Datadog Continuous Profiler

Name: Datadog Continuous Profiler
Brand: Datadog
Rating: 4.4 (2770 reviews)

Leader

Datadog provides cloud-scale monitoring and security for metrics, traces and logs in one unified platform.

Est 2010|Valuation $91.4B|Raised $795.9M

LinkedIn 546K|Website|Web traffic 4.6M|FTEs 8,100|HQ New York, New York, US

MarketsApplication Performance Monitoring (APM)

Published May 16, 2026

01Common complaints

Consider alternatives or defer adoption when:

Highly predictable costs and simplified SKU models are a strict requirement. If an organization cannot invest in the governance required to control telemetry costs, consider profiling offerings with per-GB or otherwise more predictable pricing models, or negotiate contract terms that bound spend.
An open-source-first, self-hosted profiling backend is required for full operational control; an option such as Pyroscope may be more appropriate, acknowledging the trade-off of operating additional infrastructure.
The organization only needs a standalone profiling capability and does not require a full observability platform; adopting a full platform primarily for profiling may be excessive.
The environment consists of extreme scale with many short-lived containers and tight cost predictability requirements that make host/container metering difficult to forecast.

Consider alternatives or defer adoption when: - Highly predictable costs and simplified SKU models are a strict requirement. If an organization cannot invest in the governance required to control telemetry costs, consider profiling offerings with per-GB or otherwise more predictable pricing models, or negotiate contract terms that bound spend. - An open-source-first, self-hosted profiling backend is required for full operational control; an option such as Pyroscope may be more appropriate, acknowledging the trade-off of operating additional infrastructure. - The organization only needs a standalone profiling capability and does not require a full observability platform; adopting a full platform primarily for profiling may be excessive. - The environment consists of extreme scale with many short-lived containers and tight cost predictability requirements that make host/container metering difficult to forecast.

Deal disqualifiers (poor fit)

Strict budget constraints and sensitivity to unpredictable usage-based bills.
Preference for open-source/self-hosted tooling and avoidance of SaaS vendor lock-in.
Environments where agent deployment is difficult or prohibited, or where host-based billing is uneconomical (for example, large fleets of constrained edge devices or highly transient containers without predictable metering).

Known limitations surfaced in public sources and issue trackers

Overhead can be material in some scenarios: Reports indicate that profiling can increase CPU consumption in certain large-scale deployments, particularly with specific language runtimes and configurations.
Cost management complexity: Multiple review sources and community discussions point to billing complexity and the need for governance to avoid unexpected spend.
Retention defaults: Profile retention defaults may be limited (e.g., profiles retained for a short default period and derived metrics for a limited window); buyers requiring longer retention should confirm options and costs.

- Overhead can be material in some scenarios: Reports indicate that profiling can increase CPU consumption in certain large-scale deployments, particularly with specific language runtimes and configurations. - Cost management complexity: Multiple review sources and community discussions point to billing complexity and the need for governance to avoid unexpected spend. - Retention defaults: Profile retention defaults may be limited (e.g., profiles retained for a short default period and derived metrics for a limited window); buyers requiring longer retention should confirm options and costs.

Roadmap / future features

Datadog’s product roadmap is periodically updated but not fully public. Buyers should request current roadmaps and clarification on:

Supported languages and profile types and their GA/beta status.
Long-term retention options for profile data and associated pricing.
Any advanced automated insights or anomaly-detection features for profiling and the empirical accuracy of such features.

02Why they buy

Validated strengths drawn from product capabilities and reviewer patterns:

Trace-to-profile correlation: The platform links traces and spans to profiling data when both are enabled, which supports trace-driven root cause workflows.
Continuous production profiling across multiple supported languages: The service collects profiles continuously across environments, subject to language-specific support and configuration.
Profile comparison and tag-based slicing: Comparison features help identify regressions across deployments, time ranges, and environments.

