SaaS True Cost Open calculator

Datadog features: tier-by-tier list (2026)

What Datadog does

Datadog pricing depends on product mix: infrastructure hosts, logs, APM, synthetics, RUM, security, and retention can all contribute.

Last verified: 2026-05-27 115 features tracked 7 categories See Datadog pricing →

Core capabilities (88)

See across systems, apps, and services

With turn-key integrations, Datadog seamlessly aggregates metrics and events across the full devops stack.

Source · core integrations

Analyze and explore log data in context

Quickly search, filter, and analyze your logs for troubleshooting and open-ended exploration of your data.

Source · core

Correlate frontend performance with business impact

Prioritize business and engineering decisions with user experience metrics. Visualize load times, frontend errors, and resources for every user session Slice and dice data using custom attributes Troubleshoot quickly with frontend, backend and business metrics

Source · core

Get alerted on critical issues

Datadog notifies you of performance problems, whether they affect a single host or a massive cluster.

Source · core

Get complete visibility across all your environments with

Get complete visibility across all your environments with Datadog Network Monitoring.

Source · core

Capture events and metrics from your own applications

Capture events and metrics from your own applications using our client libraries

Source · core

Company growth

[ edit ] In 2015, Datadog opened a research and development office in Paris. [ 6 ] Datadog at Google Cloud Summit In 2016, Datadog moved its New York City headquarters to a full floor of the New York Times Building to support its growing team, which do

Source · core

Technology

[ edit ] Datadog high-level architecture Datadog uses a Go -based agent, rewritten from scratch since its major version 6.0.0 released on February 28, 2018.

Source · core

In 2016, Datadog moved its New York City

In 2016, Datadog moved its New York City headquarters to a full floor of the New York Times Building to support its growing team, which doubled over the course of the year.

Source · core

As of 2024, the company had approximately 6,500

As of 2024, the company had approximately 6,500 employees in 33 countries, with offices in New York, Boston, Denver, San Francisco, Paris, Dublin, Amsterdam, Sydney, Tokyo, and Singapore.

Source · core

Prior to Datadog's initial public offering (IPO) in

Prior to Datadog's initial public offering (IPO) in September 2019, the company was offered over $7 billion to be acquired by Cisco , but rejected it in favor of going public.

Source · core

Datadog uses a Go -based agent, rewritten from

Datadog uses a Go -based agent, rewritten from scratch since its major version 6.0.0 released on February 28, 2018. It was formerly Python -based, forked from the original created in 2009 by David Mytton for Server Density (previously called Boxed Ice).

Source · core

In 2014, Datadog support was broadened to multiple

In 2014, Datadog support was broadened to multiple cloud service providers including Amazon Web Services (AWS), Microsoft Azure , Google Cloud Platform , and Red Hat OpenShift .

Source · core

Support Plans & Post Sales Services

From basic plans to bespoke offerings, Datadog offers the right level of support & services for any organization.

Source · core

See inside any stack, any app, at any scale, anywhere

575) { var maxHeight = 0; var planNames = this.$el.querySelectorAll('.plan p.plan-name') // Reset all heights and padding planNames.forEach(function(element) { element.style.height = '' element.style.paddingTop = '' }) // Force reflow void planNames[0].offsetH

Source · core

Datadog for Startups

Stay lean while you grow. Get one year of free Datadog Pro. Apply now . { const selectContainer = document.querySelector('.select-container'); const activePlan = document.querySelector('.content-section.active'); const commonQuestions = activePlan.querySelecto

Source · core

Serverless Function Pricing Example

HOUR 1 HOUR 2 HOUR 3 Function A 10 invocations 0 invocations 1M invocations Function B 1 invocation 5M invocations 1M invocations Function C 0 invocations 0 invocations 1 invocation Total Active Functions for Hour 2 active functions (Function A and B) 1 active

Source · core

*Per GB of uncompressed data ingested for processing

*Per GB of uncompressed data ingested for processing, or compressed data scanned for rehydrating. **Billed annually or $ 0.10 on-demand

Source · core

*Minimum retention of 1 month up to 15

*Minimum retention of 1 month up to 15 months **Billed annually or $ 0.075 on-demand

Source · core

*Minimum retention of 3 months, options for 6

*Minimum retention of 3 months, options for 6, 12, and 15 months **Billed annually or $ 0.90 on-demand

Source · core

The retention period is the number of days

The retention period is the number of days your logs are searchable and analyzable.

