Back to articles

How to take down fake domains and phishing sites: dismantle the operation, not just the URL

A fake domain is usually one visible piece of a larger setup. The same operator often controls the registrar account, the hosting, the lookalike pages waiting in reserve, and the ads or messages that send victims there. If you remove only the URL you were sent, the operator can point traffic at a replacement and carry on.

This guide covers the part most takedown checklists skip: mapping the infrastructure behind a fake domain before you file anything, then removing the connected pieces together. For the basic abuse-report and legal routes (UDRP, URS and court), see our complete guide to taking down a fake domain.

Key takeaways

  • One-off takedowns fail because registering a replacement domain takes minutes and costs very little. The operator loses one URL, not the operation.
  • Map before you report. Registrar, host, certificates, analytics IDs, page code and linked accounts often tie several domains to one operator.
  • File against every connected asset in parallel, at the registrar, host or CDN, browser blocklists and the channels sending traffic.
  • Track two numbers: time to takedown and the share of reports confirmed as removed. A sent report is not a removal.

Why one-off takedowns fail

Most teams treat a phishing site as a ticket. Someone reports a URL, an analyst files an abuse complaint, the page goes offline and the ticket closes. The problem is what the ticket never captured.

Phishing operations are built to absorb losses. Kits can be redeployed on a new domain quickly, and operators often register several lookalikes at once so a spare is ready when one is suspended. Outtake's investigations regularly surface this kind of pre-staged infrastructure: domains that are registered and configured but not yet active.

When a takedown removes one domain and leaves the rest, three things usually stay in place:

  • The operator's accounts. The registrar and hosting accounts used to buy and serve the site.
  • The reserve. Other registered lookalikes, often sharing the same name servers, hosting or page template.
  • The traffic sources. The ads, social profiles, SMS campaigns or emails that sent victims to the page. If they still run, they can be repointed.

The goal of a takedown is to make the operation expensive enough that the attacker moves on. That means removing what they would need to start again, not only what you saw first.

How to map the infrastructure behind a fake domain

Mapping turns one URL into a list of connected assets you can act on together. A threat graph is the record of that work: each domain, server, account and ad is a node, and each shared attribute is a link between them.

Start from the domain you found and collect these attributes.

Attribute Where to find it What a match suggests
Registrar and registration date WHOIS or RDAP lookup Domains bought together at the same registrar, often days apart
Name servers DNS lookup Domains managed from the same DNS account
Hosting IP and network owner DNS A records and IP WHOIS Pages served from the same server or provider
TLS certificates Certificate transparency logs Certificates issued for several lookalike names at once
Analytics and tag IDs Page source One operator tracking several sites with the same account
Page template and kit files Page source, file paths, image hashes The same phishing kit reused across domains
Form-post destination Page source or network capture Stolen data sent to the same collection server
Linked accounts and ads Links on the page, ad libraries, social search The traffic sources feeding the site

Treat each match as a lead, not proof. Shared hosting providers serve thousands of unrelated sites, and redacted WHOIS data rarely names a person. A link between two domains is strong when several attributes line up, for example the same template, analytics ID and registration week.

Record what you observed, where and when. Screenshots with the address bar visible, saved page source with a SHA-256 hash, DNS records and the redirect chain are the evidence each provider will ask for.

Tip

Check for domains that resolve but show no content yet. A configured domain with the same name servers and certificate pattern as a live phishing site is a candidate for the operator's reserve.

A step-by-step takedown workflow across registrars and hosts

Once you have the map, work through it as one case. The steps below assume you have confirmed the activity is malicious rather than a parked or disputed domain.

1. Group the assets by provider

List every connected domain, server and account, then group them by who can act: each registrar, each hosting provider or CDN, each ad network or social platform. One report per provider can cover several assets and shows the reviewer the pattern.

2. File with registrars

Send each registrar's abuse contact the domains, the specific phishing URLs, your evidence and the brand being impersonated. Since ICANN's 2024 contract amendments, generic top-level domain registrars must take prompt action to stop or disrupt well-evidenced DNS abuse, which includes phishing. If a registrar does not respond, ICANN accepts DNS abuse complaints against it.

