What makes a good status page design?
A good status page tells a worried customer, at a glance, whether their part of your product works, what's being done if it doesn't, and when they'll hear more. It's one of the few pages in a web app's UI that people open only when something has gone wrong, so calm, clarity and speed matter more than brand flair.
Status pages have a small set of jobs: a current-status banner, a component list, active incidents with timestamped updates, scheduled maintenance, an incident history, and subscriptions. Below we look at how three well-known public pages handle those jobs, as they appeared in October 2026, then turn the patterns into a checklist you can apply to your own. Status pages change, so check the live pages before copying any detail.
What does GitHub's status page get right?
GitHub's status page gets the component list right: it names the features developers actually use, such as Git Operations, Webhooks, API Requests, Issues, Pull Requests, Actions, Packages, Pages, Codespaces and Copilot. A developer whose CI is failing can find Actions in a glance, without decoding internal service names.
Each component carries a 90-day uptime bar with a legend that separates no downtime, no data, and partial or major outage days. Component states use a fixed vocabulary: Operational, Degraded Performance, Partial Outage, Major Outage and Maintenance. Incident updates are tagged Investigating, Update, Monitoring and Resolved, so a reader can tell where an incident stands from the label alone.
Subscriptions cover email, SMS, Slack, webhooks and Atom or RSS feeds, and past incidents sit on the same page with a link to the full history. The lesson for your page: a fixed, small set of states, used the same way every time, is easier to read under stress than custom wording.
What does Stripe's status page get right?
Stripe's status page gets grouping and scope right. Components sit under product-area groups such as Stripe API, Global payments, Revenue and finance automation, Banking-as-a-service, core components, and acquirers and payment methods. A business that only uses one product area can skip straight to it.
Its incident wording is the part worth studying. Updates describe the specific symptom (higher error rates on a named product), the time window in UTC, and which records were affected, rather than a vague "some users may experience issues". Updates on incidents caused by payment-method partners say plainly that the issue is external to Stripe, which tells the reader whether the fix is in Stripe's hands.
Stripe also publishes partner maintenance windows on the page, with a plain disclaimer that partner-supplied information may not be complete or current. That honesty about what it doesn't control is a pattern any company that depends on third parties can copy. Subscriptions include email, SMS, Slack, feeds and a dedicated social account.
What does Cloudflare's status page get right?
Cloudflare's status page gets scale right. A global network has hundreds of locations, so the page offers separate views by service and by network location, and lists scheduled maintenance per data center with its code, city, type, and start and end times in UTC. A customer can check the one location or product they care about instead of scanning a wall of green.
Maintenance notes say what the customer will feel, such as traffic being rerouted with possible added latency, and give extra instructions to customers with private network connections. Each incident has its own page with a full timeline.
It also treats machines as readers. The page points automated clients to a status API that returns JSON, which suits infrastructure customers who wire status into their own monitoring. Notification settings live in the Cloudflare dashboard. For devtools and infrastructure products, a status API is often as useful as the page itself.
How do these status page examples compare?
The three pages share the same core jobs but each solves a different problem well. Use the table to pick the pattern that matches your product's shape.
Pattern | GitHub | Stripe | Cloudflare |
|---|---|---|---|
Component naming | Features developers use | Grouped by product area | Split by service and by location |
Incident wording | Fixed update labels | Specific symptom, window and scope | Customer impact in plain words |
Third-party issues | Not a major theme | Stated as external to Stripe | Private-connection notes |
Maintenance | Listed when scheduled | Partner windows with a disclaimer | Per data center, UTC times |
History | 90-day bars, full history page | 90-day bars, past incidents | History and per-incident timelines |
Machine access | Webhooks and feeds | Feeds | JSON status API |
Best model for | Products with a short feature list | Multi-product platforms | Global or regional infrastructure |
Our recommendation: most SaaS products should copy GitHub's structure, borrow Stripe's incident wording, and add Cloudflare's location view only if customers really experience outages by region.
How should you word status page incident updates?
Write every incident update in the same four parts: what's affected, what users will notice, what you're doing, and when the next update will come. The fixed shape means a reader who opens the page mid-incident understands it from the latest update alone.
- Affected: name the component as it appears in your list, so the two match.
- Symptom: describe what users see ("API requests are returning errors"), not internal causes.
- Scope: give the start time with a time zone and, when you know it, who or what is affected.
- Next update: commit to a time and keep it, even if the update is "still investigating".
The same rules that make error messages useful apply here: plain language, no blame, and a next step. Avoid "some users may experience" when you know more, and don't mark an incident resolved until you've seen it stay fixed.
Hypothetical example: rewriting a vague incident update
For example, suppose a project management tool posts: "We are aware of an issue affecting some users and are working on a fix."
Rewritten in the four-part shape: "File uploads (Attachments component) are failing for workspaces in the EU region since 14:05 UTC. Existing files can still be opened. We've identified a storage configuration change and are rolling it back. Next update by 14:45 UTC."
The second version tells a reader whether they're affected, what still works, and when to check back, which is usually enough to stop them filing a support ticket.
Status page best practices: a checklist for your own
Use this checklist when you design or review a status page. It covers structure, wording and the parts teams forget until the first real outage.
- Open with one clear banner for overall status, with the time it was last updated.
- List components named after what customers use, and group them once the list gets long.
- Use a fixed set of component states and incident labels, with a visible legend.
- Show active incidents above everything else when one is open, and render status as static text so no loading state hides it.
- Give every update a timestamp with a time zone.
- Keep at least 90 days of visible history, and link to a full incident archive.
- List scheduled maintenance with start and end times and the expected customer impact.
- Offer subscriptions by email plus the channels your users live in: Slack, webhooks, RSS.
- Host the page on separate infrastructure and a separate domain or subdomain from your app.
- Meet the same accessibility bar as the product: status shown in text, not color alone.
- Link to the status page from your app's error states, help center and footer.
Item 10 matters more than it looks: red and green dots alone fail for color-blind readers, so pair every color with a word. Our accessible web app checklist covers the contrast and labelling rules in detail.
Should you use a hosted status page or build your own?
Most startups should use a hosted status page product and spend their design time on component names and incident wording, which is where the value is. Build a custom page only when the status page is part of the product experience, as it often is for developer tools and infrastructure.
Pros
- Hosted tools give you subscriptions, history and uptime bars on day one.
- They run on separate infrastructure, so the page survives your outage.
- Customers already know the layout, so there's nothing to learn.
Cons
- Branding is limited to logos, colors and a few layout options.
- Location or account-specific views usually need a custom build.
- Status data often lives apart from your own monitoring, so updates stay manual.
If you're a devtools company and status is part of how developers judge you, a custom page that matches your product's design system and exposes a status API can be worth it. We design and build these as part of our product design work, alongside the in-app states that point to them.
