Nobody can guarantee a takedown
The registrar, the host or the platform makes the decision, every single time. A vendor promising a 100% removal rate is promising something it does not control. Here is what is actually in our hands.
Anyone selling you a 100 percent removal rate is selling you a sentence, not a service.
The final decision never belongs to the vendor you hired. It belongs to the registrar, the host, the platform, the app store or the marketplace, and it belongs to them every single time. No amount of experience, tooling or relationship changes who signs off.
That is an uncomfortable thing to put on a pricing page, so most providers do not. It is still the truth of how this works, and a buyer who understands it will choose better than one who is shopping for the biggest number.
Why the verdict is never ours
A takedown is not an action you perform on someone else's system. It is a request for enforcement, decided by the party that operates the service.
Three things follow from that.
They enforce their own rules, not yours. A platform acts when its own policy is breached. Your trademark, your losses and your urgency are inputs to that decision, not the standard it applies. This is why an impersonation report can be rejected while everyone involved agrees the account is fake.
You are not their counterparty. Their contract is with the customer who pays them: the registrant, the site owner, the developer, the seller. You are a third party asking them to take something away from the person they have an agreement with. That asymmetry is built into every abuse queue in existence.
They can be overruled, and they can change their mind. Appeals exist. A suspended domain can be reinstated, a removed app can be restored, an account can come back after review. A vendor that promises permanence is promising to control an appeal it is not even a party to.
What a guarantee usually turns out to mean
Guarantees in this market are rarely lies. They are almost always definitions.
| The claim | What is usually underneath it |
|---|---|
| "100% success rate" | Cases they declined to take are not counted |
| "Guaranteed removal" | A credit or refund if it fails, not a removal |
| "Guaranteed in 24 hours" | Guaranteed filing in 24 hours, not verified removal |
| "Removed" | Sometimes means deindexed from search, or blocked in one country, while the page stays live |
| "Unlimited takedowns" | Unlimited submissions, which is a measure of effort rather than outcome |
None of that is useful to you at the moment a fake login page is collecting your customers' credentials. The only number worth reading is verified removals, checked from outside the provider's own ticket view, with the failures still in the denominator.
If you are comparing vendors, what to check before you buy goes through the rest of the fine print.
They own the verdict. We own the case.
The honest split looks like this.
Not ours to promise
- Whether the report is accepted.
- How long the review queue takes.
- Which policy they choose to apply.
- Whether they reinstate on appeal.
Ours, every time
- The quality of the evidence.
- Filing with the right party, first time.
- Escalation when the first route stalls.
- Watching for the re-spin afterwards.
A takedown is a decision made by somebody else, on the strength of what you put in front of them. That second part is entirely in our hands, and it is most of what separates a case that resolves in a day from one that never moves.
The four things that actually move the odds
Evidence. A complete abuse report carrying the live URL, the harvesting endpoint, WHOIS and DNS records, captures with timestamps and the specific policy clause breached gets decided in one review cycle. A thin one gets a question, and a question costs a whole cycle on their business days. What belongs in the pack is in what evidence you need, and the writing that keeps it from bouncing is in how to write a takedown notice that does not bounce.
Routing. The same abuse filed at the wrong layer stalls for a week. Host, registrar, CDN and platform each have a different remedy, a different evidence bar and a different speed, which is the whole argument in who to report to first and how long a takedown takes, layer by layer.
Escalation. Silence past a published window is a routing failure, not a verdict. The case moves up: upstream provider, registrar suspension, the relevant CERT, the store's rights channel, and where the facts support it, a UDRP instead of an abuse report. That judgement is covered in why takedowns stall and picking the remedy.
Persistence. Removal destroys a deployment, not a capability. Kits redeploy, handles get replaced and apps return under a new publisher, so the asset stays on a watch list after it comes down. That is the subject of monitoring for recurrence.
Do those four well and most cases come down. Do them badly and some of the same cases never will. That is influence over an outcome, which is a real thing to sell. It is not a guarantee, which is not.
Three things we do guarantee
You pay for removals, not effort. On a paid plan, quota is consumed only when a takedown is confirmed. If it does not come down, it does not bill. The reasoning is in why we only charge for successful takedowns, and the plans are on pricing.
We do not stop at one vendor. Registrar, hosting provider, CDN or reverse proxy, DNS provider, social platform, ad network, app store, marketplace and search, worked in parallel where it helps and escalated when a route stalls. The full set of channels is on the platforms page.
We do not watch one place. Domains, social, apps, marketplaces, ads, chat and code, and the asset stays watched after it is removed. That is what monitoring is for, and why it sits inside brand protection rather than next to it.
Those three are ours to control, which is exactly why they are the only things we put a promise on.
Pay for what comes down. Nothing else.