Hacked website: what to do in the first hour, and who to call
What you do in the first hour after a website is hacked shapes how much you can recover and explain later.
This guide puts that hour in order: stop the harm to visitors, keep the evidence, make the reporting decisions with the right advice, and hand the cleanup to a specialist. Clever Ops does not clean up hacked sites, so the guide also tells you who does.
In this guide
First-hour checklist for a hacked website
Print this, or keep it open on a phone that was not used to manage the site, and tick each step as you go. Each step is tagged by type.
How to use itHide the instructions
Keep and Contain steps are for you and your host. Report steps are decisions to take with advice. Specialist steps belong to the cleanup provider. Verify steps come after the site is clean. The times are a guide: if a step stalls, write down why and move on. These are common first steps to weigh with your host or responder, and a specialist's instructions come first once one is involved.
My website has been hacked: start here
The first hour is where reasonable actions taken in the wrong order do lasting harm: deleting strange files before anyone has copied them, restoring last month's backup over the evidence, or changing passwords on the laptop that may be how the attacker got in.
The steps on this page are ordered so that each one keeps the next one possible.
Clever Ops does not provide incident response, forensic investigation, malware cleanup or website monitoring, and is not the right first call for a site that is compromised now. Call your hosting provider first, then an incident response provider or a website malware cleanup specialist; the section on choosing a specialist explains which kind suits which situation.
What Clever Ops does provide comes later. Once the site has been cleaned, a security audit or test looks for the weakness the attacker may have used, if it is still open, and for whatever else the site exposes.
If you read only one part of this page, make it the first-hour checklist. It is written to be printed or kept open on a phone, and to be worked through by an owner or office manager with no technical background.
- Do not delete files, user accounts or database records yet. Record them first, so whoever investigates can see what the attacker changed.
- Do not restore a backup over the live site yet. A backup helps only if it was made before the attack, and nobody knows when that was until someone looks.
- Do not cancel the hosting plan or ask for the account to be wiped. Ask your host what would happen to the files and logs before anything like that is considered.
- Do not sign in to the site's admin area from a computer you have doubts about, and do not open suspicious files from the site on an everyday computer.
- Do not run security scanners or malware cleanup plugins against the live site until a specialist has a copy. Google's recovery guidance warns that vulnerability scanners can be far more invasive than antivirus scanners, with greater potential to cause unwanted damage to a site, and says to back the site up before running one.
How to tell a hacked website from a broken one
Not every strange page is an attack.
An expired domain, a failed plugin update or a hosting outage can also take a site down or make it look wrong, and each has a quicker fix. Signs of a compromise are different in kind: content you did not write, behaviour that changes depending on who is looking, and accounts or files that nobody on your side created.
Google's warnings are one of the clearer signals. When a Google evaluation finds that a site was hacked, or that it behaves in a way that could harm a visitor or their computer, the Security Issues report in Search Console shows the findings. Affected pages can carry a warning label in search results, or show a warning page in the browser when someone tries to visit.
Hacked content can also hide from the owner. Google describes cloaking, where a page shows different content to different visitors: you might see an empty page at an address where Google sees spam words and links. Google also notes that its browser warnings depend on the browsing context, so you may not be able to reproduce a warning a customer saw, and it says to rely on the Security Issues report as the source of truth.
- A warning in Google search results, or a browser warning page, when you or a customer visit the site.
- A message in Search Console, or an email from your host saying the site has been flagged or suspended.
- Pages or links you did not create. A Google search for site: followed by your domain shows the site's pages, including pages a hacker might have added.
- Visitors being sent to another site, sometimes only when they arrive from a search result or on a phone.
- Administrator accounts, scheduled tasks or recently changed files that nobody on your side created.
- Customers reporting card fraud after buying from you, or a checkout page that looks or behaves differently.
The first hour, step by step
The timeline runs in four blocks of roughly fifteen minutes.
The times are there to keep things moving: if a step stalls, write down why and move to the next one. These are common first steps to weigh with your host or responder, not a fix in themselves, and if a specialist is already involved, their instructions come first.
