Cloud Cost Recommendations

Docs > Cloud Cost Management > Cloud Cost Recommendations

Overview

Cloud Cost Recommendations provides recommendations on reducing your cloud spending by optimizing the usage of your cloud resources. Datadog generates a set of recommendations by combining your observability data with your underlying cloud provider billing data to identify orphaned, legacy, or over-provisioned cloud resources.

Recommendations are run on a daily basis and are automatically refreshed in your account as soon as the recommendations are released.

For all resources, cloud cost metrics are also pulled for that resource
For all AWS resources besides Kubernetes and EC2, AWS metrics are also pulled from AWS CloudWatch

Overview tab with potential monthly savings, potential annual savings, and total number of open cases on the Cloud Cost Recommendations page

You can see the detailed logic for each recommendation type, along with observability metrics or cost data shown on this page.

Recommendations support Tag Pipelines, allowing you to filter, group, and analyze recommendations using your organization’s standardized tags. Any tag rules configured in Tag Pipelines are automatically applied to recommendations and are normalized.

You can also query your recommendations from an AI agent with the cost_recommendations tool in the Datadog MCP Server.

Recommendation categories

Below are the available cloud cost recommendation categories and their descriptions.

Recommendation Category	Description
Terminate	Resources with signals that the resource is unused or very low utilization signals. Consider terminating or deleting these resources to reduce your costs.
Migrate	Resources with moderately low utilization signals or other inefficiencies. Consider adjusting the instance type or other parameters.
Downsize	Resources that are under-utilized or over-provisioned. Consider adjusting the size or other parameters to reduce costs.
Purchase	Resources with on-demand charges and extended uptime. Purchasing a reservation or Savings Plan can reduce the amortized cost of the resource.
Configure	Resources with configuration options that can be adjusted to reduce costs without changing capacity or terminating the resource.

Prerequisites

The following are requirements necessary to receive Cloud Cost recommendations:

Cloud provider accounts (for all desired Cloud Cost recommendations)
AWS integration and resource collection (for AWS recommendations)
Azure integration and resource collection (for Azure recommendations)
GCP integration and resource collection (for GCP recommendations)
Datadog Agent integration (for Downsize recommendations)

Setup

For each cloud account that you would like to receive recommendations for:

Configure Cloud Cost Management to send billing data to Datadog.
- For Azure, this requires using the App Registration method to collect billing data.
Enable resource collection for recommendations.
- For AWS, enable resource collection in the Resource Collection tab on the AWS integration tile.
- For Azure, enable resource collection with the appropriate integration. If your organization is on the Datadog US3 site, the Azure Native Integration enables this automatically through metrics collection. For all other sites, enabling resource collection within the Azure integration tile is required.
- For GCP, enable resource collection in the Resource Collection tab on the Google Cloud Platform integration tile.
Install the Datadog Agent (required for Downsize recommendations).

Note: Cloud Cost Recommendations supports billing in customers’ non-USD currencies.

Recommendation statuses

Assign a status to each recommendation to track cost optimization progress across your teams. Statuses persist when recommendations regenerate daily. You don’t need to re-triage the same recommendations.

Status	Description
Open	(Default) The recommendation has not been triaged.
In Progress	Work is underway to address this recommendation.
Completed	The recommended action has been taken or is no longer relevant.
Dismissed	No work is planned for this recommendation over the time frame specified when dismissing.

Filter recommendations by status

Use the status tabs at the top of the Cloud Cost Recommendations page to filter the list by status. The available tabs are Open, In Progress, Completed, and Dismissed. Each tab displays the total estimated savings for recommendations in that status.

Track savings by status

Each status tab displays the total estimated savings for recommendations in that status:

Open: Potential savings from recommendations that have not been triaged.
In Progress: Estimated savings from recommendations with work underway.
Completed: Realized savings from recommendations where the recommended action has been taken.
Dismissed: Estimated savings from recommendations that have been dismissed.

Change a recommendation status

You can change a recommendation status in three ways:

Bulk update: Select one or more recommendations in Active Recommendations, then choose a status from the toolbar above the table to apply it to all selected recommendations.
From the table: Use the status dropdown in the Status column to select a new status directly from the recommendation list.
From the side panel: Click a recommendation to open the side panel, then use the status dropdown to select a new status.