Source · core

Logs removed from your index(es) do not have

Logs removed from your index(es) do not have to be gone forever! Check out our Log Archives for long-term storage options.

Source · core

Log events come in all shapes and sizes

Log events come in all shapes and sizes, which is precisely why we offer event-based pricing!

Source · core

By committing to a monthly volume of logs

By committing to a monthly volume of logs, you aren’t rate-limited based on a single day s volume. Send as much as you need on any given day without losing visibility.

Source · core

In those cases where you have a big month, not to worry

In those cases where you have a big month, not to worry! You maintain complete visibility and pay for additional services (+50% of the annual price) only on the volume that exceeds your commitment.

Source · core

Volume discounts available

Volume discounts available. Contact us or reach out to your Sales or Customer Success rep for more details.

Source · core

Feature Flags pricing is based on your monthly

Feature Flags pricing is based on your monthly volume of flag configuration requests. Usage below 1 million monthly flag configuration requests is included at no cost.

Source · core

Includes all features in Infrastructure Enterprise and Cloud

Includes all features in Infrastructure Enterprise and Cloud Security Management (CSM) Enterprise

Source · core

Resolve issues faster with end-to-end distributed traces and

Resolve issues faster with end-to-end distributed traces and service health metrics

Source · core

Note: The allotted 150GB per host is summed

Note: The allotted 150GB per host is summed over all the APM hosts. For example, if in a given month you have 2 APM hosts and one ingests 250GBs and the other 50GBs (300GB in total), you will not be billed for ingestion that month.

Source · core

If you need longer retention, please reach out

If you need longer retention, please reach out to your Customer Success or Sales rep to customize your plan.

Source · core

Note 1: The allotted 1 million Indexed Spans

Note 1: The allotted 1 million Indexed Spans per host is summed across all your APM hosts.

Source · core

Note 2: Indexed Spans that are auto-retained by

Note 2: Indexed Spans that are auto-retained by Datadog do not count towards the 1 million Indexed Spans that are allotted per host.

Source · core

On-demand use of APM is counted on an hourly basis

On-demand use of APM is counted on an hourly basis. Both ingested and indexed span allotment are metered each hour and are not aggregated across all use throughout the month.

Source · core

For additional volume, in a given hour, customers

For additional volume, in a given hour, customers will be charged $ 0.10 per additional GB of ingested spans and the indexed spans rate for every additional 1M indexed spans.

Source · core

Datadog for Startups Stay lean while you grow

Datadog for Startups Stay lean while you grow. Get one year of free Datadog Pro.

Source · core

Every tier includes the full agent engineering platform

Every tier includes the full agent engineering platform from day one, including tracing, evaluations, playground, datasets, experiments, annotations, dashboards, and CLI and MCP access. Extended data retention is available.

Source · core dashboard

Trace data is retained for 15 days by

Trace data is retained for 15 days by default, and trace metrics are retained for 15 months.

Source · core

*Billed annually or $ 18 on-demand **Additional containers

*Billed annually or $ 18 on-demand **Additional containers billed at $ 0.002 / hr per container

Source · core

Run multiple tests at the same time across

Run multiple tests at the same time across your test environments to expedite release cycles and reduce test durations

Source · core

Analyze tests across your pipelines in the Synthetic

Analyze tests across your pipelines in the Synthetic Monitoring & Continuous Testing Explorer and search using facets on test durations, commit author, pipeline, and more

Source · core

Datadog Continuous Testing is a comprehensive toolkit in

Datadog Continuous Testing is a comprehensive toolkit in the development cycle that offers integration and end-to-end testing with codeless test setup, self-healing capability, and seamless integration with CI providers (GitHub Actions, Azure DevOps, CircleCI

Source · core integrations

For example, 1 Parallel Testing additional speed offers

For example, 1 Parallel Testing additional speed offers an additional parallel slot, allowing you to run 2 tests in parallel; 2 Parallel Testing additional speed offers 2 additional parallel slots, allowing you to run 3 tests in parallel; and so on.

Source · core

For example, if you want to run 100

For example, if you want to run 100 tests in a CI pipeline and each test takes 10 seconds to run:

Source · core

Parallel tests are tests that run simultaneously in

Parallel tests are tests that run simultaneously in your continuous integration and delivery pipelines to ensure that you can:

Source · core integrations

As an example, if you needed to run

As an example, if you needed to run 60 tests and each test took one minute to run, it would take you one hour to run those tests sequentially.