The steps map to incident response outcomes in the NIST Cybersecurity Framework 2.0, which include containing incidents, recording the actions performed during an investigation, and collecting incident data and metadata with their integrity and provenance preserved.
Start a written log before you touch anything: the time, what you saw, the page address, who noticed and how. Screenshot the warning or the damage, ideally from a phone or computer you do not use to manage the site, because a hacked page can serve harmful scripts to whoever opens it.
Every later step goes into the same log with the time and the name of the person who did it.
Then call your hosting provider on the support number in your hosting account, not one shown on the hacked site, and tell them the site has been compromised. Google's recovery guidance also says to tell your host. Ask whether the host can stop the site serving visitors without deleting anything, and ask that the files and logs be kept as they are.
Taking a site offline should mean it stops serving pages, not that anything is removed. Google's guidance describes stopping the web server, or pointing the domain's DNS entries to a static page on a different server that returns a 503 status. It warns that an error code from the compromised site itself is not enough, because harmful content can still be sent with it, and that a robots.txt rule blocks only search engine crawlers, not people.
On rankings, it says temporary downtime during recovery is unlikely to affect where the site appears in search results later.
Get a full copy of the site files, the database, the web server access and error logs and the hosting control panel's login history, dated and stored away from the live server. Mark the copy as infected and keep it off everyday computers.
Google's clean-up guidance gives the same instruction for a site whose backup is clean but outdated: make a disk image of the current site even though it is still infected, and mark the copy as infected.
In the same block, write down every user account on the site, the hosting panel, FTP or SFTP and the database, marking any that nobody recognises. Google's guidance says to note unwanted account names for later investigation. Then list your backups with their dates and where each one is kept, and restore none of them yet.
Ask your host, or check your backup tool, whether older backups are deleted automatically, and ask for that to be paused so a backup from before the attack is not lost while this is sorted out.
Change passwords in an order that cuts off the ways back in: first the email account that receives password reset links, then the hosting control panel and the domain registrar, then the site's administrator accounts, FTP or SFTP and the database.
Google's guidance names FTP, database, system administrator and content management system logins among the passwords to change. Turn on two-factor sign-in wherever it is offered, and expect your host or developer to update the site's settings after the database password changes.
Do this from a computer you have reason to trust. Google lists a virus-infected administrator's computer among the potential ways in, since the hacker might have installed spyware that records the administrator's keystrokes, and recommends running several reputable antivirus scanners on every computer an administrator has used to sign in to the site.
The last block is for decisions rather than technical work: whether personal information may be involved, whether to report to police, what your payment provider and insurer need to hear, and which specialist does the cleanup. Google's clean-up guidance makes a similar point: where confidential user information was obtained from a site, it says you may want to consider any business, regulatory or legal responsibilities before you begin cleaning the site or deleting any files.
Offline now, or a controlled copy first?
Two good instincts pull against each other here.
Taking the site down protects visitors; leaving everything untouched protects the evidence. A holding page served from outside the compromised files lets you do both, because the site stops serving while its files and logs stay as they were. When that cannot be arranged straight away, work through these questions in order and take the first answer that applies.
1. Is the site harming visitors right now?
Harm means visitors being redirected to scams, offered downloads, shown fake login or payment forms, or put through a checkout that may be copying card details. If yes, stop the site serving first and copy second. The files will still be there to copy once it stops serving, as long as nothing is deleted.
If no, and what you see is a defaced page, spam pages or a warning, go to question 2.
2. Can your host or developer take a full copy within minutes?
If yes, copy first, then stop serving: files, database and logs, labelled with the date and marked infected. If no, or nobody is sure how, stop serving now through a holding page outside the compromised files and take the copy as soon as someone can. Do not let the copy become the reason the site stays up for hours.
3. Does the attacker still seem to be active?
Changes that keep appearing, passwords that stop working or new accounts showing up mean someone still has a way in. Stop serving now and ask your host to take a snapshot of the account as it stands, because an attacker who is locked out may try to delete files or logs.