Areas where competitors may have an advantage:

Pricing transparency and predictability: Some competitors price profiling as data ingestion (per-GB) or emphasize hourly host models that are easier to model for certain organizations.
Alternatives that avoid host-count or container-count billing mechanics can be appealing to teams sensitive to host-based SKUs.

Validated strengths drawn from product capabilities and reviewer patterns: - Trace-to-profile correlation: The platform links traces and spans to profiling data when both are enabled, which supports trace-driven root cause workflows. - Continuous production profiling across multiple supported languages: The service collects profiles continuously across environments, subject to language-specific support and configuration. - Profile comparison and tag-based slicing: Comparison features help identify regressions across deployments, time ranges, and environments. Areas where competitors may have an advantage: - Pricing transparency and predictability: Some competitors price profiling as data ingestion (per-GB) or emphasize hourly host models that are easier to model for certain organizations. - Alternatives that avoid host-count or container-count billing mechanics can be appealing to teams sensitive to host-based SKUs.

03Field reports

Because the Continuous Profiler is offered as a capability within Datadog’s broader observability platform, public review platforms typically host reviews for Datadog as a whole. Reviews that reference APM and profiling often reflect experiences with the combined suite, including cost governance and support responsiveness.

Gartner Peer Insights (Datadog overall)

Accessible review excerpts include both positive and critical remarks. Examples of visible comments:

Positive: "The ability to link a failed synthetic API test directly to a backend trace ID has saved us countless hours…" (review excerpt)
Critical: "…the billing model is complex… cost will explode." (review excerpt)

Note: Full reviews on some platforms are gated; this report relies on accessible excerpts.

G2 (Datadog overall)

Recent visible review excerpts include:

Positive: "Its integration capabilities … one stop shop." (review excerpt)
Critical: "...difficult to control and predict costs… usage can grow quickly and lead to unexpected spend." (review excerpt)
Critical: Reports of support being email-first with multi-day response times, which some reviewers noted as impacting response to incidents.
Mixed: Some reviewers report installation complexity in containerized environments.

Recent visible review excerpts include: - Positive: "Its integration capabilities … one stop shop." (review excerpt) - Critical: "...difficult to control and predict costs… usage can grow quickly and lead to unexpected spend." (review excerpt) - Critical: Reports of support being email-first with multi-day response times, which some reviewers noted as impacting response to incidents. - Mixed: Some reviewers report installation complexity in containerized environments.

Capterra (Datadog overall)

Publicly visible excerpts emphasize consolidated visibility:

Positive: "...better one stop shop and one single pane of glass" (review excerpt)

Community discussions (Reddit, GitHub)

Community forums and issue trackers contain candid feedback, often focused on pricing and operational behavior:

Pricing skepticism and concerns about unexpected billing changes appear in several community threads.
Some engineering communities praise the profiler functionality in language-specific contexts.
Technical concerns include reported CPU impact when running continuous profiling in certain large-scale .NET deployments, as recorded in an open issue.

Community forums and issue trackers contain candid feedback, often focused on pricing and operational behavior: - Pricing skepticism and concerns about unexpected billing changes appear in several community threads. - Some engineering communities praise the profiler functionality in language-specific contexts. - Technical concerns include reported CPU impact when running continuous profiling in certain large-scale .NET deployments, as recorded in an open issue.

Why customers choose Datadog Continuous Profiler

Common reasons cited in vendor materials and by reviewers:

Integration between profiling and APM that enables investigation workflows linking traces to profiles.
The convenience of a managed, always-on profiling service that removes the need to operate a separate profiling backend.

Why some customers leave / churn drivers

Common churn drivers reported across review platforms and community sources include:

Pricing unpredictability at scale and complexity of billing SKUs and metering.
Concerns about vendor lock-in and platform complexity when an organization prefers API-first, open-source-first, or more modular approaches.

Datadog case study: Cvent + AWS

Critical evaluation:

This is vendor-published material; the case study does not provide a detailed independent methodology or quantitative baselines for overhead measurements in the visible excerpt.
The case study is useful as an architecture-level example of always-on profiling correlated with traces, but buyers should request measurement details and independent validation where available.

