How to Fix a Broken Website Step by Step

Start by defining what "broken" means for your business: checkout down, forms failing, blank pages, or critical admin access lost. Write the top three user journeys that must work today, not the full backlog. That single list keeps the first day from turning into random plugin toggles and database experiments. If revenue or leads depend on specific URLs, treat those as the incident boundary. Everything else is secondary until the boundary is stable again.

Second, gather evidence before you change production. Capture error messages, server logs, recent deploys, plugin or theme updates, and hosting status. If you skip this, you will fix symptoms and reintroduce the real cause within days. In parallel, take a quick backup snapshot if the host still allows it, so you can roll back a bad change without negotiating with panic. Evidence first, patches second.

Third, fix the critical path in order: hosting and DNS, PHP or runtime errors, database connectivity, then application-level issues (theme, plugins, build). Each layer has different owners and tools. Jumping to "rewrite the theme" when PHP memory is exhausted wastes time and budget. For static or headless stacks, verify build output, environment variables, and edge cache before rewriting components.

Fourth, after the site loads reliably, run a short regression pass: main conversion flow, search, key landing pages, and analytics or tag firing. Many "fixed" sites still leak leads because events, thank-you pages, or email hooks broke quietly. Use a checklist, not memory. Document what you changed and why so the next deploy does not undo the fix.

Finally, turn the incident into prevention: staging workflow, update policy, monitoring alerts, and a single owner for releases. A step-by-step fix is only complete when the same failure mode is harder to repeat. If you need outside help, bring someone who can explain the root cause in plain language, not only ship a patch and disappear.