Assess quality of proxy

[Brianvault]

Real-world testing is still the best check, but for initial filtering I usually look at ASN type, IP reputation, DNS consistency and whether the IP matches the claimed geo/provider. Tools can help, but they’re never perfect. A clean-looking IP can still perform badly depending on the target. For static residential, provider quality matters a lot since you keep the same IP long term.

 

[Mango Proxy]

Honestly a lot of ‘proxy quality’ issues are actually signal consistency issues - ASN, DNS, timezone, WebRTC, browser stack all need to make sense together

 

[flatdiskprod]

Curious if anyone has found anything around JA3 and JA4 detection as a signal etc. Almost no proxy detect site or fingerprinting site is really surfacing that. it's newer more sophisticated. I did come across one site that was showing it but I'll have to find it again.

 

[Papa Mexicana]

Normally, people will use more than one technique instead of focusing on a single metric because a proxy label might not be an accurate way to determine whether someone is a bot or not. The first thing that comes to mind for most people would be using Iphey as a quick identifier along with an ASN sanity check, followed by cross-reference from another tool such as MaxMind or Ipinfo. However, the true litmus test will ultimately be the behavior on the target platform.

 

[Mango Proxy]

Normally, people will use more than one technique instead of focusing on a single metric because a proxy label might not be an accurate way to determine whether someone is a bot or not. The first thing that comes to mind for most people would be using Iphey as a quick identifier along with an ASN sanity check, followed by cross-reference from another tool such as MaxMind or Ipinfo. However, the true litmus test will ultimately be the behavior on the target platform.

ASN, DNS, timezone, WebRTC, browser stack all need to make sense together.

 

[Feijo]

Frauddefense I tried the first time. it's much better than the others.