Automate detection, keep updates human
The gap between an outage starting and a status page reflecting it is where trust is won or lost. Tying the page to automated monitoring — so a detected outage flips a component to "degraded" or "down" within minutes, not whenever someone gets around to it — closes most of that gap. The written explanation of what's happening is still worth having a human compose, but the initial "something is wrong" signal shouldn't wait on a person noticing first.
Break status down by component, not one big red light
A single "everything is fine / something is wrong" toggle for an entire product hides useful information. If email delivery is degraded but the core application is fine, a single combined status either falsely alarms users of the unaffected parts or — worse — hides a real problem under an overall "operational" status because most components are fine. Separate status per meaningful component (API, web app, email, each major integration) lets affected users find their specific answer immediately without reading through unrelated incident history.
Historical incident data is worth more than it looks
A visible incident history — not just current status — lets a customer evaluating reliability (or already skeptical after a recent problem) actually verify a track record instead of taking a vendor's word for it. It's also the practical foundation for the postmortem and SLA-credit conversations that follow any serious incident: dates and durations that are already public and timestamped, not reconstructed from memory afterward.
Host it separately from the infrastructure it reports on
A status page hosted on the same infrastructure as the product it monitors has an obvious single point of failure: exactly the outage that most needs communicating is the one most likely to also take the status page itself down. Hosting the status page on genuinely separate infrastructure (a different provider, a different region, at minimum a different deployment) means it stays reachable during the incidents that matter most.
Subscriptions turn a status page from pull to push
A status page only helps people who think to check it. Email or SMS subscriptions to status changes — ideally scoped to specific components rather than all-or-nothing — turn the page into something that reaches affected users proactively instead of requiring them to go looking for an answer during an outage they may not even know is happening yet.