“Legit” gets used for two different questions. One is whether a company is real: whether it exists, can be reached, and will still be there next month. The other is whether what it sells is what it says: residential addresses from people who agreed to share their connection, in the countries claimed, that work on the sites you need.
Both questions have checkable answers. This guide lists the evidence worth collecting before you pay, in the order we would collect it, whether you are buying residential proxies for scraping or monitoring for the first time or replacing a provider that let you down.
Start with who you are paying
The simplest checks are the ones most often skipped.
- A legal entity you can name. The terms of service and invoices should name a registered company and a jurisdiction. A brand with no entity behind it leaves you with nobody to hold to anything.
- Terms that say what you are buying. Look for an acceptable use policy, a refund policy and a statement on how traffic is billed. Vague terms are not proof of bad intent, but they leave you nothing to point to when a dispute arrives.
- A support channel that answers a technical question. Ask something specific, such as how a session is held or which countries support city targeting, and judge the answer, not the speed of the greeting.
- A payment flow that behaves normally. Real providers take ordinary payment methods and issue proper invoices. Pressure to pay through unusual channels, or “lifetime” deals that only make sense if the company plans to disappear, deserve caution.
None of these proves quality. They establish that there is someone accountable, which every later check depends on.
Ask where the addresses come from
For residential proxies this is the central question. Residential exit addresses are real people’s home connections. A legitimate pool is built from people who knowingly agreed to share bandwidth, usually through an app or software kit, in exchange for payment or a free service, and who can stop at any time.
A provider should be able to explain, in plain words:
- How participants join and what they are told on the consent screen.
- What they receive in return.
- How they opt out, and whether that removes them promptly.
- What limits protect the participant, such as running only when the device is idle or on Wi-Fi.
- How customers are vetted, and which uses are refused.
Evasive answers here are the strongest negative signal in the whole process. We cover what good and bad sourcing look like in how providers ethically source residential IPs; the short version is that a provider that cannot describe its own supply chain either does not know it or does not want you to.
Test the claims you can measure
Some claims can only be taken on trust. Others can be checked in an afternoon, and those should be.
Is the pool actually residential? Send requests, record the exit addresses, and look up who each address is registered to. Hosting and cloud networks inside a “residential” pool are a known pattern, and datacenter addresses get flagged by the sites you are trying to reach. The checks are set out in real vs fake residential proxies.
Is it where it says? Request exits in a country you care about and compare the location the target sees with the one you asked for.
Does it work on your targets? Run a sample of your real requests and count usable responses, not status codes. A challenge page can arrive with a success code.
Do this during a trial or on the smallest plan, with the same requests you will send in production. A provider that discourages testing before purchase is telling you something.
Check whether two brands are one network
A provider can be entirely legitimate and still not independent. Much of the residential market is resale: a brand buys access from a larger operator and sells it under its own name.
That matters for legitimacy in two ways. First, the reseller’s promises about sourcing are only as good as its upstream’s, which you cannot inspect. Second, if you buy a second provider as a backup, it may draw on the same addresses as your first. Atlas measures this address by address on its proxy pool overlap page, including which direction the overlap runs.
Resale is not fraud. Paying a premium for it without knowing is the problem.
Read the evidence that already exists
You do not have to run every check yourself. The question is which existing evidence is independent of the seller.
- Marketing pages and comparison tables written by the provider describe intent, not results.
- Affiliate “best of” lists are paid by the providers they rank, which does not make them wrong but does make them a weak source on their own.
- Customer reviews are useful on support and billing, and weak on performance, because each reviewer saw one use case.
- Measurements from a party that discloses its interests are the strongest outside evidence, if you can see how they were taken.
Atlas is run by Shifter, which also sells residential proxies and is ranked in it. Our disclosure page sets out how that conflict is handled: the same tests and formula for every network, Shifter included, access bought the way a customer buys it, no paid placement, and an open-source collector you can run on your own account to check our figures. You should apply the same scepticism to us that this guide asks you to apply to anyone else.
Watch how disputes are handled
Every provider has a bad week. What separates a trustworthy one is what happens next.
Look for a public status page with real incident history, not a permanent green light. Look for changelogs that admit when something changed. When you raise a problem, notice whether the reply engages with your evidence or deflects it. The same standard applies to anyone publishing numbers about providers: corrections should be dated and visible, and a network that disputes a figure should get a measured answer either way.
A checklist before you pay
- Name the legal entity and read the terms, refund and acceptable use policies.
- Ask the five sourcing questions above and keep the answers in writing.
- Run a trial with your own requests: confirm residential registration, location and usable responses.
- If you are adding a backup, check how much of its pool it shares with your current provider.
- Read reviews for support and billing, and measurements for performance.
- Start on a plan you can leave, and only commit to volume after a month of real traffic.
What “legit” will not tell you
Passing every check above means a provider is accountable and broadly sells what it claims. It does not mean it is the right one for your targets, or the best value. Those are separate questions, answered by measuring success on your own sites and dividing what you pay by what works. The network profiles in Atlas are a starting point for both, and your own trial is the finish.