3. File with hosts and CDNs

Report each server to its network owner, citing the provider's acceptable use policy. If a CDN sits in front of the site, report to the CDN as well; many forward reports to the origin host. Our guide to reporting phishing involving Cloudflare covers that case.

4. Protect users while removal is pending

Report every live phishing URL to Google Safe Browsing and other browser blocklists. A browser warning is not a removal, but it reduces harm while providers review your reports.

5. Cut the traffic sources

Report the ads, social profiles and messaging accounts sending traffic to the domains. If they stay up, the operator can repoint them at a new site within minutes.

6. Confirm and watch for return

Recheck each asset after the provider responds. Record the outcome as removed, suspended, blocked or no action. Then monitor for new registrations that share the attributes you mapped, since a returning operator often reuses its template, hosting or tag IDs.

Benchmarks to track

Two numbers tell you whether a takedown program removes threats or only files paperwork.

  • Time to takedown. Measure from detection to confirmed removal, not from detection to report sent. Exposure keeps building until the page is actually down.
  • Confirmation rate. The share of submitted reports that end in a confirmed removal. A report that sits unanswered, or a page that comes back under the same domain, should not count.

For reference, Outtake's Brand Protection reports a 99.4% confirmation rate across submitted reports. Track your own baseline first, then compare it against vendor figures using the same definitions.

A third number is worth tracking once you work at the campaign level: how many connected assets each case removes. A program that removes one domain per case is likely leaving the reserve in place.

Fake domain takedown checklist

  1. Confirm the domain is impersonating your brand and causing harm now.
  2. Capture timestamped screenshots, page source with a hash, DNS records and the redirect chain.
  3. Look up the registrar, name servers, hosting IP and certificates.
  4. Extract analytics IDs, kit files and the form-post destination from the page source.
  5. Search for other domains that share those attributes, including ones not yet live.
  6. Find the ads, profiles and messages sending traffic to each domain.
  7. Group every asset by the provider that can act on it.
  8. File with each registrar, host and CDN, with evidence for every asset.
  9. Report live URLs to browser blocklists while removal is pending.
  10. Report the traffic sources to their platforms.
  11. Confirm each outcome and record time to removal.
  12. Monitor for new registrations that reuse the mapped attributes.

FAQ

How long does it take to take down a phishing site?

It depends on the provider, the evidence and whether the activity is clearly malicious. Hosts and registrars that act on clear phishing reports can remove a site within hours. Weak evidence, unresponsive providers and legal disputes take longer.

Why does a phishing site come back after it is taken down?

Usually because only one domain was removed. The operator still has its accounts, its reserve of lookalike domains and its traffic sources, so it can redeploy the same kit. Mapping and removing connected assets together reduces the chance of a quick return.

What is a threat graph?

A threat graph is a map of the assets behind an attack and the attributes that connect them, such as shared hosting, registration patterns, analytics IDs and linked social accounts. It turns a single fake domain into a list of everything the operator would need to keep going.

Should I report to the registrar or the host first?

Report to both at the same time. The host can remove the content and the registrar can suspend the domain. Running the reports in parallel avoids waiting on one provider before contacting the other.

What counts as a confirmed takedown?

A confirmed takedown is one where you have rechecked the asset after the provider's response and verified that the content is gone or the domain no longer resolves. A sent report, an automated acknowledgment or a browser warning alone does not count.

How Outtake dismantles fake domain operations

Outtake starts from a single signal, such as one flagged domain, and its agents trace shared hosting, registration patterns, analytics IDs, linked social accounts and operator activity. The average campaign dismantles 40 or more connected nodes across five or more platforms. Remediation agents file with registrars, hosts, social platforms, app stores and ad networks in parallel, and every outcome feeds the threat graph used for the next case.

See how it works on the Brand Protection page or book a demo.

Sources