Within hours, not days. A cloned casino or exchange site is a working phishing tool from the moment it goes live, and every hour it stays up is another batch of deposits, logins, or wallet seed phrases captured under your brand's name. The question isn't whether to take it down, it's whether your process is fast enough for the damage window that's actually in play.
The speed problem is getting worse, not better. AI-driven impersonation scams, face-swapped support agents, cloned sites generated from templates in minutes, grew more than 1,400% in 2025 according to Chainalysis.
Fraud linked to AI-assisted operations now generates several times the revenue of manual scams for a fraction of the effort. Total crypto scam losses hit roughly $17 billion in 2025, and the average amount stolen per incident climbed from $782 to $2,764 over the same period.
None of that is because detection got worse. It's because clones are faster and cheaper to stand up than ever, which means the only lever a brand controls once one exists is how quickly it comes down.
A clone that's live for six hours and a clone that's live for six days aren't the same incident. One is a near-miss. The other has a growing list of real customers who typed in a password or sent a deposit to the wrong place, and when it's eventually discovered, they blame the real brand, not the copy.
Four things, in practice: how quickly the clone is detected in the first place, whether the evidence package handed to the host or registrar is strong enough to act on without back-and-forth, whether the brand has legal standing behind the request, and how responsive the specific host or platform is.
The first three are within a brand's control. The fourth varies: some registrars act within hours, others need repeated escalation.
Legal standing matters more than most operators expect. A takedown request backed by a registered trademark is something a host or platform is obligated to act on. The same request without one is a favour they can choose to grant, deprioritize, or ignore, and that difference alone can add days to a case that should have closed in hours.
As a working benchmark: a confirmed case should be actioned within 24 hours of verification, measured from the moment a human analyst confirms the threat is real, not from whenever someone happens to notice it. That's the standard CloneGuard holds itself to on monitoring and takedowns, and it's a reasonable bar for any brand to measure its own process against, in-house or outsourced. Anything measured in business days rather than hours means the clone has had time to do real damage before it's gone.
Hitting a 24-hour response time isn't a tooling problem, it's a staffing problem. It means someone is watching, or something is watching reliably enough to wake someone, at 2am on a Sunday, not just during business hours.
It means having a human review every flag before acting, because a false positive filed against a genuine partner or affiliate costs real relationship damage. It means knowing which registrars and hosts respond to what kind of evidence, built up case by case, not something a generic monitoring tool comes pre-loaded with.
Generic brand-monitoring platforms, the kind built for retail and consumer goods watching for counterfeit products or logo misuse, simply aren't tuned to the patterns iGaming and crypto clones actually use: domain permutations around a casino or exchange name, fake deposit and login pages built to harvest credentials in real time, impersonating app store listings, and fake support accounts running on Telegram and Discord rather than the channels a generic tool watches.
Building that coverage properly in-house means hiring and rostering a team for a problem that's adjacent to, not core to, running a casino or exchange. Most operators who try end up with partial, business-hours coverage, which is exactly the gap a clone launched overnight is built to exploit.
This is the problem CloneGuard exists to take off a brand's plate. Monitoring runs continuously across every channel clones actually use, web, apps, social, and messaging platforms like Telegram and Discord, not just the ones a generic tool checks.
Every flagged case is reviewed by a human analyst before any action is taken, so nothing gets actioned on a false positive. Confirmed cases are actioned within 24 hours, backed by registered trademarks where a brand has them, which is what turns a takedown request from a favour into a legal obligation.
And when a host doesn't respond, cases escalate straight into formal legal enforcement rather than sitting in a queue. It's delivered as one service with one point of contact, not three separate vendors a brand has to coordinate between.
Monitoring is the base of every Protection Plan, with takedowns metered in and takedown packs available on their own if a brand isn't ready for a full subscription.
Don't engage with it directly, and don't spend hours building a bigger case before reporting. Screenshot the site with a visible timestamp and URL, note the domain's registrar and host if you can find them quickly.
Report it immediately through whatever enforcement channel you have, whether that's an internal process or a provider. If the host doesn't respond within a reasonable window, the next step is formal legal escalation, not waiting longer in the hope it resolves itself.
Speed is the only variable that's fully in a brand's control once a clone exists, and reliably hitting a 24-hour response time takes dedicated, round-the-clock coverage most operators aren't staffed to run themselves.
The gap between a six-hour takedown and a six-day one is measured in captured deposits and damaged trust, not inconvenience. If your current process can't tell you, with confidence, how a clone found at 3am tonight would get handled, see what continuous coverage actually costs before you find out the hard way.
Typically within 24 hours of confirmation once the host or registrar cooperates. Cases where the host is slow or unresponsive take longer and usually need to escalate into formal legal action to force resolution.
Yes. A registered mark turns a takedown request into a notice a host or platform is legally obligated to act on, rather than a favour they can choose to ignore. Brands without a registered mark routinely wait longer for the same clone to come down.
Waiting. Teams often try to gather more evidence or get internal sign-off before reporting, which only gives the clone more time to capture deposits or credentials. Report as soon as a case looks credible and let a verified process take it from there.
* Stats: Chainalysis, 2026 Crypto Crime Report.