Then change the email and hosting passwords from a trusted device, and call the incident response specialist now rather than at the end of the hour. If you hold cyber insurance and the policy asks you to call the insurer before engaging a responder, make that call first.
4. Does the site take payments or hold customer accounts?
If it does, contact your payment provider through its own support channel, tell it what you saw and when, and ask whether online payments should be paused. Treat personal information as possibly involved until a specialist says otherwise, and read the reporting section below before anything is cleaned.
5. Still unsure?
Treat the site as harmful. Stop serving through a holding page, copy everything, change the passwords and let the specialist decide what is safe to bring back. A short outage is easier to explain to customers than harm the site kept serving after you noticed.
What evidence to keep, and why it matters later
Evidence lets a specialist answer the two questions that decide everything after the cleanup: how the attacker got in, and when.
The answer to when tells you which backups are safe to restore; Google makes the same link, noting that understanding when a hack first took place helps determine which backups might still be clean. The answer to how tells you what has to change before the site goes back up.
Logs need the earliest attention. Ask your host today how long it keeps web server access logs and control panel login records, and ask for a copy now rather than later. Google also warns that an attacker might have altered log files, which is one more reason to copy them before anything else changes and to leave the originals with the host.
Keep suspicious files where they cannot run: inside the infected copy, not on an office computer, and unopened. The NIST framework lists, under incident analysis, that incident data and metadata are collected and their integrity and provenance preserved. In plain terms, you can show where each file came from and that nobody has changed it since.
- Your written log, with times, actions and the person who took each one.
- Screenshots of warnings and of defaced or spam pages, plus any emails from Google, your host or customers.
- A full copy of the site files and database, dated, marked infected and stored away from the live server.
- Web server access and error logs, the hosting control panel's login history and any FTP or SFTP logs your host keeps.
- The site's own activity log if it has one, and the user lists for the site, hosting panel and database with unknown accounts marked.
- A list of backups with their dates and locations, left untouched.
- Payment provider alerts, customer reports of card fraud, and any message from someone claiming responsibility or asking for money.
The reporting decisions: privacy, police and payments
These are decisions to make with advice, not steps to tick off alone, and the right answer depends on your business and on what the site held.
What follows is what each body says, so you know who to ask.
Personal information and the Notifiable Data Breaches scheme
If the site held personal information, such as enquiry form submissions, customer accounts, bookings or orders, a privacy question sits beside the technical one. The Office of the Australian Information Commissioner (OAIC) explains how the Notifiable Data Breaches scheme applies to entities covered by the Privacy Act 1988. Under the scheme, an organisation or agency covered by the Privacy Act 1988 must notify affected individuals and the OAIC when a data breach is likely to result in serious harm to a person whose personal information is involved. Part IIIC of the Privacy Act 1988 covers notification of eligible data breaches.
Whether your business is covered by the Act, and whether this incident meets that test, are legal questions about your circumstances, and this page cannot answer them. Ask a privacy lawyer, and ask early, while the facts are still being gathered: what was on the site, what the attacker could reach, and when. The OAIC's own page on the scheme explains how it applies.
Police and ReportCyber
The Australian Federal Police asks victims of cybercrime to report it at ReportCyber, which the AFP describes as part of the Australian Cyber Security Centre, with resources to help people report and stay aware of cyber threats. Your log is the material a report draws on: what happened, when you noticed, and what you have done since.
The Australian Cyber Security Centre's report page separates the choices. One option reports a cybercrime, including scams, to police. Another reports a cyber security incident to the centre itself, which it describes as an event that has compromised, or could disrupt, an organisation's systems, data, communications or operations through unauthorised access, modification or interference. Its contact page lists the Australian Cyber Security Hotline as 1300 CYBER1 (1300 292 371).
The same report page has an option for reporting a ransomware or cyber extortion payment to the Department of Home Affairs, and says a reporting business entity has an obligation to report such a payment under section 27 of the Cyber Security Act 2024. Whether your business is a reporting business entity is a question for your lawyer, and one to ask before any payment is discussed.
