Best ISP proxies for sneaker botting
05 Sept 2026 · NatProxies
ISP proxies win checkout tasks because the session survives and the connection is fast. Residential wins monitoring. Using one for both is why most tasks fail.
If you are botting sneaker or Pokémon drops and your tasks keep failing at checkout, the proxy type is usually the reason. This is the version I wish someone had given me before I wasted money on the wrong kind.
Short answer
Use ISP proxies for the checkout tasks. Use residential for monitoring.
Most people buy one type and use it for everything, which is why they get inconsistent results and cannot work out what changed.
What an ISP proxy actually is
An ISP proxy is a static IP that belongs to a consumer internet provider — AT&T, T-Mobile, Comcast — but is hosted in a datacentre.
That combination is the whole point. The IP's ASN lookup says "residential broadband customer", so it passes the checks that instantly reject datacentre ranges. But it is physically in a datacentre on a fast connection, so it behaves like server infrastructure: low latency, high throughput, and available every time you connect.
A rotating residential proxy is the opposite trade. It is a genuine home connection, borrowed for a moment, on someone's actual broadband. Maximum authenticity, unpredictable speed, and a different IP most requests.
Why ISP wins for checkout
Speed decides drops. A retail drop is decided in the first few seconds. Residential proxies route through a real home connection, and you inherit that person's upload speed and latency. ISP proxies sit on datacentre links. On a drop measured in milliseconds, that difference is the difference.
Sessions survive. Adding to cart, filling checkout and submitting payment is one continuous session. If your IP changes halfway through, you look like an account being passed between machines — which is exactly the pattern fraud systems exist to catch. A static IP holds the session together.
You can warm them. The same IP every time means you can build history on it: browse normally, make a small purchase, look like a returning customer. You cannot warm an IP you will never see again.
Consistent results are debuggable. When every task uses a different random IP, a failure tells you nothing. With static IPs you can see which ones perform and retire the ones that do not.
Where residential is genuinely better
Anyone who tells you ISP is always better is selling you something.
Monitoring and scraping. Watching product pages for stock changes means thousands of requests. Do that from one static IP and you will be rate-limited within minutes. Rotation is exactly what you want.
Very aggressive sites. A few retailers block datacentre ASNs so thoroughly that even ISP ranges get caught. Residential is the fallback.
Geographic spread. If you need to appear in twenty countries, residential covers the map in a way a handful of static IPs never will.
The working pattern for most people: residential for the monitor, ISP for the tasks.
How many proxies do you need
The honest answer is fewer than most guides suggest, because most guides are written by people selling proxies by the hundred.
- 1 IP per task is the safe assumption. Two tasks sharing an IP doubles the request rate from that address and roughly doubles the chance of a flag.
- 10 IPs is a reasonable start for a serious attempt at a single drop.
- 25 to 50 if you are running several sites or several sizes.
Buying 200 proxies for your first drop is how people lose money before learning anything. Start small, see which sites work, scale what does.
What actually matters when choosing a provider
ASN, not marketing. "Premium residential" means nothing. Ask which carrier the IPs sit on. AT&T and T-Mobile ranges are the ones that hold up on US retail.
Fresh versus reused. A "fresh" ISP subnet has not been hammered by hundreds of previous botters. A cheap reused one may already be blocked at the sites you care about, and you will not find out until the drop.
Billing model. ISP is billed per IP with unlimited bandwidth — heavy use costs the same as light use. Residential is billed per gigabyte, so a monitor left running overnight can be genuinely expensive. Match the model to the workload.
Delivery speed. If provisioning takes hours or waits on a human to approve your order, that is a provider you cannot use when a drop is announced with a day's notice.
Check the fraud score yourself. Run a sample through Scamalytics or IPQualityScore before buying at volume. Any provider unwilling to give you a test IP is telling you something.
Common mistakes
Using residential for checkout. The most expensive mistake, and the most common. Rotating IPs on a logged-in session is the exact signature fraud systems look for.
Sharing IPs across accounts. Ten accounts behind one IP links them all. When one is flagged, they all are.
Buying by price alone. A $1 proxy that is already blocked at your target site costs more than a $3 one that works, because you also lost the drop.
Not testing before the drop. Load the site through each proxy the day before. Finding out at release that half your IPs are blocked is entirely avoidable.
What we sell, plainly
We sell AT&T Fresh and T-Mobile Legacy ISP proxies — static US carrier IPs, unlimited bandwidth, billed per IP for a 30-day term, from $2.50 per IP with a 10 IP minimum. Provisioning is automatic and takes under a minute, and you can switch between IP whitelist, user:pass and SOCKS5 authentication whenever you like.
We also sell rotating residential by the gigabyte for the monitoring side.
We are not the cheapest and we are not going to pretend our proxies work on every site on earth. If you are scraping at volume, buy residential — from us or anyone. If you are running checkout tasks and losing to session flags, ISP is the fix.
Try it on your own workload
Start from a single gigabyte. No subscription, no minimum commitment.
See pricing