Recommendation action-taking

You can act on recommendations to save money and optimize costs. Cloud Cost Recommendations support Jira, 1-click Workflow Automation, and Datadog Case Management. Unused EBS and GP2 EBS volume recommendations also support 1-click Workflow Automation. See the following details for each action-taking option:

Jira: Create Jira issues directly from the recommendation side panel or by selecting multiple recommendations in the Active Recommendations list and clicking Create Jira issue. Created issues are tagged and link back to the recommendation in Datadog.
To filter recommendations by Jira status, use the following query options:
- @jira_issues.issue_key:* - Show only recommendations with a Jira issue
- -@jira_issues.issue_key:* - Show only recommendations without a Jira issue
- jira_issues.issue_key:ABC* - Filter by specific Jira project prefix
Bits Code code fixes: Code fixes are available for applicable S3 and DynamoDB recommendations, as well as the Downsize Kubernetes Deployment recommendation. In these situations, Bits Code creates production-ready pull requests to implement cloud resource changes and cost optimizations in Terraform or Helm charts, respectively. Set up Bits Code to use this feature.
1-click Workflow Automation actions: Actions are available for a limited set of recommendations, allowing users to execute suggested actions, such as clicking Delete EBS Volume, directly within Cloud Cost Management.
Cost Optimization Automation: Set up automations that act on recommendations continuously on a recurring schedule. Automations are scoped to specific accounts, regions, and tags and include safeguards such as pre-action snapshots and optional human approval through Slack or Microsoft Teams.
Datadog Case Management: Users can go to the recommendation side panel and click Create Case to generate a case to manage and take action on recommendations.
Dismiss: Use Dismiss in the recommendation side panel to hide a recommendation for a chosen time frame and provide a reason. Dismissed recommendations move to the Dismissed tab.

