Free vs paid proxies: risks and real costs
Free proxies from public lists are run by operators you can't identify, so anything you send through them can be read or changed unless HTTPS with certificate checks protects it. A paid service costs money, but you can check its prices, terms and limits before you rely on it.
- Category
- Fundamentals
- Published
- Updated
- Reading time
- 6 min
- Checked against
- 2 sources
Free vs paid proxies is less a question of price than of who runs the proxy. A free proxy from a public list is a server someone has left open, on purpose or by accident, and you usually can't tell which. A paid service, sometimes called a premium proxy, has terms you can read, credentials tied to your account and someone you can contact when something goes wrong.
This post covers what any proxy operator can see, why free lists are unreliable, when a free proxy is acceptable, and what to check before you pay for a service.
What a free proxy list is
Free proxy lists are aggregator pages that publish IP:port pairs for open proxies, often found by scanning the internet or submitted by visitors. Nobody vouches for the entries. An entry might be a misconfigured server, a machine that was compromised, a test box someone forgot about, or a proxy set up deliberately to watch the traffic that passes through it. The list can't tell you which.
What a proxy operator can see
Every proxy sits between you and the website, so the real question is how much of your traffic it can read:
| Traffic | What the operator can see | What the operator can change |
|---|---|---|
An http:// URL | The whole request and response: URL, headers, cookies, form fields and page content | Anything, including scripts or ads injected into pages and files swapped in downloads |
An https:// URL through a CONNECT tunnel | The hostname and port you connect to, when, and how much data passes | Nothing it can read, as long as your client verifies the website's certificate |
For HTTPS, your client sends a CONNECT request with only the host and port, and the proxy relays encrypted data in both directions from then on. TLS runs between your client and the website, so the operator can't read the pages. An operator can still try to intercept TLS by presenting its own certificate. A client that verifies certificates refuses that connection, which is why you should never turn verification off to make a proxy "work".
This isn't hypothetical. Researchers who measured more than 107,000 proxies listed on aggregator sites over 50 days reported open proxies that modified HTML to mine cryptocurrency in visitors' browsers, proxies that attempted TLS man-in-the-middle interception, and downloads altered to include remote access trojans and other malware (Mani et al., ACSAC 2018; see Sources).
Reliability and availability
Free proxies come and go. In the same study, more than 92% of the open proxies listed on aggregator sites didn't respond to proxy requests at all. The ones that work are shared with everyone who reads the list, have no stated capacity, and can disappear between one request and the next.
For anything that runs on a schedule, that means constant list maintenance and results you can't compare from one run to the next, because you don't know where each request exited or who saw it on the way.
Where free proxy IPs come from
Usually, nobody knows. A free list doesn't say who operates each address, or whether the owner of the machine knows it's being used as a proxy. Some entries may be misconfigured servers or compromised devices, which would mean your traffic passes through someone else's machine without their consent. You can't check either way, and that alone rules free proxies out for work you might have to explain later.
When a free proxy is acceptable
Rarely, and only for throwaway work:
- Learning how proxy settings work in a client, with requests to a public test endpoint such as
https://api.ipify.org?format=json. - Checking that your own code copes with a proxy that times out or refuses connections.
Never send credentials, cookies, tokens, personal data or anything you wouldn't publish through a free proxy. Never use one for plain http:// pages you care about, and never turn off TLS certificate verification to get a connection through.
What to check before paying for a proxy service
A paid service should pass these checks before you send real traffic through it:
- Published prices and billing model. Per GB or per IP, the smallest order, and whether purchased traffic expires.
- Limits in writing. Targeting granularity, session lifetimes, concurrency, and what the service doesn't do.
- Terms and acceptable use. Which uses are allowed and what happens when someone breaks the rules. Ours are in section 2 of the Terms of Service, and a separate page explains how we expect netdash to be used.
- Documentation with working examples for the clients you actually use.
- Your own credentials. Authentication tied to your account, so your traffic and usage are separate from everyone else's.
- Reachable support. A contact address you can test before you buy.
- Answers about IP sourcing. How the provider obtains its exit IPs, and whether the people whose connections carry your traffic have agreed to it. The questions to ask are in what is a residential proxy?
For rotating services specifically, the rotating proxy checklist goes through each point in more depth.
Free vs paid proxies: the costs you can calculate
A free proxy costs nothing to buy and a lot to depend on: time spent finding entries that work, data you can't trust, and the risk of exposing something you sent. Those costs are hard to put a number on. A paid service's costs are easy to work out in advance:
- Per-GB billing: you pay for the traffic you transfer. Estimate it as pages per run × average page size × runs.
- Per-IP billing: you pay for each address per period, whatever the traffic.
- Smallest order: how much you have to buy to try the service on your own workload.
netdash residential and mobile plans are one-time GB purchases with graduated pricing per order, starting from a 1 GB order. Purchased GB never expire, and 1 GB of traffic is 1 GB from your balance. Static ISP and datacenter IPs are billed per IP every 30 days; self-serve checkout for those is paused. See residential proxy pricing for current rates and a calculator, or compare proxy types to see how each netdash network is billed.
FAQ
Not for anything that matters. HTTPS with certificate verification protects the content of your requests, but the operator still sees which sites you visit and when, and anything sent over plain HTTP can be read or changed. You also can't tell whose machine your traffic passes through.
Over plain http://, yes: the operator sees the whole request, including form fields and cookies. Over https:// with certificate verification on, the password travels encrypted inside the tunnel. Either way, don't sign in to anything through a proxy you can't identify.
No. Paying gets you terms you can read and support you can contact, but the same rules apply: use HTTPS, keep certificate verification on, and send only what the task needs. Run the checklist above before you rely on any service, including ours.
Related
- What is a residential proxy?
How residential IPs route your requests, and what to ask a provider about sourcing.
- Choosing a rotating proxy provider
A 2026 checklist for pricing, sessions, targeting, documentation and terms.
- Residential proxies
netdash residential plans: country targeting, rotating or sticky sessions, one credential pair.
Sources
Third-party behaviour described on this page was checked against these documents.
- 01RFC 9110 HTTP Semantics, section 9.3.6: CONNECTAccessed 2026-10-09
- 02Mani, Vaidya, Dworken, Sherr: An Extensive Evaluation of the Internet's Open Proxies (ACSAC 2018; arXiv:1806.10258)Accessed 2026-10-09
Spotted something out of date? Tell us.