How to Put a WordPress Site in Maintenance Mode
Updated
· 10 min read
Key takeaways
- Use a plugin for quick updates, a .maintenance file when wp-admin is down, and your host's dashboard toggle for migrations.
- A correct maintenance setup returns a 503 status with a Retry-After header so Google treats the downtime as temporary.
- Many maintenance plugins render a branded page but return a 200 status underneath, which can cost you rankings.
- The most common cause of a site stuck in maintenance mode is a forgotten .maintenance file left in the site root.
- At 10+ client sites, bulk update tools handle maintenance windows while monitoring and after-hours triage catch breakage afterwards.
To put a WordPress site in maintenance mode, use a plugin for quick updates, drop a .maintenance file in the site root for a no-plugin fix, or enable it from your host's dashboard for migrations. The right method depends on the job, and serving a proper 503 keeps Google from dropping the site mid-update. This guide covers all of them, plus how to roll maintenance mode out across dozens of client sites at once.
What is WordPress maintenance mode and when should you use it?
Maintenance mode replaces your site with a short notice page and tells search engines the downtime is temporary. Done correctly, WordPress serves a 503 Service Unavailable status so Google knows to come back later rather than drop the page from results, per Google Search Central's guidance on site downtime. Done incorrectly, you serve a 404 or 200 and risk rankings.
Use it in four scenarios: major plugin or core updates on a live site, migrations, redesigns, and emergency repairs. Skip it for a quick security fix or a routine update round that takes seconds.
When a quick update does not need maintenance mode
A single plugin update or a staging deployment finishes before anyone notices. Flipping maintenance mode on for a 30-second change creates more client emails than it prevents. Reserve it for work long enough to show visitors a broken page.
When downtime is planned versus accidental
Planned downtime gets a scheduled window and a notice page. Accidental downtime, like a plugin conflict at 2am, gets monitoring and a fix first, and our process covers how agencies hear about it before clients do.
Which maintenance mode method fits your situation?
Four scenarios cover almost every case. Match yours to a method:
| Scenario | Best method | Why |
|---|---|---|
| Plugin or core update round | Built-in mode or plugin | Short window, quick to toggle off |
| Redesign or rebuild | Plugin with custom page | Clients see a branded page for days, not minutes |
| Migration to new host or stack | Hosting dashboard toggle | Works even when WordPress itself is unreachable |
| Emergency (hack, broken update) | .maintenance file | No dashboard access needed, fastest to deploy |
The plugin route wins on convenience but requires a working admin. The .maintenance file works when wp-admin is down. Hosting-level maintenance covers migrations where you need the whole domain to return a 503.
If you run care plans across many client sites, the real question is which method you can trigger in bulk. That is the problem AccessDossier is built around, and it is covered in the final section.
How do you enable maintenance mode with a plugin?
Install a maintenance plugin, activate it, and the site shows a holding page to visitors while you work. Most plugins let you set the HTTP status to 503 so search engines know the outage is temporary, and let you edit the page's text and logo without touching code. For one-off client work this is the fastest route.
Whitelisting your own IP and logged-in users
Before you activate anything, check the plugin's exclusion settings. You want logged-in administrators to bypass the holding page so you and your team can test the live site while clients see the notice. Whitelisting your office IP does the same for tools that hit the site without a login, like uptime monitors. If the plugin only offers one or the other, prefer the logged-in bypass and test it in a second browser where you are signed out.
Scheduling maintenance windows for client sites
Some plugins let you schedule a window in advance, so the page goes live at 2am during your update round and switches off automatically. If yours does not, a cron job or the built-in .maintenance flag covered in the WordPress developer documentation handles the same job. Whatever method you use, tell the client the window beforehand and confirm the site is back before their morning. For repeatable checklists across many client sites, see our pre-maintenance review.
How do you add maintenance mode without a plugin using the.maintenance file?
Drop a file named .maintenance into the site root (the same directory as wp-config.php) containing <?php $upgrading = time();. WordPress checks for that file on every load and serves the default maintenance screen. Delete the file to bring the site back. Because it needs no plugin install or activation, this is the fastest method for a routine update round: update plugins, clear the file, done.
Creating a custom maintenance.php for a branded page
The default screen is a blank white page with one line of text. To brand it, create wp-content/maintenance.php. WordPress loads that file instead of its default output whenever maintenance mode is active, so you can style it with your agency's logo, a phone number, and an estimated completion time. Include <?php http_response_code(503); ?> at the top so search engines see the correct status. Test the page on a staging copy first, then copy it across client sites so every care plan site shows the same branded screen during an update round.
Common mistakes that leave the site stuck in maintenance mode
Three errors account for most stuck sites. First, a leading dot matters: maintenance.txt does nothing, while a stray .maintenance file left from last week's update round keeps the site down silently. Second, wrong file permissions after an SFTP upload can prevent WordPress from reading or deleting the file. Third, a PHP syntax error inside .maintenance or maintenance.php triggers a fatal error that looks like a hacked site.
Always delete the file immediately after the update round finishes, and check the site's front end before you close the ticket. If a client reports downtime hours later, a forgotten .maintenance file is the first thing to check, before you blame the hosting or a plugin conflict. For sites on a care plan with monitoring, the uptime alert usually catches this within minutes; unmonitored sites can sit broken overnight.
How do you turn on maintenance mode from your hosting dashboard?
Most managed WordPress hosts include a maintenance or "coming soon" toggle in their dashboard. It flips the whole site offline at the server level, before WordPress loads, so it works even if a plugin conflict locks you out of wp-admin. That makes it the right tool when you are migrating a site, restoring a backup, or changing DNS and nothing at the WordPress layer can be trusted.