Recommendation and resource descriptions

Recommendation Category	Cloud Provider	Resource Type	Recommendation Type	Recommendation Description	Recommendation Prerequisites
Configure	Anthropic	Anthropic API Key	Enable Anthropic Prompt Caching	Identifies Anthropic API keys with no prompt caching usage and recommends enabling prompt caching to reduce input token costs.	Anthropic integration
Migrate	AWS	Auto Scaling Group	Migrate ASG Legacy Instances	An Auto Scaling group that includes legacy instance types.
Downsize	AWS	Auto Scaling Group	Reduce Minimum Capacity	An Auto Scaling group with a minimum capacity of instances that can be reduced.	Datadog Agent
Terminate	AWS	CloudTrail Trail	Delete Unnecessary CloudTrail trails	CloudTrail trails with paid events can be deleted to reduce costs.
Terminate	AWS	DynamoDB Table	Delete DynamoDB Global Secondary Index	A DynamoDB table’s Global Secondary Index (GSI) has 0 consumed reads.
Terminate	AWS	DynamoDB Table	Delete DynamoDB Table	A DynamoDB table has 0 consumed reads and 0 consumed non-replica writes.
Terminate	AWS	DynamoDB Table	Delete Extra On-Demand Backups	A DynamoDB table has charges for more than 2 on-demand backups.
Downsize	AWS	DynamoDB Table	Downsize DynamoDB Capacity	A provisioned DynamoDB table using less than 80% of its read and write capacity more than 80% of the time.
Migrate	AWS	DynamoDB Table	Migrate DynamoDB to Infrequent Access Table Class	Migrating to the Infrequent Access (IA) table class offers more potential savings from storage rates compared to the additional costs from capacity rates.
Migrate	AWS	DynamoDB Table	Migrate DynamoDB to On-Demand Capacity Mode	A provisioned DynamoDB table has an hourly read and write capacity consumption below 18% at least once in the last two weeks.
Migrate	AWS	DynamoDB Table	Migrate DynamoDB to Provisioned Capacity Mode	An on-demand DynamoDB table has an hourly read and write capacity consumption that is always greater than 18%.
Migrate	AWS	DynamoDB Table	Migrate DynamoDB to Standard Table Class	Migrating to the Standard table class offers potential savings from capacity rates compared to the additional costs from storage rates, or it uses the Standard table class’ free tier for storage.
Terminate	AWS	EBS Snapshot	Delete Old EBS Snapshots	EBS Snapshots that are at least 90 days old and can be deleted.
Terminate	AWS	EBS Volume	Delete Unattached EBS Volume	Volume that is not attached to an EC2 instance.
Terminate	AWS	EBS Volume	Delete Unused EBS Volume	Volume that has no read or write activity.
Downsize	AWS	EBS Volume	Downsize EBS Volume Provisioned IOPS	An EBS volume using less than 80% of the provisioned IOPS for reads and writes.	Amazon EC2 integration
Downsize	AWS	EBS Volume	Downsize EBS Volume Provisioned Throughput	An EBS volume using less than the configured threshold of the provisioned throughput for reads and writes.	Amazon EC2 integration
Downsize	AWS	EBS Volume	Downsize EBS Volume Storage Capacity	An EBS volume with less than 20% of its storage capacity used.	Amazon EC2 integration
Migrate	AWS	EBS Volume	Migrate EBS Volume from GP2 to GP3	EBS volumes that are GP2 and can be upgraded to GP3 for cost reduction and performance improvement.
Migrate	AWS	EBS Volume	Migrate EBS Volume from IO1 to GP3	EBS volumes that are IO1 and can be upgraded to GP3 for cost reduction and performance improvement.
Terminate	AWS	Elastic IP	Delete Idle Elastic IP	Elastic IP addresses with idle charges in your AWS cost and usage report.
Downsize	AWS	EC2 Instance	Downsize EC2 Instance	EC2 instances with CPU and memory utilization less than the available resources of the next smallest instance in the family. Without the Datadog Agent, this recommendation is generated using CloudWatch metrics.	Datadog Agent
Migrate	AWS	EC2 Instance	Migrate EC2 Instance	EC2 Instances of a previous generation that can be upgraded to a newer instance type.
Migrate	AWS	EC2 Instance	Migrate EC2 Instance to Graviton Type	EC2 Instances that can be migrated to an equivalent Graviton instance type.
Terminate	AWS	EC2 Instance	Terminate EC2 Instance with Stuck Node	EC2 instances hosting Kubernetes nodes that are stuck in the pending phase, indicating the node is not functioning properly.	Datadog Agent
Terminate	AWS	EC2 Instance	Terminate Unused EC2 Instance	EC2 instances with CPU and memory utilization under a customizable threshold. Without the Datadog Agent, this recommendation is generated using CloudWatch metrics.	Datadog Agent
Terminate	AWS	ECR Repository	Delete ECR Repository	ECR repository with no pull activity that can be deleted to reduce costs.
Downsize	AWS	ECS Task Definition	Downsize ECS Task Size	An ECS task using less than 50% of its requested CPU or memory.	Container Monitoring
Terminate	AWS	ElastiCache Cluster	Delete Unused ElastiCache Cluster	An ElastiCache Redis cluster with no cache hits and no replication or a Memcached cluster with no cache hits.
Terminate	AWS	OpenSearch Domain	Delete Unused OpenSearch Domain	OpenSearch domain with no request activity.
Terminate	AWS	Classic Load Balancer	Delete Unused Classic Load Balancer	Classic Elastic Load Balancer with no active connections that is not attached to an EC2 instance.
