7 status page examples and what to borrow from them
See how GitHub, PayPal, Heroku, Google Workspace, DigitalOcean, Slack, and Reddit show service status, maintenance, incident history, and ways to get help.
A Git push hangs while Actions is still running. The checkout works against PayPal’s sandbox, but payments are failing in production. In both situations, a single green or red banner leaves the reader with another question. Which part of the service is affected?
These seven public status page examples show how teams answer that question. Look at what each page puts up front, what it saves for the incident history, and where it sends a reader who still needs help.
1. GitHub gives each kind of status its own place
GitHub puts a branded illustration above a list of components such as Git Operations, API Requests, and Actions. If you’re waiting for a build, you can go straight to Actions. The component’s 90-day bar gives you some recent context, while the incident history is where you go to find updates for a specific disruption.
Those are three distinct reading tasks. A summary tells someone where to look, a component bar shows status over a defined period, and an incident record explains what the team communicated as the situation changed. Keep the paths to all three short. GitHub Status · Incident history
2. PayPal separates the maintenance window from the expected impact
PayPal’s scheduled maintenance notices identify the products involved and give start and end times. Some also specify an impact window. A maintenance window might reserve time for an entire change while the part that affects transactions is shorter. Giving each period its own line helps a reader decide when to defer a job or expect a disruption.
PayPal also separates production from sandbox. A developer investigating a test checkout can check the sandbox before drawing a conclusion about live payments. Follow the same principle in your notices. Name the affected environment and describe what a customer might see. Update the notice as the expected impact becomes clearer. Production history · Scheduled maintenance · PayPal status entry point
3. Heroku separates running apps from the tools that manage them
The legacy Heroku Status page used Apps, Data, and Tools as its main categories. Heroku now directs readers to Salesforce Trust as the primary status source and keeps the old page as a backup. Trust starts with an instance or region, then shows services such as Apps, CLI, Data, and Dashboard with regional history.
There’s a practical reason to keep the application and its tooling separate. A CLI outage and an outage in the app your customer uses call for different decisions. The regional view adds a step, so make it easy for people to recognize their instance. If the status page moves, point old bookmarks to the new primary source and explain which page remains a backup. Heroku on Salesforce Trust · Heroku’s status guide
4. Google Workspace gives unlisted problems a route to support
The Google Workspace Status Dashboard puts a contact-support prompt above its recent service grid. A reader can look up Gmail or Drive and still find a next step when the symptom they’re seeing is missing from the page.
Support and incident communication handle different parts of that situation. A report gives the team something to investigate, and a public incident communicates the scope it can substantiate. Put the support route where people reach the end of the public information. Google Workspace Status Dashboard
5. DigitalOcean follows the reader’s product and region
A Droplet in Singapore can have trouble while the rest of a cloud provider’s service is healthy. DigitalOcean puts locations under the relevant product. You can open Droplets and look for SGP1, while services such as billing have their own entries. Past incident updates live in a separate history.
Structure your page around what customers already know about their deployment. If they have a product name and a region code at hand, those are useful navigation labels. DigitalOcean Status · Incident history
6. Slack gives readers a next step in familiar language
Slack puts connection troubleshooting and a feedback email near the status summary. Component names such as Login/SSO, Messaging, and Huddles describe things people already do in Slack.
That placement matters when the whole company uses the same service. If the summary looks healthy but Slack won’t connect, the reader can take another step without knowing how to run a network diagnostic. Use names the people opening the page will recognize, and make the next step visible. Slack Status
7. Reddit makes recent uptime legible by component
Reddit shows 90-day status bars for components such as Desktop Web, Native Mobile Apps, and Comment Processing, with a separate route to historical uptime. The bars show reported component status. Someone checking a problem with comments can look there for a relevant public record.
If you publish a similar view, label the component, the time window, and what missing data looks like. An internal queue-depth chart answers a different question and needs its own explanation before it can help a reader. Reddit Status · Uptime history
Three things to check before you build
- Can readers find the URL? A short address such as
status.example.comis useful if customers can also find it from the help center and remember it during an outage. - Can people read and update it during an outage? Trace both paths. Separate hosting helps, but shared DNS, sign-in, or publishing dependencies can still block customers or the on-call team.
- Can readers subscribe to updates? An opt-in channel such as email can save repeated checks. Which updates go out depends on the notification settings.
When you show regional status, line up the scope of your evidence with the label customers see. Guance synthetic tests let you check an endpoint from selected locations, so record which locations, requests, and assertions your status judgment covers. Detection and notification timing depend on probe and monitor settings. If you connect a service on a Guance custom status page to an SLO, map the service explicitly and choose an enabled SLO without grouping dimensions. Guance synthetic tests · Configure a status page
Frequently asked questions
What should a small team copy first?
Start with the few services your customers can recognize, and write maintenance notices with a clear window and scope. Set aside time for someone to post updates and keep the page current.
Should we publish internal metrics like queue depth?
Start with the decision a customer needs to make. Public component uptime may be enough to check whether a feature had a known disruption. If an internal metric would add something, explain its units, scope, and how to interpret it.
How much history should we keep visible?
There’s no common window across all pages and record types. GitHub and Reddit show recent 90-day component bars on their home pages and keep incident histories separately. Tell readers what period you show, where to find incident updates, and how you mark missing data.
Further reading
Contact us
Join the community
to join the community
Try Guance
Start online and pay only for what you use.
Get startedChoose a Guance plan