Maintenance mode versus a staging copy: which to use
A staging copy is usually the better answer for update rounds and risky changes: clone the site, test, then push. Host-level maintenance mode is for the short window when the live site itself must be unreachable, such as a migration cutover or a database restore. For routine plugin updates on client sites, you rarely need to take the live site down at all, which is why we built AccessDossier Care around update workflows on live sites with off-site backups rather than downtime. If you are comparing maintenance tooling across a portfolio, see our breakdown of AccessDossier vs ManageWP (https://accessdossier.com/vs/managewp).
How do you customize the maintenance mode page visitors see?
A bare "briefly unavailable" screen tells clients nothing. Use your maintenance page to do three things: match the site's branding, state when the site will be back, and give a way to reach you if the downtime is urgent for them.
Most maintenance plugins handle all three. With LightStart (Maintenance & Coming Soon page), you can edit the page content, upload a logo, and add your own HTML and CSS, so the screen looks like the client's site rather than a plugin default.
Keep the copy short: what happened, the expected return time, and a contact link or phone number. For agency work, put the agency's contact on the page, not the client's, so after-hours questions come to you first.
How does maintenance mode affect SEO and the 503 status code?
The status code decides how Google treats your downtime. A maintenance page that returns 200 tells Googlebot the site is fine, just empty, and it can start dropping pages from the index. A correct setup returns 503 with a Retry-After header, which tells search engines the outage is temporary and to come back later.
Checking what status code your maintenance page actually returns
Test it before you rely on it. Run curl -I https://yoursite.com from the terminal, or use a browser extension that shows response headers, and confirm you see HTTP/1.1 503 Service Unavailable plus a Retry-After value in seconds. Many maintenance plugins render a pretty page but return 200 underneath, which is the failure mode that hurts.
What Google Search Central says about site downtime
Google's guidance for planned outages is to serve 503 for anything longer than a few hours and keep the header's Retry-After realistic, then drop it as soon as the site is back. For short deployments, a brief 503 window is a non-event; Googlebot retries and reindexes without lasting effect. The real risk is a maintenance page left on for days returning 200, or a 503 with no Retry-After during a long outage.
How do you take a WordPress site out of maintenance mode safely?
Removal differs by method. Delete the .maintenance file from the site root, or deactivate the plugin if you used one. If your host toggled it, switch the site back to live from the same dashboard; most hosts document this alongside their staging tools, as in WP Engine's support documentation.
Post-maintenance checklist: cache, uptime check, functional test
Clear every cache layer: page cache, object cache, and any CDN. Then load the site in an incognito window to confirm the 200 response replaced the 503.
Run a functional test on the pages you touched: forms, checkout, logins. Check uptime from an external monitor, not your own browser, so cached results don't fool you. Across 10 to 200 client sites, run this checklist per site and log the result in your care plan notes, so the next update round starts from a verified state.
How do you handle maintenance mode across dozens of client sites at once?
Running update rounds with maintenance windows
At 10+ client sites, per-site plugin toggles stop scaling. Bulk tools let you schedule an update round across the whole portfolio: WP Umbrella, ManageWP, MainWP, WP Remote, and FleetWP all manage updates, backups, and maintenance mode from one dashboard. The pattern that works: stage updates, run the round off-peak, flip maintenance mode on for the affected sites, then verify and close the window. Keep the window short and documented so a client who notices sees a plan, not an outage.
Monitoring after the window closes
Maintenance mode only covers the work you scheduled. Breakage usually shows up hours later, which is why agencies on Reddit ask for a 24/7 helpdesk rather than another update tool. Uptime monitoring with alerts to you first, not the client, closes that gap. That monitoring-plus-after-hours-triage is where we built AccessDossier Care; the bulk update tools above do not staff a 2am phone. Run both: tools for the round, humans for what follows.
FAQ
How long can a WordPress site stay in maintenance mode before it hurts rankings?
A correct 503 setup with a Retry-After header tolerates hours, even a couple of days, without lasting SEO damage. Problems start when the page returns 200 instead, or when a 503 sits for days with no Retry-After. Finish the work, remove the mode, and verify the site responds 200 before closing the window.
Why is my WordPress site stuck in maintenance mode after an update?
Usually a forgotten .maintenance file left in the site root from the update round. WordPress checks for it on every load and keeps serving the maintenance screen until it is deleted. Check for the file first, including the leading dot, before blaming hosting or a plugin conflict. On unmonitored sites this can sit broken overnight.
Can I put my WordPress site in maintenance mode without logging in?
Yes. Drop a .maintenance file into the site root over SFTP; no wp-admin access needed. That makes it the fastest option when a hack or broken update has locked you out of the dashboard. Alternatively, most managed hosts offer a maintenance toggle at the server level, which works even when WordPress itself is unreachable.
Do visitors see a 503 error or a normal page during WordPress maintenance mode?
They see a normal-looking notice page, but the server underneath should return a 503 Service Unavailable status. The page can be branded with your logo and contact details. The catch: many plugins render a pretty page but return 200 underneath, which tells Google the site is fine but empty. Test the actual headers before relying on any tool.
Is maintenance mode the same as staging?
No. Staging is a private clone of your site where you test changes safely, then push them live; the live site stays up. Maintenance mode takes the live site itself offline for visitors. Staging suits update rounds and risky changes; maintenance mode suits the short window when the live site must be unreachable, like a migration cutover or database restore.