My understanding is that Netflix mostly detect VPNs by flagging ips being used by multipe accounts at the same time.<p>I also know that some (expansive) VPNs are still working with Netflix by proving a unique ip for each client, solving that issue.<p>The problem is, with the shortage of available ip v4, they've become quite expensive and no VPN can provide you a unique ip in each country. Most of them provide you a unique US ip and some a unique UK ip and that's pretty much it.<p>With the virtual infinity of ip v6, wouldn't that be cheaper to provide unique ip adresses in all countries and make it nearly impossible for Netflix to flag it?
Who IP address ranges are given to is not any secret (arin, ripe and so on had databases on this) and there is nothing special to block a block of IPs instead of just one.
I don't think that's the only way they flag VPNs. If that was the case, student accommodations, hotels, offices and the like would be flagged pretty quickly.<p>I do believe one of the things they check is if the IP address belongs to a datacenter range (e.g. AWS, Azure, GCE, OVH, etc) or to an ISP providing Internet access. It is just the opposite of what some anti-spam RBLs do :)