If your WordPress site is showing redirects, unfamiliar pages, spam links, strange administrator accounts, browser warnings, unexpected downloads or unexplained changes, treat it as a possible security incident—not merely a broken plugin.
The safest first response is to reduce harm while preserving information about what happened. Avoid making several unrecorded changes at once. Do not paste “quick fix” recovery code into a theme or plugin, and do not restore a backup simply because it is available. A backup made after the compromise, or one that contains the same malicious files, can put the infection back.
This checklist is general information, not a diagnosis. The right response depends on the evidence, hosting arrangement, site purpose, affected accounts and applicable legal or contractual duties. WordPress describes security as risk reduction rather than a promise of perfect security; its Hardening WordPress guidance also recommends trusted software sources, current software, reliable recovery methods, logging and monitoring.
1. Contain the site before trying to clean it
If visitors may be served malware, phishing content or deceptive redirects, limit exposure. Contact the hosting provider and ask what emergency controls are available. Depending on the host and the nature of the incident, that may mean taking the site offline temporarily, restricting access, placing it behind a maintenance response, or using another controlled containment measure. Google’s hacked-site guidance says taking a hacked site offline can prevent malware from being served and stop further abuse; it also discusses a 503 response where taking the site offline is not possible.
Do not assume that a visible homepage is the whole incident. Attackers may target less obvious URLs, uploaded files, scheduled tasks, database content, administrator accounts, hosting panels, FTP/SFTP, DNS, email or other sites on the same account. If the site processes payments, forms, customer details or staff credentials, pause the affected workflow until its safety is assessed.
If you must inspect a suspicious URL, use a device and network that are not used to administer the site where practical. Do not click unknown links, download files or enter credentials into a page that may be impersonating a trusted service.
2. Preserve evidence before deleting or overwriting anything
Record what was observed and when: browser warnings, redirect destinations, affected URLs, new users, altered pages, unfamiliar plugins or themes, file timestamps, host alerts and relevant screenshots. Save the warning text and the exact URL, but do not repeatedly browse a dangerous page from a normal work computer.
Ask the host to preserve relevant access, error, firewall and malware-scan logs according to its retention policy. Keep a copy of the current site state and database through a trusted, documented process if your incident responder advises it. Label copies with their date and source. The purpose is investigation and comparison—not to put a potentially infected copy back online.
Record who has access to the WordPress administrator, hosting, domain, DNS, email, analytics, Search Console, payment and third-party accounts. This list helps identify what must be secured and what may have been exposed. It also prevents the team from losing access while trying to recover the site.
3. Work out the likely scope
Look for evidence across more than the WordPress dashboard:
- unexpected administrator or editor accounts, password-reset emails or changed recovery addresses;
- modified core, plugin, theme, upload or configuration files;
- unfamiliar scheduled tasks, server users, SSH keys, FTP/SFTP accounts or hosting-panel sessions;
- database content containing injected links, scripts, redirects or new options;
- changes in DNS, email forwarding, payment integrations or external services;
- Google Search Console security warnings, suspicious indexed URLs or unexplained search snippets; and
- other sites, subdomains or applications sharing the same hosting account.
A scanner result is useful evidence, not by itself proof that every malicious persistence mechanism has been removed. Likewise, a site that looks normal to an administrator is not necessarily clean. A qualified responder should correlate timestamps, files, database records, logs, user accounts and the initial access path before declaring the incident closed.
4. Protect credentials—but do it in the right order
If credentials may have been exposed, rotate them from a known-clean device. Prioritise WordPress administrator accounts, hosting and control-panel access, domain registrar and DNS, SFTP/FTP/SSH, database users, email, payment services, analytics, Search Console and any API or application secrets. Remove unauthorised users and review legitimate users’ roles and recent activity.
Use unique, strong passwords and multi-factor authentication where the service supports it. A password change alone does not prove that an attacker has been removed: active sessions, application passwords, API keys, SSH keys, email forwarding rules and persistence mechanisms may also require review or revocation. The official WordPress password-reset documentation includes multiple reset paths and warns that emergency scripts must be removed when finished. Do not leave a temporary reset file, debugging file or recovery tool on a live server.
Do not use a public “emergency” snippet or modify functions.php, a plugin or the database casually to regain access. Such actions can destroy evidence, create a second vulnerability, expose credentials or cause an outage. If administrative access is lost, use the host’s documented recovery process or a qualified responder who can preserve and record the change.
5. Update and recover from a trusted state—not an infected restore
Once containment and evidence collection are planned, identify the recovery route. In some incidents, a clean rebuild from trusted WordPress core, known-good themes and plugins is safer than trying to remove every malicious change. In others, a carefully investigated file and database recovery may be appropriate. The choice depends on the evidence and the site’s complexity.
Never assume the newest backup is clean. Establish when the compromise likely began, inspect candidate backups for suspicious files and content, and confirm that the backup includes the database and required uploads. Do not restore an infected or unverified backup merely to get the site working again. Keep the original evidence and a separate copy of any candidate recovery material.
Use WordPress and plugin/theme packages from trusted sources. WordPress’s hardening documentation says to keep WordPress current and not obtain official releases from untrusted sites. Apply updates in a controlled recovery process, preferably on an isolated staging or recovery environment first where that is feasible. Check compatibility and critical journeys—login, forms, checkout, booking, email delivery and integrations—before exposing the recovered site to visitors.
Remove software that is unused, unsupported, nulled or obtained from an untrusted source. Reinstall core, plugins and themes from verified packages when the evidence warrants it rather than trusting files that were present during the compromise. Keep a change record: what was replaced, from which source, by whom and when.
