What an ISP proxy actually is
An ISP proxy — also called a static residential proxy — is an address registered to a consumer internet provider but hosted on datacenter infrastructure. To the target it looks residential, because the ASN behind it belongs to an ISP rather than a hosting company. To you it behaves like a server: fast, stable, and assigned to you rather than rotating away.
That combination is the entire point. Residential trust normally costs you speed and permanence, because the address belongs to a real household device that will go offline. ISP capacity removes both compromises, at the cost of a smaller and less diverse pool than a rotating residential network.
When you need a persistent identity
Reach for ISP proxies whenever the target ties something to the address rather than merely inspecting it.
Account management is the clearest case: a platform that sees a login from a new country every few minutes will challenge it, however clean each individual IP is. The same applies to dashboards and internal tools that whitelist by address, to long-running sessions that must survive for hours, and to any workflow where you will need to return tomorrow and look like the same visitor you were today.
If your work is stateless — collecting independent pages at volume — a persistent address is the wrong shape and actively worse than rotation, because you concentrate all your traffic on one IP. Rotating residential gives you far more address diversity for that job. Choosing a proxy type covers the trade-off.
Static versus rotating, in practical terms
Rotation spreads requests so no single address carries enough traffic to trip a rate limit. Persistence does the opposite deliberately: it keeps one address so the target can recognise you.
Those goals conflict, and picking the wrong one produces failures people routinely misdiagnose as an unreliable proxy. A stateful flow on a rotating pool drops its session mid-checkout; a bulk collection job on a single static address gets rate limited within the hour. Neither is a fault in the network. How rotation works and when to use a sticky session set out which shape fits which job.
How the billing differs
ISP proxies are billed per address for a fixed period rather than per gigabyte, because you are holding specific IPs rather than drawing on a shared pool. Traffic through them is not metered separately.
That changes the arithmetic against rotating residential. High, predictable volume through a small number of addresses is usually cheaper here — you are not paying for every byte. Low volume that needs wide address diversity is usually cheaper on a rotating plan, because holding enough static IPs to match that diversity would cost far more. Estimate your monthly transfer and your required address count before choosing; how pricing works explains both models.
The failures we see most
A challenge after weeks of clean running. Usually accumulated behaviour on a persistent address rather than a sudden problem with the IP. Reduce request rate on that identity and check that the rest of your fingerprint is still consistent.
Rate limiting under load. Expected on a static address — that is the trade you accepted for persistence. Either slow down or move that portion of the work to rotating capacity.
403 on an address that worked yesterday. Frequently the client rather than the network: TLS and HTTP/2 fingerprints are scored alongside the IP, and a library upgrade can change yours overnight. The five detection layers explains what a proxy does and does not cover.
Our five-step test separates the two in a couple of minutes, and support is staffed 24/7 if it turns out to be ours.
Acceptable use
A proxy changes the network path, not the rules. A robots.txt disallow is a refusal whether or not anyone enforces it, and using our capacity to circumvent an explicit refusal is outside our terms, as are the targets listed in our acceptable use rules. What we log, and for how long, is set out in our logging policy.
