Our Website Passes Malware Scans. Why Are Visitors Still Being Redirected?
Your hosting company says the website is clean. The malware scanner is not finding anything. Then a customer tells you that opening a page sent them somewhere else.
You check the website yourself, and it looks fine.
That is a frustrating place to be, especially when you have already spent time getting the problem investigated.
I have worked on websites where the explanation was an infected copy still sitting in the server cache. I have also found a malicious external script after the scans being used failed to identify the problem.
If visitors are still being redirected, I would keep investigating what their browsers are actually receiving. A clean scan is useful information, but the redirect still needs an explanation.
A cleanup left malicious content in the server cache
In one recent case, a client’s WordPress website had been compromised through a plugin.
The hosting company identified the plugin as part of the Essential Plugin supply chain attack disclosed in April 2026. The incident affected more than 30 plugins from a portfolio that had previously been legitimate.
WordPress.com’s account of the incident explains how malicious code was introduced following a change in ownership. Wordfence’s record for WP Slick Slider and Image Carousel, the plugin identified by the host, lists the injected backdoor.
The hosting company performed the malware removal.
People were still seeing redirects and malicious content afterward.
When I investigated, I found that the server cache had not been cleared after the cleanup. It was still serving content from before the removal.
Clearing that cache resolved the remaining redirect problem.
From the visitor’s perspective, the website was still behaving like a hacked website. The cleanup needed to account for the saved content being delivered to them.
How can that happen?
A page cache stores a generated copy of a page so the website can serve it quickly on later visits.
If that copy was created while malicious content was present, cleaning the underlying website may leave the stored copy in place until it expires or is purged.
The browser can keep receiving the old content during that time.
Depending on the setup, there may be separate copies held by a WordPress caching plugin, the hosting platform, a content delivery network, and the visitor’s browser. Clearing one layer may leave another untouched.
Hosting platforms document these as separate caching layers with different purge controls.
This is why I would ask the host to confirm which caches were cleared after malware removal. “We cleared the cache” can leave a lot unanswered.
I covered the broader version of this problem in Why Aren’t My Website Changes Showing Up for Customers?. After a security incident, that same delivery problem can keep exposing visitors to malicious content.
Another website had a redirect script the scans missed
In a separate case, a client’s website had a malicious redirect that the hosting support team and the scans being used had not identified.
While reviewing the page source, I found an external script that concerned me.
The source had changed since the previous time I inspected the page. That made it harder to trace what was being loaded and where it was coming from.
When I checked the script’s source separately, it was flagged as malware.
Scanning the main website had not given us that answer.
That gave the investigation something specific to follow: the suspicious script, the address it loaded from, and the code or setting responsible for adding it to the page.
Websites routinely load scripts for analytics, forms, bookings, maps, and other features. An unfamiliar address alone needs investigation before anyone removes it.
In this case, the suspicious script and the separate malware finding provided evidence that the earlier checks had missed.
What does a clean malware scan tell you?
It tells you that the tool did not detect a problem within the checks it performed at that time.
The scope of those checks matters.
A remote website scanner examines what it can access from outside the site. A scanner operating within the hosting environment may have access to files or other information that the remote tool cannot see. The exact coverage depends on the product and configuration.
Sucuri’s SiteCheck documentation specifically notes that remote scanners have limited access and that results are not guaranteed.
That is a reason to understand what was checked and follow up on evidence from visitors.
I would ask:
- Did the scan check the page where the redirect happened?
- Did it examine the external script that page loaded?
- Did the cleanup include the relevant files, database content, and settings?
- Was the visitor receiving a cached page?
- Can the reported redirect still be reproduced?
The answers help explain how a clean report and a real customer problem can exist at the same time.
Start with the page and conditions the customer reported
“The website redirected me” is a useful starting point. A few more details make it much easier to investigate.
Ask for:
- The page they opened
- The address they were sent to, if they already have it
- The approximate time it happened
- Their device and browser
- Whether they arrived through search, an advertisement, or a direct link
- A screenshot or recording, if they already captured one
Use the information they have. There is no need to ask a customer to keep opening a suspicious destination.
I would then test the reported path in a controlled environment.
Checking the homepage alone can miss a problem on another page. Checking while logged in to WordPress can also produce a different result from an ordinary visit. Some hosting configurations bypass the page cache for logged-in users.
A private window can help compare sessions. It still receives whatever the hosting platform or content delivery network serves.
If one person sees the problem and another does not, those differences are useful clues. Save them for the person doing the investigation.
Ask for a specific explanation of what was cleaned
If the host has already completed a cleanup, I would ask for a short written account of what they found and changed.
That should help answer:
- What was compromised?
- What malicious content or code was removed?
- What happened to the plugin, account, or other access point involved?
- Which caches were purged afterward?
- Which affected pages were tested?
- What should happen if the redirect returns?
For a compromised plugin, there needs to be a clear decision about its removal or replacement, along with a review of anything the attacker may have added elsewhere.
WordPress’s hacked-site guidance covers documenting the symptoms, using different types of scans, and reviewing access as part of the response.
A business owner should be able to understand the outcome without interpreting a security log.
I want to know what caused the problem, what was changed, and how the result was checked.
Follow suspicious scripts back to their source
If a redirect remains unexplained, a developer or security specialist may need to inspect the page source and the requests made as the page loads.
The investigation should establish which script or response triggers the redirect and how it gets onto the website.
A suspicious script might be referenced by a theme, a plugin, stored page content, or a setting used to insert code. The investigation should establish which of those, if any, is responsible.
Save the relevant evidence before making changes.
If the script’s address or contents change between inspections, preserve those observations too. A screenshot of one version may only describe what appeared on that visit.
Once the responsible source is identified, the fix needs to address that source and any related compromise. Removing one visible script tag can leave the mechanism that adds it unresolved.
This is work I would hand to someone who can trace the behavior without disrupting legitimate website features.
Verify the cleanup through the affected visitor experience
After the cleanup and relevant cache purges, go back to the pages and conditions that produced the original complaint.
I would want the follow-up to confirm that:
- The affected pages load without the reported redirect.
- The identified malicious script or content is no longer being delivered.
- The result holds across repeated visits and the relevant devices or sessions.
- Important features still work after the changes.
Agree on a follow-up check as well. If the redirect returns after the cache has been rebuilt, that is new evidence to investigate.
Cache clearing resolved the remaining redirects in the first case because stale content was the problem left after cleanup. It belongs alongside the work to remove the compromise and address how the attacker gained access.
The second case required closer inspection of what the page was loading before the scans’ blind spot became apparent.
Both investigations needed an explanation for what visitors were experiencing.
Still getting reports after the website was declared clean?
Send the support team the affected URL, the behavior people reported, and the details of the work already completed.
Those details help the next person investigate the behavior that remains unresolved.
If you need help tracing an unresolved WordPress problem, our WordPress support and troubleshooting service is a place to start.
For a more involved issue spanning the website, hosting, and other services, a Technical Website Assessment can help establish what is happening and what needs attention.
Written by Joey Zuccarini
COO at Kismet Creative Co.