If someone is threatening you or demanding money to stop an attack or keep data private, treat it as a police matter. The AFP lists 000 for emergencies, asking for police, and the Police Assistance Line on 131 444 for non-urgent help, which connects you to your local police.
Payments, insurance and the people you deal with
If the site takes card payments, your payment provider needs to hear from you through its own support channel, with what you saw and when. If you hold cyber insurance, read what the policy says about notifying the insurer and choosing a responder before you engage anyone, because the insurer may have its own process.
Check customer and supplier contracts for clauses about telling the other party of a security incident, and take advice before anything goes out to customers, so that what you say matches what the specialist finds.
Hacked website repair: choosing who cleans it up
Two kinds of provider do this work, and the facts in your log decide which one you need.
An incident response provider, often described as digital forensics and incident response, investigates how the attacker got in, what they reached and whether they are still present, and keeps evidence in a form a lawyer, insurer or police officer can use. A website malware cleanup service removes malicious code, restores the site and hardens it, and its focus is getting a clean site back online rather than a full investigation.
Lean towards incident response when personal or payment information may have been reached, when the attacker seems to be still active, when the hack has touched email or other systems beyond the website, or when a lawyer, insurer or police report is likely to need the findings.
A cleanup service can suit a site that holds no personal information, where the damage is spam pages or a defacement and the attacker's access has been cut off. If you are unsure, start with incident response; a cleanup can follow its findings.
Website malware removal and fixing the hack are different jobs. Removing malicious files does not close the gap the attacker used, and Google's recovery guidance warns that several independent hacks may be in place, so it recommends continuing to search even after one weakness is found and fixed.
Its clean-up steps include checking that any backup was made before the site was hacked, correcting the vulnerability, changing every site-related password again, and doing a clean installation rather than just an upgrade, because upgrades can leave files from a previous version.
- Will you take and keep a copy of the site, its database and its logs before you change anything?
- Will you look for how and when the attacker got in, and tell me plainly if you cannot work it out?
- Will you reinstall the platform, plugins and themes from clean sources, or only remove the files you find?
- Will you check user accounts, scheduled tasks, the database and other sites on the same hosting account, and not only the files?
- What will your written report record: what you found, what you removed or changed, the likely way in, and what you could not check?
- What will you hand back at the end, including the evidence copy?
Hacked website recovery: back online and clearing the warnings
Bring the site back only when the specialist confirms two things: the malicious content is gone, and the weakness it came through is fixed or, if it could not be found, what was done instead.
The NIST framework lists recovery outcomes along the same lines: the integrity of backups is verified before they are used for restoration, and the integrity of restored assets is verified, systems and services are restored, and normal operating status is confirmed.
If Google flagged the site, the warning does not clear on its own. Google says you must request a review to have a page or site unflagged as dangerous or possibly deceptive, and lists four things to have done first: verified ownership of the site in Search Console, cleaned the site, corrected the vulnerability and brought the clean site back online.
Google's help page for the Security Issues report adds that the issue must be fixed on every affected page, because fixing only some pages does not earn a partial return to search results, and that a good request explains the issue, describes the fixes and documents the outcome.
Do not ask too early. Google says that requesting a review while the problem still exists only prolongs the time the site is flagged, and that a request made before the issue is fixed can lengthen the turnaround of the next one. Its help page says most reconsideration reviews can take several days or weeks, and you receive an email when the review is complete.
After the cleanup: checking the way in is closed
This is where Clever Ops fits.
Once a specialist has cleaned the site, a security audit or test of the cleaned site looks for the weakness the attacker may have used, if it is still open, and for what else the site exposes: outdated components with public advisories, administrator accounts or access keys nobody can account for, files and backups that should not be public, and, where customers sign in, access rules that let one person reach another person's data.
The audit is not forensics. It tests the site as it stands after the cleanup; it cannot reconstruct what happened during the attack or prove that nothing was left behind, and its report records what was checked and what was not. Read beside the specialist's report, it answers a narrower question: is the door the attacker probably used still open, and are there others like it?
