ML brings RLOps onto Datadog's telemetry substrate.
Valuation Score
DDOG closed Aug 5 at $283 (~$101B) into the print, then gapped ~18% to ~$232 (~$83B) on Aug 6 despite beating revenue/EPS and raising FY26 to $4.45–4.47B / $2.50–2.54 EPS — the market focused on a Q3 guide ($1.135–1.145B) that is only ~2% sequential off Q2. At ~$232 the stock sits ~24% below our raised $305 base and trades ~18.5× forward sales / ~92× forward non-GAAP P/E on the raised guide (PEG ~3.1× on ~30% growth) — cheaper than the ~23× / ~119× setup at $285, with the AI-winner re-rating partially unwound into a still-rich but no-longer-fully-priced multiple.
The Observability Embedment Moat
Datadog's moat is built on Agent Embedding, Multi-Product Bundle Lock-In, and AI-Native Observability:
- Agent Embedding & Operational Embedding: Datadog's lightweight agent runs on every host, container, serverless function, and Kubernetes pod across customer infrastructure — over 850+ integrations span every cloud, OS, database, and SaaS. Once instrumented, every alert, dashboard, runbook, and on-call rotation references Datadog metrics. Ripping out Datadog requires re-instrumenting thousands of services and rebuilding institutional muscle memory across SRE teams — a multi-year program.
- Multi-Product Bundle: 8+ Products, Land-and-Expand: Customers using 8+ Datadog products represent a steadily growing share of the base, and $100k+ ARR accounts reached ~4,720 (+23% YoY). The cross-product correlation value — APM traces linked to logs, infrastructure metrics, security signals, and now LLM observability — cannot be replicated by single-product competitors (Splunk for logs, Grafana for metrics, New Relic for APM).
- AI-Native Observability Beachhead: AI workloads remain the growth accelerant: Q2 revenue re-accelerated to +36% YoY as customers build and deploy with AI on the Datadog platform. DASH 2026 capabilities are now GA — fully autonomous Bits AI (detect → investigate → remediate), AI Guard against prompt injection, Bits Agent Builder, and Bring Your Own Cloud — while the Adaptive ML acquisition adds RLOps and agentic LLM post-training on Datadog's real-world infra and security data. Generative AI is observability-hungry: prompt logs, token usage, model drift, hallucination rates, and GPU utilization all become billable telemetry.
- Compounding Data Volume from AI & Agents: AI workloads generate exponentially more telemetry than traditional apps — every LLM call produces traces, every agent run produces step-level spans, every model output requires evaluation logs. Datadog's consumption-based pricing captures this expansion natively. Q2's beat (+36% vs a +29–31% guide) confirmed the pull-through; the Q3 guide ($1.135–1.145B, only ~2% sequential) is the next falsifiable read on whether largest-customer conservatism and summer digestion mute that trajectory into H2.
Ten Moats Verdict
Datadog's durability is the agent plus the multi-product bundle: transaction embedding and bundling are strong; the other AI-resilient pillars (proprietary data, system of record, network effects, regulatory lock-in) are intact. AI workloads generate more telemetry than traditional apps, and consumption pricing captures that expansion — confirmed by Q2's +36% re-acceleration and a FY26 raise to ~+30%. Bits AI GA and Adaptive ML deepen attach on the same substrate without turning customer telemetry into a unique dataset. Primary risks are largest-customer concentration (visible in the soft Q3 sequential guide) and Splunk-Cisco bundle pressure; after the ~18% post-print gap those risks are partially in the price rather than fully ahead of it.
70.8 resilient · 74.5 vulnerable · 80/20 = 71.6 · = 72
Datadog dashboards, query language (DDQL), and notebook workflows require fluency that SRE teams build over years; Bits AI Chat and autonomous agents are partially abstracting this, but advanced incident analysis and custom monitor design still require platform expertise.
Customers encode thousands of monitors, SLO definitions, dashboards, runbooks, and incident workflows in Datadog — real switching cost, but it is customer-owned configuration rather than Datadog-owned logic competitors cannot replicate. Same bar as Snowflake SQL/dbt (intact): portable with effort, painful, not a vendor franchise.
Datadog operates on private customer telemetry, not public datasets.
SREs and platform engineers fluent in Datadog command premium salaries and remain in short supply; AI-assisted observability (Bits AI) is augmenting rather than replacing senior reliability engineers.
Datadog sells 20+ products (Infra, APM, Logs, RUM, Synthetics, Security, LLM Observability, GPU Monitoring, Bits AI suite, etc.) on a single agent and unified data model — 8+ product adoption drives outsized retention and expansion. Bits AI GA and Adaptive ML's RLOps layer add another attach surface on the same telemetry substrate; the cross-product correlation (traces ↔ logs ↔ metrics ↔ security signals) remains a structural advantage no single-domain competitor can match.
Datadog ingests trillions of telemetry events daily across 30,000+ customers, and Adaptive ML trains agents on that infra and security data. The flywheel is real at scale, but the underlying data is the customer's and can be dual-homed or exported — not a unique corpus like CrowdStrike's Threat Graph or S&P's benchmarks. Scale of ingestion is not uniqueness.
FedRAMP, HIPAA, SOC 2, ISO 27001, PCI DSS certifications support regulated industries; not as deep a lock-in as ServiceNow's federal moat but meaningful for healthcare and finance customers.