Datadog publishes a customer case study describing the use of APM combined with Continuous Profiler for a large-scale platform. The case study emphasizes always-on profiling and claims of minimal overhead. Critical evaluation: - This is vendor-published material; the case study does not provide a detailed independent methodology or quantitative baselines for overhead measurements in the visible excerpt. - The case study is useful as an architecture-level example of always-on profiling correlated with traces, but buyers should request measurement details and independent validation where available.

Datadog internal savings claim

Datadog has published a company figure attributing a substantial annual recurring cost savings to internal use of its observability products, including profiling.

Critical evaluation:

The figure is an internal claim and is not independently audited in the publication. The claim attributes savings to a combination of tools, making it difficult to isolate the portion attributable specifically to profiling.
Buyers should request a breakdown that distinguishes savings from profiling-driven code changes, infrastructure rightsizing, and other observability initiatives.

Datadog has published a company figure attributing a substantial annual recurring cost savings to internal use of its observability products, including profiling. Critical evaluation: - The figure is an internal claim and is not independently audited in the publication. The claim attributes savings to a combination of tools, making it difficult to isolate the portion attributable specifically to profiling. - Buyers should request a breakdown that distinguishes savings from profiling-driven code changes, infrastructure rightsizing, and other observability initiatives.

04Best fit

Company sizes (typical)

Mid-market to enterprise organizations running multiple services and frequent releases are primary candidates for always-on profiling.
Smaller organizations with acute performance requirements may also derive value but should evaluate incremental platform cost relative to needs.

Industries / verticals

Common adopters include SaaS providers, fintech, marketplaces, and high-traffic consumer web services where latency and cloud costs have measurable business impact.

Roles and personas

Decision makers: VP Engineering, Head of Platform/SRE, Observability lead.
Champions/end users: backend engineers, performance engineers, SREs.

Pain points addressed

Diagnosing production performance regressions that are hard to reproduce locally.
Identifying CPU or allocation hotspots and lock contention in high-throughput services.
Correlating slow traces to specific code locations and deployment changes.

Qualification criteria (good fit)

Organization already using Datadog APM and the Datadog agent across the majority of services.
Mature tagging and versioning practices to support meaningful profile slicing.
Governance and operational discipline to manage telemetry volume and associated costs.

05The pitch

What the product does (objective)

Key capabilities commonly described by Datadog:

Code-level breakdowns (method, class, and line-level) to identify bottlenecks.
Correlation with distributed traces so spans can link to profiling views when both are enabled.
Profile comparison across timeframes, versions, environments, or tag dimensions.
Tag-based filtering across service, version, environment, host, and container dimensions.

Datadog Continuous Profiler is an always-on production profiling capability within the Datadog observability platform. It collects profiling signals such as CPU usage, memory allocations, lock contention, wall time, and some I/O-related profiles depending on language, and presents them as flame graphs and related views. Profiles are tagged by service, version, host, and container to support code-level bottleneck analysis without requiring local reproduction. Datadog describes the profiler as having low overhead and integrates it with APM/tracing to enable navigation from traces and spans to profiles. Key capabilities commonly described by Datadog: - Code-level breakdowns (method, class, and line-level) to identify bottlenecks. - Correlation with distributed traces so spans can link to profiling views when both are enabled. - Profile comparison across timeframes, versions, environments, or tag dimensions. - Tag-based filtering across service, version, environment, host, and container dimensions.

Target market and positioning

Primary buyers/users: SRE and platform teams, backend engineers, performance engineering teams, and teams already running Datadog APM.
Typical environments: production microservices, Kubernetes/containerized environments, and cloud workloads where continuous sampling is desirable.
Positioning: Datadog refers to profiling as an additional pillar of observability alongside metrics, logs, and traces, and offers the capability through an agent-based SaaS model.

Get a comprehensive analysis of Datadog Continuous Profiler including market position, competitive landscape, adoption trends, and peer benchmarks.

Download full report →