Which service fits depends on the site. A website security check suits a brochure site with no customer logins. A WordPress security audit covers WordPress sites and WooCommerce stores role by role. A web application and API penetration test suits a custom application with customer accounts, payments or an API.
Live Shopify storefronts and Laravel applications are outside the service. If Clever Ops built or hosts the site, the work is a security review rather than an independent test, and the report says who built and who hosts it.
- The specialist's report, or their notes on what they found, removed and changed, and the likely way in.
- Confirmation that the site was cleaned or rebuilt and that passwords and keys were changed afterwards.
- Written authorisation from the site owner, and your host's conditions for testing if it has any.
- Test accounts for each kind of user, if customers or staff sign in.
- The questions you most need answered, such as whether the same plugin, form or account could be used again.
Making the next incident smaller
Nobody can rule out another attack, but the first hour goes very differently when the groundwork is already done, and most of that groundwork is administrative rather than technical.
- Keep a one-page contact sheet offline: your host's support number and account ID, your domain registrar, your developer, a chosen incident response provider, your insurer and a privacy lawyer.
- Know where your backups are kept, how far back they go and how to restore one, and keep at least one copy outside the hosting account itself.
- Turn on two-factor sign-in for the hosting panel, the domain registrar, the email account that receives reset links and every site administrator.
- Keep a list of everyone with administrator access, and remove people promptly when they leave.
- Ask your host how long it keeps access logs, and whether that is long enough to cover an attack noticed weeks after it began.
- Publish a way for people to report security problems to you, so the next warning reaches you directly. A vulnerability disclosure policy sets out how.
- Print the first-hour checklist on this page and keep it with the contact sheet.
Frequently asked questions
Usually yes, if it is harming visitors, but take it offline in a way that keeps the evidence. A holding page or 503 response served from outside the compromised files, for example through a DNS change to a static page on another server, stops visitors reaching the site without deleting anything. Google's guidance notes that an error code from the compromised site itself is not enough, and that temporary downtime during recovery is unlikely to affect future search rankings.
Not straight away. A backup helps only if it was made before the site was hacked, and nobody knows when that was until someone has looked at the logs. Restoring over the live site also overwrites evidence, and an older copy of the site can still contain the weakness the attacker used. Google's clean-up guidance says to check that the backup predates the hack, then to install updates and correct the vulnerability after restoring.
It depends on what the site held and on your business, and this page cannot tell you. If personal information may have been reached, ask a privacy lawyer early. The OAIC explains that the Notifiable Data Breaches scheme requires an organisation or agency covered by the Privacy Act 1988 to notify affected individuals and the OAIC when a data breach is likely to result in serious harm to a person whose personal information is involved. Separately, the Australian Federal Police asks victims of cybercrime to report it at ReportCyber.
Hacked content can be cloaked, showing different content to different visitors, so you may see an empty or normal page where Google sees spam words and links. Google also notes that its browser warnings depend on the browsing context, so a warning a customer saw may not appear for you. A site: search for your domain lists pages a hacker might have added, and the Security Issues report in Search Console shows a sample of the affected addresses.
It is an important step, but not a fix. An attacker can leave other ways back in, such as extra administrator accounts, modified files or scheduled tasks, which keep working after the passwords change. Change them from a trusted device during the first hour, then again once the cleanup is done; Google's clean-up guidance repeats the password change as part of the cleanup.
No. Clever Ops does not provide incident response, forensic investigation, malware cleanup or monitoring. Start with your host and an incident response provider or website cleanup specialist. Once the site is clean, Clever Ops can audit or test it, looking for the weakness the attacker may have used and anything else the site exposes, and the report records what was and was not checked.
It depends on what the specialist finds: how the attacker got in, how much they changed, whether a clean backup exists and how much of the site needs rebuilding. If Google flagged the site, the review comes after that, and Google's help page says most reconsideration reviews can take several days or weeks. Plan around the specialist's findings rather than a fixed date.
Scope the right test for your application
Tell us what you have built, who uses it and when it needs to launch. You will get a plain-English recommendation: a website audit, a web application test or a referral to a specialist.