Data Sources

No single signal is enough to determine whether a service is healthy or down. Isitreallydown? blends several independent sources to make a more honest call.

We monitor official public status pages where they're available. When a provider posts an incident, that's a strong signal — though component-scoped incidents may not affect every user.

For cloud providers, Isitreallydown? may probe public regional endpoints to determine whether those endpoints are reachable. These checks are one signal among many and do not represent a complete health guarantee for the provider.

Reports submitted by visitors are aggregated to detect spikes above a typical baseline. A single report doesn't move the needle; consistent reports from different sessions and regions do.

We watch for unusual regional behavior using public internet measurement signals to help distinguish localized issues from global outages.

Public news and incident reporting can provide context — for example, news of physical damage to infrastructure or major provider incidents.

AI may be used to summarize signals, classify incident confidence, group related findings, and explain why something is being watched. AI output is reviewed by system rules and should be treated as informational.

Isitreallydown? separates service health into:

• Healthy — no problems detected.
• Watching — something unusual was detected, but it may be transient or unconfirmed.
• Degraded — multiple signals suggest reduced functionality.
• Confirmed outage — only used when strong supporting evidence exists.

External sources can be delayed, rate-limited, missing, or wrong. We surface contradicting signals where we see them so you can judge the picture for yourself.

If you believe a signal is incorrect or missing, contact support@downrightnow.app.