Is Vercel down right now?

While you wait: the Vercel incident checklist

1. Separate serving from shipping. Existing production deployments serve from the edge independently of the build/deploy pipeline. 'Builds failing' and 'sites down' are very different incidents — check which components the status page lists.
2. Freeze auto-deploys. Hold merges to production branches while the deploy pipeline is degraded. A build that dies halfway through the incident is noise, and you want a clean state when it recovers.
3. Know your rollback options. Rollbacks and promotions go through Vercel's control plane. If the dashboard and API are down, you may not be able to roll back either — factor that into whether you ship anything at all.
4. Check functions separately from static. Function incidents 5xx your API routes and SSR pages while static assets keep serving — a half-down site that confuses users and monitors alike.
5. Mind the meta-trap. If your status page or incident tooling is itself hosted on Vercel, it's down too. Communicate from somewhere independent of the platform that's failing.

What breaks when Vercel has issues

If your frontend lives on Vercel, an edge incident is a production outage: pages 5xx or time out for real users. Function incidents take down API routes, SSR and ISR revalidation while static content keeps flowing — partial failures that look confusing in monitoring. Build-plane incidents don't touch running traffic but block every ship, including the hotfix for whatever else is broken, which is exactly when you'll notice. Preview deployments stall too, slowing review of in-flight fixes.