Compare

CostPulse vs AWS Cost Anomaly Detection

AWS Cost Anomaly Detection is free, built into AWS, and worth turning on whether or not you ever use CostPulse.

This page is for deciding whether it is all you need. For a lot of teams it is.

AWS Cost Anomaly Detection facts checked against its own pages on October 13, 2026. Sources are at the bottom. Prices and features change: check the source before you decide.

QuestionAWS Cost Anomaly DetectionCostPulse
PriceFree. The optional Amazon Q investigation can add CloudWatch Logs Insights charges.$49/month Starter (1 AWS account) or $99/month Team (up to 5 accounts), after a 14-day trial with no card
How often it checksAbout three times a day. AWS says it can take up to 24 hours to detect an anomaly after the usage.Once a day per account, at your chosen local hour. Billing data lands hours to a day after the spend, so expect an alert within about a day.
New servicesNeeds 10 days of history for a new service before it can detect anomalies on it.Compares against a 7-day baseline, and recent spend is backfilled from Cost Explorer when you connect.
How it decidesMachine learning that accounts for weekly or monthly seasonality and growth. You set absolute or percentage thresholds for alerts.A rule you can read: by default 20% above the 7-day average, with weekday and month-start guards. The threshold is configurable.
Explaining the causeRoot causes ranked by dollar impact across service, account, Region and usage type, in the console. An Amazon Q investigation, started from the console or a Q conversation, can add who made the change from CloudTrail.The alert itself names the service, Region and usage type, the quantity change in billed units (for example +2.4K Hrs), and a tag breakdown when your tags cover the spend.
Where alerts goEmail, or an SNS topic for individual alerts. AWS documents chat delivery to Slack and Amazon Chime through Amazon Q Developer in chat applications, which also supports Microsoft Teams.Slack, Microsoft Teams or email, sent directly. Starter: Slack or Teams. Team: both.
Setup for chat alertsCreate a monitor, an individual-alert subscription, an SNS topic, and a chat configuration in Amazon Q Developer.One read-only CloudFormation stack in your payer account (us-east-1), then connect Slack or Teams.
SummariesDaily summary email (top 10 alerts by cost impact) or weekly summary email.A weekly digest every Monday: your top services and week-over-week drift.
AWS Marketplace chargesNot monitored, except third-party models on Amazon Bedrock. AWS suggests AWS Budgets for the rest.Not something we have tested yet, so we make no claim about it.

When AWS Cost Anomaly Detection is enough

  • You read email or the AWS console every day, and an anomaly summary there is enough.
  • You want detection across many accounts, tags or cost categories at no cost (AWS managed monitors cover those dimensions).
  • You are comfortable setting up SNS and Amazon Q Developer in chat applications for Slack or Teams delivery.
  • You want CloudTrail attribution of who made a change, which the Amazon Q investigation provides and CostPulse does not.
  • Your budget for this is zero. Turn it on. It is the right first step.

What CostPulse adds

  • The explanation arrives in the Slack or Teams message: service, Region, usage type, the change in billed units, and tags. No console login to find out what moved.
  • A detection rule you can read and tune, instead of a model you cannot inspect.
  • Mute a service for 24 hours, or mark an alert as not an anomaly, from the message.
  • Bedrock spikes arrive with model, invocation and token context.
  • No SNS topics or chat configuration to maintain: connect a channel and you are done.

Can you use both?

Yes, and plenty of teams should. CostPulse reads your billing export through a read-only role and does not change anything in Cost Anomaly Detection. Keep AWS's detector on as a second opinion.

The fastest way to judge is to look at real alerts. The demo is a synthetic AWS account with new anomalies every day, no signup.

Sources