Source · core

Build apps using a low code drag-and-drop interface

Build apps using a low code drag-and-drop interface with blueprints and over 1,750+ out-of-the box actions to handle complex tasks and take action without having to leave Datadog

Source · core

A Seat is a license for a user

A Seat is a license for a user to actively participate in the on-call or incident management process in Datadog.

Source · core

*Error Tracking is included for APM traces and

*Error Tracking is included for APM traces and RUM events at no additional cost. **Billed annually or $ 36 on-demand

Source · core

Error Tracking is billed according to your monthly

Error Tracking is billed according to your monthly volume of errors. Monthly error volumes below 50k errors will be billed at the flat rate of $ 25 per month or $ 36 on-demand.

Source · core

Understand user behavior across the entire journey with

Understand user behavior across the entire journey with funnels, segments, and pathway diagrams.

Source · core

Boost adoption and revenue by identifying golden paths

Boost adoption and revenue by identifying golden paths and key conversion drivers.

Source · core

Visualize the full user journey across web and

Visualize the full user journey across web and mobile to identify drop-off points, optimize flows, and drive conversion.

Source · core

Uncover UX opportunities by combining session replays with

Uncover UX opportunities by combining session replays with heatmaps to visualize engagement and behavior.

Source · core

Collaborate with full context by linking replays to

Collaborate with full context by linking replays to funnels, retention, and product metrics for faster, aligned decision-making.

Source · core

*Average over the 5 minute interval of active

*Average over the 5 minute interval of active application instances **Billed annually or $ 9 on-demand

Source · core

*Average of the hourly number of active step

*Average of the hourly number of active step functions **Billed annually or $ 7.20 on-demand

Source · core

*Triggered by active workflows. 1 step function state

*Triggered by active workflows. 1 step function state transition is equivalent to 1 step function span. **Billed annually or $ 15 on-demand

Source · core automation

*Triggered by active functions. 1 traced lambda invocation

*Triggered by active functions. 1 traced lambda invocation is equivalent to 1 top level span. **Billed annually or $ 15 on-demand

Source · core

*denominator for normal month would be the number

*denominator for normal month would be the number of hours in that month. For example, in a 30-day month, it would be 720 hours.

Source · core

Azure Functions can run on different platforms depending

Azure Functions can run on different platforms depending on how they re deployed, including App Service Plans, Consumption Plans, Premium Plans, or inside Azure Container Apps.

Source · core

Monitor the availability and performance of your application

Monitor the availability and performance of your application across user sessions

Source · core

Enhance RUM with video-like replays to see user

Enhance RUM with video-like replays to see user issues firsthand across web and mobile

Source · core

Yes—RUM Measure and RUM Investigate are designed to

Yes—RUM Measure and RUM Investigate are designed to work together to give you full visibility and cost control.

Source · core

Note: if you plan on collecting both Sessions

Note: if you plan on collecting both Sessions with and without a Replay, you can consider committing to both “RUM” and “RUM & Session Replay” plans to optimize your usage.

Source · core

* Sampling must be disabled to obfuscate sensitive

* Sampling must be disabled to obfuscate sensitive data **Billed annually or $ 0.045 on-demand

Source · core

A database host is any physical or virtual

A database host is any physical or virtual instance hosting your database on a server that you monitor with Datadog.

Source · core

Note: Azure SQL Database deploys a database in

Note: Azure SQL Database deploys a database in an isolated network; each database is treated as a single host.

Source · core

Billed per GB of uncompressed data ingested **Billed

Billed per GB of uncompressed data ingested **Billed annually or $ 0.12 on-demand

Source · core

Per vCPU pricing, recommended for volumes above 30TB/month

Per vCPU pricing, recommended for volumes above 30TB/month, is also available*** *** Billed annually per vCPU allocated to Observability Pipelines Worker

Source · core

This is suited for organizations with volumes below

This is suited for organizations with volumes below 30TB/month. It is cost effective at lower volumes and easy to understand, as well as allowing you to pay based on your usage as you scale or ramp up early on.

Source · core

However, your bill will increase at a linear

However, your bill will increase at a linear rate as your volume increases, so if you are expecting your observability volume to balloon in the near future, the vCPU pricing may work better for your organization.

Source · core

Even if you re paying based on ingested

Even if you re paying based on ingested volume, by committing to a monthly volume of data, you are not rate-limited based on a single day’s volume. Send as much data as you need on any given day without losing visibility.