Terminate	AWS	Load Balancer	Delete Application Load Balancer	An application load balancer with no traffic being processed.
Terminate	AWS	Load Balancer	Delete Network Load Balancer	A network load balancer with 0 processed bytes.
Downsize	AWS	Lambda	Downsize Lambda Provisioned Concurrency	AWS Lambda function with over-allocated provisioned concurrency.
Terminate	AWS	CloudWatch Log Group	Delete Lambda CloudWatch Logs and write permissions	Remove write permissions for Lambda CloudWatch Logs to prevent further unnecessary logging.
Downsize	AWS	CloudWatch Log Group	Set CloudWatch Logs Retention Policy	Reduce CloudWatch Logs storage costs by setting appropriate retention policies.
Terminate	AWS	MQ Broker	Delete Unused MQ Broker	An MQ broker with 0 connections.
Terminate	AWS	RDS Instance	Delete Unused RDS Instance	RDS instance with 0 database connections and 0 replica lag.
Downsize	AWS	RDS Instance	Downsize RDS Instance	RDS instances that AWS Compute Optimizer suggests downsizing to a smaller instance type.	AWS Cost Optimization Hub permissions
Downsize	AWS	RDS Instance	Downsize RDS Instance Provisioned IOPS	RDS instances using less than 80% of provisioned IOPS over the past two weeks.
Migrate	AWS	RDS Instance	Migrate RDS Instance Engine	An RDS running an engine version that is no longer supported and incurring extended support charges.
Migrate	AWS	RDS Instance	Migrate RDS Instance to Graviton	RDS Instances that can be migrated to an equivalent Graviton instance type.
Terminate	AWS	Redshift Cluster	Delete Redshift Cluster	Redshift cluster with 0 database connections.
Migrate	AWS	S3 Bucket	Clean up old versions to reduce storage costs	A bucket with versioning enabled has significant storage costs from old object versions.	Storage Management
Terminate	AWS	S3 Bucket	Delete S3 Bucket	An S3 bucket with minimal storage costs and no meaningful object API usage (Get, Put, Copy, Head, or multipart upload activity).	Cloud Cost Management or Storage Management
Terminate	AWS	S3 Bucket	Delete S3 noncurrent version objects	A standard S3 bucket without a noncurrent version expiration lifecycle and that does not serve a website contains noncurrent version storage bytes older than 30 days.	Storage Lens
Terminate	AWS	S3 Bucket	Delete abandoned S3 multipart uploads	S3 buckets with incomplete multipart uploads older than 7 days that are consuming storage space.	Storage Lens
Migrate	AWS	S3 Bucket	Reduce small file count to reduce storage costs	A bucket has a significant percentage of small files in infrequent access storage classes, increasing storage costs due to minimum billing size.
Migrate	AWS	S3 Bucket	Transition S3 IA and Glacier objects to Intelligent-Tiering	A bucket has large early deletion charges.
Migrate	AWS	S3 Bucket	Transition S3 Standard objects to Intelligent Tiering	A bucket’s costs are almost entirely in per-GB standard storage, but GET requests indicate few objects are accessed.
Migrate	AWS	S3 Bucket	Transition S3 objects to Infrequent Access by Prefix	A bucket prefix’s costs are almost entirely in per-GB standard storage, but GET requests indicate few objects in the prefix are accessed.	Storage Management
Migrate	AWS	S3 Bucket	Transition old Standard-class data	A bucket prefix has old Standard-class data with no lifecycle transition rule to move it to cheaper storage.	Storage Management
Terminate	AWS	VPC NAT Gateway	Delete Unused NAT Gateway	A NAT Gateway that has no bytes sent through it.
Downsize	AWS	VPC NAT Gateway	Reduce NAT Gateway Cross-Zone Transfers	Resources that need a NAT gateway should use one that is in the same availability zone, or they can incur unnecessary cross-zone transfer charges.
Downsize	AWS	VPC NAT Gateway	Reduce NAT Gateway Within-VPC Transfers	Resources in the same VPC should avoid communicating with each other through a NAT gateway because that incurs unnecessary NAT gateway processing charges.	NPM
Terminate	Azure	AKS Cluster	Delete AKS Cluster	An AKS cluster with less than 5% CPU usage.
Terminate	Azure	App Service Plan	Delete Unused App Service Plan	App Service Plans with no deployed apps that Azure Advisor recommends deleting.
Migrate	Azure	Managed Disk Snapshot	Migrate Disk Snapshot to Standard Storage	Snapshot is stored on Premium storage. Migrating to Standard storage reduces cost by 60% with no change in data durability.
Downsize	Azure	Container App	Downsize Container App	A Container App has higher than necessary minimum replicas.
Terminate	Azure	Data Explorer Cluster	Terminate Unused Stopped Data Explorer Cluster	A cluster is considered unused and stopped if it has been stopped for at least 60 days. The recommendation is to delete the cluster to reduce cost.
Terminate	Azure	Load Balancer	Delete Load Balancer	Load Balancer with 0 bytes transferred.
Terminate	Azure	Managed Disk	Delete Unattached Managed Disk	Managed disk is unattached and can be deleted.
Terminate	Azure	Managed Disk	Delete Unused Managed Disk	Managed disk with no read/write operations, which can be deleted.
Downsize	Azure	Managed Disk	Downsize Managed Disk IOPS	Managed disk using less than the configured threshold of provisioned IOPS.
