GitHub serves as the backbone for millions of developers worldwide, hosting critical code and collaboration workflows. With that central role comes a deep responsibility to maintain reliability—and to communicate clearly when things go wrong. Earlier this year, GitHub acknowledged recent availability challenges and outlined ongoing reliability investments. Now, the platform is delivering on its promise of better incident communication by rolling out three major improvements to its status page.
These changes are guided by three core principles: transparency, accuracy, and timeliness. By providing more nuanced incident classifications, publishing per-service uptime metrics, and offering granular insights into specific dependencies like AI model providers, GitHub aims to give developers a clearer, more honest picture of platform health.
More Accurate Incident Classification
Previously, any service disruption was labeled at least a “Partial Outage,” even if the impact was minimal. This often led to confusion, making it appear that a service was completely down when it was merely sluggish or experiencing intermittent glitches. To fix this, GitHub introduced a new incident severity level: Degraded Performance.
Now the status page uses a three-tier system that spans the full spectrum of possible issues:
- Degraded Performance – The service is operational but impaired. Users may notice elevated latency, reduced functionality, or sporadic errors affecting a small percentage of requests.
- Partial Outage – A significant portion of the service is unavailable or severely impacted for a meaningful number of users.
- Major Outage – The service is broadly unavailable, affecting most or all users.
This refined classification ensures that developers receive a more truthful signal about what’s actually happening. A slight slowdown won’t be mistaken for a full-blown outage, reducing unnecessary alarm while still prompting caution.
Per-Service Uptime at a Glance
Another major addition is the publication of per-service uptime percentages directly on the status page. For each listed service, you can now see its reliability track record over the last 90 days. This gives developers a quick, data-driven way to assess how stable a given GitHub feature has been recently.
How Uptime Is Calculated
The uptime percentage is derived from the number, severity, and duration of incidents affecting that service. Weighted calculations follow industry-standard practices:
- Major Outage – 100% of the incident duration counts as downtime.
- Partial Outage – 30% of the duration counts as downtime (reflecting significant, but not total, service loss).
- Degraded Performance – 0% counts as downtime, since the service remains functional.
For example, a one-hour Partial Outage over a 90-day period would contribute only 18 minutes to the downtime calculation—not the full hour. This methodology ensures that brief, severe incidents are weighted fairly, while sustained minor degradation does not artificially depress uptime numbers.
By making these numbers transparent, GitHub empowers teams to make informed decisions about relying on specific services and to better understand any historical patterns of instability.
Granular Insights into Service Dependencies
Modern software platforms often depend on third-party providers, and GitHub is no exception. To improve clarity around such dependencies, GitHub has started with a dedicated status component for Copilot AI Model Providers. This component appears on the status page whenever a model provider experiences issues, offering a clearer breakdown of the root cause.
Previously, a problem with an underlying AI provider might have been lumped into a generic “Copilot” incident. Now, users can quickly see whether the disruption is caused by the model provider itself or by GitHub’s own infrastructure. This separation helps developers decide whether to wait for the provider to recover or to explore alternative workflows.
GitHub plans to extend this model to other critical dependencies over time, ensuring that incident communication remains as precise as possible.
Why This Matters for Developers
Clear, honest communication during incidents is more than a courtesy—it’s essential for developer productivity and trust. With the new three-tier severity system, teams can prioritize their response: a Degraded Performance alert might simply mean “expect slower responses, but keep working,” while a Major Outage prompts a full switch to offline modes.
Per-service uptime data also supports better decision-making for CI/CD pipelines, automated deployments, and third-party integrations. When you know that a particular service has a 99.9% uptime track record, you can calibrate your own error handling accordingly.
Finally, the dedicated component for Copilot AI model providers sets a new standard for transparency around dependencies. As platforms become increasingly interconnected, understanding where a failure originates is crucial for rapid incident response.
GitHub’s updates demonstrate a commitment to treating developers as informed partners. By providing more accurate, granular, and timely information, the company is not only improving its own status page but also raising the bar for incident communication across the industry.
Learn more about GitHub’s reliability work by exploring the official status page.