Source · core

*No minimum scan size

*No minimum scan size. No scan limits. Customers are billed proportionally to scan volume.

Source · core

Data Observability: Jobs Monitoring for Clusters is billed

Data Observability: Jobs Monitoring for Clusters is billed per host, per hour. This uses an average host count per hour, by sampling the number of unique hosts instrumented every five minutes and taking an average of those samples.

Source · core

Example : Suppose we observe: 1:00-1:05 pm: 100

Example : Suppose we observe: 1:00-1:05 pm: 100 unique hosts 1:05-1:10 pm: 300 unique hosts 1:10-1:15 pm: 150 unique hosts 1:15-1:20 pm: 50 unique hosts 1:20-2:00 pm (last 8 intervals): 0 unique hosts

Source · core

The sum of unique host counts in each

The sum of unique host counts in each five-minute interval for the hour is 100 + 300 + 150 + 50 = 600.

Source · core

Assume we see this usage pattern once a day for 30 days

Assume we see this usage pattern once a day for 30 days. This would result in a monthly total of 50 x 30 = 1,500 billable Data Observability Job Host Hours.

Source · core

Data Observability: Jobs Monitoring is billed per job

Data Observability: Jobs Monitoring is billed per job, per hour for both Serverless and Orchestrators (charged separately).

Source · core

Example : Suppose we observe: Job 1, Run

Example : Suppose we observe: Job 1, Run 1 duration: 3.4 hours Job 2, Run 2 duration: 0.1 hours Job 1, Run 2 duration: 5 hours Job 2, Run 2 duration: 0.15 hours Job 2, Run 3 duration: 0.25 hours

Source · core

The sum of the hours for all the

The sum of the hours for all the runs is 3.4 + 0.1 + 5 + 0.15 + 0.25 = 8.9. We round to the next full hour making the usage for the month 9 Data Observability Job Hours.

Source · core

Hosts where Jobs Monitoring is enabled do not

Hosts where Jobs Monitoring is enabled do not require Infrastructure Monitoring and are not billed as Infrastructure hosts, except in the case of Spark on Kubernetes (K8s).

Source · core

For Spark on Kubernetes (K8s), any node running

For Spark on Kubernetes (K8s), any node running the Datadog agent is counted as an Infrastructure host, but only nodes where Spark jobs are actively running count toward Data Observability Job Host Hours.

Source · core

Automatically generate a Software Bill of Materials (SBOM)

Automatically generate a Software Bill of Materials (SBOM) to maintain visibility into all open-source dependencies

Source · core

Quickly find the source of any vulnerability with

Quickly find the source of any vulnerability with code snippets and speed up the remediation process

Source · core

Secure and maintain custom code and open source

Secure and maintain custom code and open source libraries in your repositories, along with IDE and CI/CD integrations

Source · core integrations

Yes, AWS Fargate is supported by Runtime SCA

Yes, AWS Fargate is supported by Runtime SCA. If Runtime SCA is purchased standalone, the AWS Fargate add-on must be purchased at $ 1 per task, or $ 1.40 on demand.

Source · core

If Runtime SCA is purchased as part of

If Runtime SCA is purchased as part of Code Security, it can run on any AWS Fargate task that has APM installed.

Source · core

AI features (5)

Proactively monitor your user experience

End-to-end user experience visibility in a single platform. Monitor critical user journeys captured with an easy-to-use web recorder Save engineering resources with AI-powered, self-maintaining tests Detect and alert on performance issues for users in various

Source · ai

Unlimited context for unlimited ambition

Only pay based on the volume of LLM spans, even as context grows, reasoning gets heavier, and agents use more tools. How we price Per LLM spans only — not tool spans, workflow spans, or retrieval spans. You pay for calls to the model, nothing else.

Source · ai automation

Per LLM spans only — not tool spans

Per LLM spans only — not tool spans, workflow spans, or retrieval spans. You pay for calls to the model, nothing else.

Source · ai automation

You pay for the actual AI work, not

You pay for the actual AI work, not every surrounding step in the workflow. That makes pricing simpler to forecast as agents grow more complex and context windows expand.

Source · ai automation

The Free tier includes up to 40K LLM

The Free tier includes up to 40K LLM spans per month with 15-day retention, unlimited context, and unlimited evaluations — no credit card required.

Source · ai

Integrations (5)

Instrument your apps, write new integrations

Datadog includes full API access to bring observability to all your apps and infrastructure.