Indirect network effects via 850+ integrations: as more SaaS/cloud providers integrate, Datadog becomes more valuable to customers; partner ecosystem (consultancies, MSPs) deepens implementation density.
Every alert, every incident page, every postmortem, every SLO calculation, and every change deployment flows through Datadog at instrumented enterprises. The agent IS the operational nervous system — every code deploy, container start, and AI inference triggers Datadog telemetry by default. Bits AI autonomous remediation (now GA) tightens that loop from detect → fix without leaving the platform.
Datadog is the operational history for metrics, traces, logs, and incidents at instrumented cloud-native shops — sticky, but OpenTelemetry exists specifically to make that record portable. CrowdStrike rates the equivalent endpoint-telemetry SoR intact; identity, payments, and ServiceNow's CMDB are the strong bar.
Growth Analysis
Growth Drivers
Key Risk
Q3 guide implies only ~2% sequential growth off a $1.12B print — consistent with management's stated conservatism on the largest customer (OpenAI per analyst estimates). If that digestion persists into Q4 or AI-lab training spend pauses, headline growth could settle in the mid-20s even as the FY26 raise sticks; combined with Splunk-Cisco AI bundle pressure on enterprise renewals, that would re-rate a multiple that just compressed from ~23× to ~18.5× forward sales on the post-print gap.
Score Derivation
88.0 base + 4.0 trajectory − 5 risk = 87
cagrEstimate 24–30% (midpoint 27% → base ~88) anchors on raised FY26 ~+30% with Q2 at +36% and a conservative Q3 sequential; all three drivers accelerating (+4 trajectory) for core re-acceleration, AI-native pull-through, and Bits AI GA / Adaptive ML; marginTrend stable (0); keyRiskSeverity moderate (−5) for Q3 sequential soft-patch / largest-customer concentration into an ~18.5× forward-sales multiple after the post-print gap = 87
Price Scenarios (12–24 Months)
Valuation Multiples
| Trailing P/E (GAAP) | ~130× |
| Forward P/E (NTM, non-GAAP) | ~92× |
| PEG Ratio | ~3.1× |
| Price / Sales (NTM) | ~18.5× |
| Price / FCF | ~68× |
The post-print gap moved Datadog from ~23× to ~18.5× forward sales — still an AI-infrastructure premium, but no longer priced as if every sequential print must clear a double-digit bar. Against revised scenarios (bear $170 / base $305 / bull $420), ~$232 offers ~31% upside to base and ~81% to bull versus ~27% downside to bear: a constructive asymmetry the $285 pre-print tape did not have, provided the Q3 sequential soft-patch is conservatism rather than a lasting digestion of AI-lab spend.
Approximate figures as of August 6, 2026.
Where We Are vs Targets
Loading live price…
Revised August 6, 2026 after the Q2 beat-and-raise and ~18% post-print gap (prior bear $175): largest customer digests further; AI-lab spend pauses; growth decelerates below 22% and the multiple compresses toward ~12× forward sales.
- Top customer (OpenAI, per analyst estimates) renegotiates pricing aggressively, digests training-related spend, or builds in-house observability — removing 4–6 points of headline growth in FY2027; Q3's ~2% sequential guide proves the start of a multi-quarter digestion rather than conservatism
- Q3/Q4 sequential growth stays muted; AI-native cohort spending digests after the 2025–26 training buildout; Splunk-Cisco AI observability bundle wins large enterprise renewals, pushing net retention below 110%
- Multiple compresses from ~18.5× to ~12× forward sales as growth decelerates toward low-20s and the AI-winner premium partially unwinds — ~$60B market cap on ~$5.0B FY2027 revenue
Revised August 6, 2026 (prior base $290): FY2026 lands at the raised $4.45–4.47B guide; Q3's soft sequential is management conservatism that H2 converts; FY2027 sustains ~25% growth and the stock holds a mid-to-high-teens forward sales multiple.
- FY2026 revenue lands at $4.46B+ (~30% growth) with Q3–Q4 delivering on AI-native demand despite the conservative sequential guide; FY2027 consensus moves to $5.6B+
- AI-native expansion continues; LLM Observability + GPU Monitoring + hyperscaler superintelligence-lab contracts contribute $300M+ run-rate; Bits AI autonomous suite begins measurable attach after GA
- Stock sustains ~16–18× forward sales on FY2027 revenue of ~$5.6B (~$95–100B+ market cap), consistent with a partial re-rating back from the post-print gap
Revised August 6, 2026 (prior bull $400): Datadog becomes the standard observability-and-autonomy layer for the AI economy; AI telemetry re-accelerates growth to 30%+ in FY2027; Bits AI + Adaptive ML emerge as a material new ARR line.
- AI workload telemetry re-accelerates total revenue growth to 30%+ in FY2027 as agentic systems and superintelligence-lab training generate exponentially more observability data — Q3's soft sequential proves one-off conservatism
- Bits AI (Code, Detection, Agent Builder, infra remediation) plus Adaptive ML RLOps scales into a $500M+ ARR product line and becomes the de facto AIOps/autonomy layer for SRE teams
- Net retention rises above 130% as 8+ product customers expand; multiple re-rates to ~22–24× forward sales (~$140B+ market cap on ~$6.2B FY2027 revenue)