AWS Health API and the Support Plan Wall: What Free Gets You, What Business Plan Adds
September 3, 2026
Every AWS account sees the Personal Health Dashboard for free. The API behind it, the one that turns health events into programmatic evidence for claims and audits, is gated behind paid support plans. That distinction decides how you build outage evidence on AWS, and most teams discover it halfway through wiring an integration.
The two-layer access model
AWS Health comes in two shapes. The Personal Health Dashboard, or PHD, is a console page showing events that affect your account: service disruptions, scheduled maintenance, and resource-level issues. Every account gets it, on every support plan, including the free Developer plan.
The AWS Health API is the programmatic interface to the same event data: DescribeEvents, DescribeEventDetails, DescribeAffectedEntities, plus EventBridge integrations that can push health events into your own automation. This is the layer with the plan gate. Access to the AWS Health API requires a Business, Enterprise On-Ramp, or Enterprise support plan; on Developer and Basic support, the API calls fail. The public Service Health Dashboard at health.aws.amazon.com, which reports provider-wide events rather than your account's, is separate from both and has no API for account-specific events at all.
| Layer | What it shows | Who can use it | Cost |
|---|---|---|---|
| Public Service Health Dashboard | Provider-wide public events, per service and region | Everyone | Free |
| Personal Health Dashboard | Events affecting your account and resources | Every AWS account, all support plans | Free, no extra charge |
| AWS Health API | The same account events, programmatically: queries, EventBridge rules, automation | Business, Enterprise On-Ramp, Enterprise, or Unified Operations support plan only | Requires a paid support plan |
| Organizational view | Health events across all accounts in an AWS Organization | Management account with the API | Paid plans only |
Why the API matters for claims
A claim to AWS needs dates, times, and request logs for each incident. The provider's own health event reference, the ARN in the event ID, is not strictly required, but attaching it correlates your logs with AWS's record of the same incident and removes the most common back-and-forth. Programmatically, that correlation looks like: an EventBridge rule on AWS Health events filtered to your services and regions, a Lambda that snapshots the event details and affected entities into an incident bucket, and your probe logs landing in the same folder. When the claim is written, the evidence file is complete and the health event ARN is already in it.
On a free support plan, the same correlation is manual: someone reads the PHD during or after the incident, copies the event details, and files them with the probe logs. Workable for one incident a quarter, fragile at higher tempo, and completely absent at 3 AM unless a human is awake.
Free gets you the dashboard; paid gets you the pipe. Everything automatable about AWS outage evidence flows through the Health API, and the API starts at the Business plan.
What the plans actually cost
The gate sits at Business support. AWS support plans are priced as a monthly minimum plus a percentage of monthly AWS spend, tiered by spend level. The exact number for your account depends on your bill, and it has changed over time, so check the current AWS support pricing page rather than a blog quote, this one included. The relevant shape: for small accounts the Business plan minimum lands in the low hundreds of dollars a month, which is why the decision is not "should we automate evidence capture" but "is our monthly spend high enough that missed credits exceed the plan cost".
The crossover math is the same one from the claim value math: the first credit tier on AWS pays 10% of the affected service's regional bill, so an account spending $5,000 a month on EC2 in one region clears $500 for a first-tier breach month. Against a Business plan minimum in the low hundreds, plus the plan's other benefits (more support cases, faster response), the API access can pay for itself on one caught breach. Below roughly $2,000 of monthly AWS spend, manual PHD checks usually win.
What to do on each side of the wall
If you are on a free or Developer plan: bookmark the Personal Health Dashboard, check it the morning after any anomaly, and rely on external probes for detection. Save event details manually into your incident folder, per the evidence checklist. This is the setup most small stacks actually need.
If you are on Business or above: put an EventBridge rule on AWS Health, filter to the services and regions that carry your bill, and land every event in storage your team can search. The integration is an afternoon, and it converts the claim process from reconstruction to assembly.
UptimeAudit takes the second path as a service: it watches the big four's health feeds including AWS's, without needing your AWS credentials or your support plan tier, matches events to your monitored services and regions, and drafts the claim. The API gate is the reason a service like this exists at all: the events are public per provider, but the convenient pipe is not.