Source · integrations api

Running tests in your CI/CD pipelines today is

Running tests in your CI/CD pipelines today is included with your API and Browser Testing test runs, and tests will run sequentially (one after another).

Source · integrations api

Note: In both cases, your 100 test runs

Note: In both cases, your 100 test runs will contribute to your API or Browser Testing test run quotas.

Source · integrations api

HTTP, SSL, DNS, TCP, and ICMP all fall

HTTP, SSL, DNS, TCP, and ICMP all fall under API test billing. Each hit being made on your endpoint counts as a single test run.

Source · integrations api

The Wireless AP SKU is focused on wireless

The Wireless AP SKU is focused on wireless APs that support OOTB API integrations built by NDM, such as Cisco Meraki.

Source · integrations api

Collaboration (8)

Build real-time interactive dashboards

More than summary dashboards, Datadog offers all high-resolution metrics and events for manipulation and graphing.

Source · collab real-time dashboard

Share what you saw, write what you did

System events and metrics are only part of the story. Datadog is built to give visibility across teams.

Source · collab

Decouple deployment from release to safely control when

Decouple deployment from release to safely control when and how changes reach users

Source · collab

Target specific user and service cohorts to control

Target specific user and service cohorts to control exposure by environment or traffic segment

Source · collab

Gain instant visibility into release impact by correlating

Gain instant visibility into release impact by correlating feature flags with service health

Source · collab

*Pricing is based on usage per 1 million

*Pricing is based on usage per 1 million Monthly Flag Configuration Requests (MFCRs).

Source · collab

For every host hour, customers get an hourly

For every host hour, customers get an hourly allotment corresponding to a share of the monthly allotment (150 GB for ingested spans and 1M indexed) split evenly by hour by month in the current year.

Source · collab

Find vulnerabilities in open source libraries and their

Find vulnerabilities in open source libraries and their dependencies in production in real-time

Source · collab real-time

Reporting and analytics (5)

Get full visibility into modern applications

Monitor, troubleshoot, and optimize application performance. Trace requests from end to end across distributed systems Track app performance with auto-generated service overviews Graph and alert on error rates or latency percentiles (p95, p99, etc.) Instrument

Source · reporting

Datadog, Inc. is an American company that provides

Datadog, Inc. is an American company that provides an observability service for cloud-scale applications, providing monitoring of servers, databases, tools, and services, through a SaaS -based data analytics platform.

Source · reporting

Customers using Test Optimization or Code Security receive

Customers using Test Optimization or Code Security receive one Code Coverage committer for every billed committer of those products, at no additional cost

Source · reporting

Multiple preconfigured compute options to suit your search

Multiple preconfigured compute options to suit your search and analytics workloads

Source · reporting

Track long-term trends with data retention starting at

Track long-term trends with data retention starting at 15 months to assess the lasting impact of product changes.

Source · reporting

Security and compliance (1)

Get full visibility into user access and platform

Get full visibility into user access and platform changes across all services, empowering DevOps and compliance teams with context-rich insights.

Source · security

Administration (3)

Common Questions

How is Code Coverage priced? Code Coverage is priced per committer, per month, based on the maximum number of active committers during the billing period. Do I need to pay separately if I already use Test Optimization or Code Security?

Source · admin

This means that ingested and indexed span use

This means that ingested and indexed span use that goes over the associated APM Host allotment in the same hour will increase billing regardless of span use during other times.

Source · admin

For more information on how seats are assigned

For more information on how seats are assigned, claimed, and billed , see the Incident Response billing documentation .

Source · admin

See Datadog pricing in detail

This page lists what Datadog does. For per-seat math, tier comparison, and the full hidden-fee list, see the Datadog pricing breakdown.

Datadog pricing → Get price alerts

Closest alternatives

Other observability tools buyers shortlist when they evaluate Datadog.

New Relic features Observability platform
Grafana Cloud Not yet covered
Honeycomb Not yet covered

Sources

Features evolve. Confirm directly with the vendor before purchase decisions. This list was last refreshed on 2026-05-27.

Frequently asked questions

What features does Datadog include?

This page tracks 115 verified Datadog features across 7 categories, sourced from Datadog's official site and Wikipedia.

Which tier do I need for Datadog?

Datadog's feature availability varies by plan. Compare the public Datadog tiers, calculator, and hidden fees on the Datadog pricing page.

How does Datadog compare on features to alternatives?

Buyers commonly compare Datadog against New Relic, Grafana Cloud, Honeycomb. Each alternative's feature page is linked below where we cover it.