Downsize	Azure	Managed Disk	Downsize Managed Disk Throughput	Managed disk using less than the configured threshold of provisioned throughput.
Terminate	Azure	MySQL Database	Delete Database for MySQL	Database server with no connections, which can be terminated.
Terminate	Azure	Azure Managed Redis	Delete Azure Managed Redis	Azure Managed Redis cache with no get or set operations.
Terminate	Azure	SQL Server	Delete SQL Server	SQL Server with no connections, which can be terminated.
Terminate	Azure	SQL Server Database	Delete SQL Server Database	SQL Server Database with no successful connections and very minimal CPU, which can be terminated.
Downsize	Azure	SQL Server Database	Downsize SQL Server Database DTU	SQL Server database with low DTU usage that can be downsized.
Downsize	Azure	SQL Server Database	Downsize SQL Server Database Storage	SQL Server Database using less than 20% of provisioned storage capacity.
Terminate	Azure	Storage Account	Delete Storage Account	Storage Account with no transactions and no used capacity over the last 14 days.
Terminate	Azure	VM Instance	Delete Azure VM Instance	VM instance with less than 5% user CPU and over 90% usable memory.	Datadog Agent
Downsize	Azure	VM Instance	Downsize Azure VM Instance	VM instance that can be downsized to a smaller instance type.	Datadog Agent
Migrate	Azure	VM Instance	Migrate Azure VM Instance to Arm	VM instance that can be migrated to an equivalent Arm instance type for a lower price.
Downsize	Azure	VM Scale Set	Downsize Azure VM Scale Set	VM instances with low usage that can be downsized
Terminate	Azure	VM Scale Set	Shutdown Azure VM Scale Set	VM instances with low usage that can be shutdown.
Downsize	AWS	Databricks Cluster	Downsize Databricks All-Purpose	Identifies overprovisioned all-purpose Databricks clusters and suggests rightsizing to smaller instance types to reduce costs.
Downsize	Azure	Databricks Cluster	Downsize Databricks All-Purpose	Identifies overprovisioned all-purpose Databricks clusters and suggests rightsizing to smaller instance types to reduce costs.
Downsize	GCP	Databricks Cluster	Downsize Databricks All-Purpose	Identifies overprovisioned all-purpose Databricks clusters and suggests rightsizing to smaller instance types to reduce costs.
Downsize	AWS	Databricks Cluster	Downsize Databricks Job	Identifies overprovisioned Databricks jobs and suggests rightsizing to smaller instance types to reduce costs.
Downsize	Azure	Databricks Cluster	Downsize Databricks Job	Identifies overprovisioned Databricks jobs and suggests rightsizing to smaller instance types to reduce costs.
Downsize	GCP	Databricks Cluster	Downsize Databricks Job	Identifies overprovisioned Databricks jobs and suggests rightsizing to smaller instance types to reduce costs.
Terminate	GCP	Compute Address	Delete Unused Compute IP Address	Unused compute IP addresses can be deleted.
Terminate	GCP	Compute Disk	Delete Unattached Compute Disk	Compute disks that are unattached and can be deleted.
Terminate	GCP	Compute Disk	Delete Unused Compute Disk	Compute disks that are unused and can be deleted.
Terminate	GCP	Compute Global Address	Delete Unused Compute Global IP Address	Unused compute global IP addresses can be deleted.
Terminate	GCP	Compute Instance	Delete Compute Instance	Compute instance with low CPU usage, high available memory, and minimal network activity.	Datadog Agent
Downsize	GCP	Compute Instance	Downsize Compute Instance	Compute instance with low CPU and memory usage that can be downsized to a smaller instance type.	Datadog Agent
Downsize	GCP	Compute Instance Group	Reduce Minimum Capacity	A Compute Instance Group Autoscaler with a minimum capacity of instances that can be reduced.
Terminate	GCP	CloudSQL Instance	Delete Cloud SQL Instance	CloudSQL instances with minimal usage that can be deleted.
Downsize	GCP	CloudSQL Instance	Downsize CloudSQL Database	CloudSQL instances that are over-provisioned and can be downsized.
Terminate	GCP	Storage Bucket	Delete Noncurrent Cloud Storage Objects	Cloud Storage buckets that benefit from lifecycle rules to automatically delete noncurrent object versions.
Migrate	GCP	Storage Bucket	Transition Cloud Storage Bucket to Autoclass	Objects in the storage bucket can be automatically migrated to archival tiers for better rates.
Downsize	AWS	Kubernetes Cluster	Reduce Cluster Idle	Kubernetes clusters with high CPU or memory cluster idle.	Datadog Agent
Downsize	Azure	Kubernetes Cluster	Reduce Cluster Idle	Kubernetes clusters with high CPU or memory cluster idle.	Datadog Agent
Downsize	GCP	Kubernetes Cluster	Reduce Cluster Idle	Kubernetes clusters with high CPU or memory cluster idle.	Datadog Agent
Downsize	AWS	Kubernetes Deployment	Downsize Deployment	Containers are using only a fraction of their requested CPU or memory.	Datadog Agent
Downsize	Azure	Kubernetes Deployment	Downsize Deployment	Containers are using only a fraction of their requested CPU or memory.	Datadog Agent
Downsize	GCP	Kubernetes Deployment	Downsize Deployment	Containers are using only a fraction of their requested CPU or memory.	Datadog Agent