6. Recheck the recovered site from more than one angle
After recovery, test the site as an administrator and as an unauthenticated visitor. Check desktop and mobile views, redirects, forms, downloads, checkout or booking flows, email notifications, robots directives, canonical URLs and important landing pages. Search for unexpected files, users, links, scripts and database entries. Run appropriate host, file-integrity and security scans, and review logs after the site returns to service.
Do not describe the site as “clean forever.” A recheck is evidence at a point in time. Continue monitoring for reinfection, failed logins, unexpected file changes, new users, outbound email abuse and renewed redirects. Make sure the recovery owner knows how to report a recurrence and where the logs and incident record are kept.
7. Harden after the immediate incident
Hardening should address the likely access path as well as general exposure. Depending on the site and host, review:
- current WordPress, PHP, plugin and theme versions and their support status;
- administrator count, least-privilege roles and dormant accounts;
- unique credentials, multi-factor authentication and session/application-token revocation;
- hosting, database, SFTP/SSH, DNS and control-panel access;
- file permissions and unnecessary write access;
- firewall, rate limiting, login protection and XML-RPC requirements;
- backups stored separately from the live site, with restoration tests and retention rules;
- security logs, malware alerts and an escalation contact; and
- the workstation and browser used to administer WordPress, which should also be checked for malware or keylogging risk.
These controls reduce risk; they do not make a site invulnerable. WordPress’s official hardening guide specifically covers passwords, file permissions, database security, backups, logging, monitoring and trusted sources. Choose controls that can be maintained and documented rather than adding unreviewed plugins or hiding the login URL as a substitute for access control.
8. Consider users, privacy and reporting duties
A hacked site may affect more than its files. Consider whether visitors could have been shown deceptive content, whether credentials may have been captured, whether contact or customer information was accessed, and whether payment or email systems were involved. Preserve evidence and involve the appropriate hosting provider, insurer, legal adviser, privacy officer, regulator or law-enforcement contact where the facts and jurisdiction require it. Do not announce a breach, blame a supplier or promise that no data was accessed before the evidence supports that statement.
If your organisation has contractual notification obligations, document decisions and dates. If a third-party processor or payment service is involved, follow its incident process. This article cannot determine whether a reportable personal-information breach occurred; that requires a fact-specific assessment under the applicable law and agreements.
9. Review Google Search Console if it applies
Verify that the correct site property is controlled and check for unfamiliar owners or users. Review the Security Issues report, indexed pages, performance changes and URL Inspection results. Google explains that hacked sites can host phishing or other deceptive content and that Search Console can show security issues and sample URLs; see its social-engineering guidance.
Only request a security review after all issues identified in the Security Issues report have been fixed throughout the site and the fixes have been tested. A review is not a guarantee of immediate removal of a warning or restoration of rankings. Google says most reviews can take several days or weeks. If hacked URLs were removed through a temporary removal process but should remain indexed, handle that carefully after the underlying issue is fixed; do not use removals as a substitute for cleanup.
When to get specialist help
Get help promptly when you cannot establish the entry point, the site keeps reinfecting, hosting access is affected, multiple sites share an account, customer or payment data may be involved, logs are unavailable, or the recovery would require database or server changes you cannot verify. A specialist should explain the evidence they need, the containment plan, what recovery state they can support and what will be rechecked. Ask for a written record of actions rather than a vague claim that the site is “guaranteed clean.”
Engage24’s live Website Security service describes security audits, malware scanning and removal, firewall configuration, SSL management, two-factor authentication, WordPress hardening, monitoring and incident response. Suitability, scope and any additional work should be confirmed for the specific site; this article does not promise a result, a ranking outcome, a particular response time or a fixed recovery timeline.
Key takeaways
- Contain possible harm and preserve evidence before deleting or overwriting files.
- Rotate exposed credentials from a known-clean device and review sessions, keys, users and related systems.
- Do not restore an infected or unverified backup, and do not use insecure recovery code on a live site.
- Recover from trusted software and a verified clean state, then recheck files, database, users, logs and visitor journeys.
- Harden and monitor the site, assess user and privacy impact, and use Search Console’s Security Issues report where applicable.
- Treat professional support as an evidence-led process—not a guaranteed ranking, cleanup or timeline promise.
Need help fixing a hacked WordPress site?
If your site is showing malware warnings, unexpected redirects or signs of unauthorised access, book time with Engage24 to discuss what happened and the safest next steps. Explain what has already been changed and confirm the scope of support before work begins. Our Website Security service also offers malware scanning and removal and incident response.
Book time with Engage24 for help
Sources and verification notes
- Engage24 — Website Security — live service page verified 17 Nov 2026; includes the Security Maintenance article archive, malware scanning/removal, audits, hardening, monitoring, incident response and quote/booking routes.
- Engage24 — IT Support — live service page verified 17 Nov 2026; includes the IT Support scope and confirmed booking URL used in the CTA.
- WordPress Developer Resources — Hardening WordPress — official guidance on risk reduction, updates, trusted sources, passwords, backups, logging and monitoring; last modified 14 Oct 2026.
- WordPress Documentation — Reset your password — official password-reset methods and cautions around emergency reset scripts; last modified 27 Mar 2026.
- Google Search Central — My site’s been hacked — now what? — official containment, damage assessment, clean recovery, patching, password changes and review guidance.
- Google Search Central — Social engineering (phishing and deceptive sites) — official Security Issues report, sample URLs, cleanup and security-review guidance; last updated 10 Dec 2025.

