Buying Guides 5 min read

Residential Proxy Quality Checklist: 9 Tests Before You Scale Up

Before you buy residential proxies in volume, run these nine tests on a small plan. Most take minutes.

A residential proxy quality checklist is what you run on a small plan before spending real money. Marketing pages all claim large pools and high success rates; the only numbers that matter are the ones you measure on your own targets. These nine tests take an afternoon at most and tell you whether a residential network is worth scaling up.

1. Success Rate on Your Real Target

Send 200 to 500 requests to the actual site you care about, at the concurrency you'll really use. Count responses that contain the data you need, not just status 200, since block pages often return 200 too. Under 90 percent on a moderately protected site deserves a closer look.

2. Geolocation Accuracy

For a sample of IPs in your target country, check what a geolocation service reports. The script below does this along with several other checks.

3. IP Type: Really Residential?

The org field from an IP lookup should name consumer internet providers, not hosting companies. A residential pool full of hosting ranges isn't residential. Read what is an ASN to interpret these fields.

4. Rotation and Uniqueness

Without a session, 100 requests should return close to 100 different IPs. Lots of repeats mean a small pool in that country.

5. Sticky Session Stability

Hold a session for 10 to 30 minutes and poll its IP every 30 seconds. It should stay the same until the time runs out. Residential IPs are real home connections, so an occasional drop is normal; frequent changes are not.

6. Speed and Latency

Measure time to first byte and total time for a small page. Residential is slower than datacenter by nature; what matters is whether it's consistent and fast enough for your concurrency.

A Script for Tests 2 to 6

import time
import statistics
import requests

USER, PASS = 'USERNAME', 'PASSWORD'
HOST = 'global.nodeproxies.xyz:8080'
COUNTRY = 'us'

def lookup(username):
    proxy = 'http://%s:%s@%s' % (username, PASS, HOST)
    start = time.time()
    r = requests.get('https://ipinfo.io/json', proxies={'http': proxy, 'https': proxy}, timeout=30)
    data = r.json()
    return data.get('ip'), data.get('country'), data.get('org'), time.time() - start

rows = []
for _ in range(30):
    try:
        rows.append(lookup('%s-country-%s' % (USER, COUNTRY)))
    except requests.RequestException:
        rows.append((None, None, None, None))

ok = [r for r in rows if r[0]]
print('success: %d/%d' % (len(ok), len(rows)))
print('unique IPs:', len({r[0] for r in ok}))
print('in country:', sum(1 for r in ok if r[1] == COUNTRY.upper()))
print('median seconds:', round(statistics.median(r[3] for r in ok), 2))
for r in ok[:10]:
    print(r[0], r[1], r[2])

sticky = '%s-country-%s-session-qc1-time-10' % (USER, COUNTRY)
print('sticky IPs:', {lookup(sticky)[0] for _ in range(5)})

Free IP lookup services rate-limit, so keep the sample modest or use your own lookup endpoint.

7. Header and DNS Leaks

Request http://httpbin.org/headers through the proxy and make sure your real IP doesn't appear in X-Forwarded-For or similar headers. On SOCKS5, use socks5h:// so DNS goes through the proxy. See check proxy anonymity headers.

8. IP Reputation

Look up a sample of IPs on a public blocklist checker. Residential IPs are shared over time, so some history is normal, but a sample where most IPs are listed predicts CAPTCHAs. The real test is still test 1. What is IP reputation explains how scores are built.

9. Documentation, Errors and Support

  • Are the endpoint, ports and username options documented clearly?
  • Are error codes explained, so you can tell bad credentials from an empty plan?
  • Does support answer a technical question with a technical answer?
  • Is pricing public, with no surprise fees?

Scoring the Results

Weight the tests by your use case. For scraping, success rate and cost per success dominate. For account management, session stability and IP type matter most. For geo-testing, geolocation accuracy is everything. Record results in a table and repeat the same tests on each provider you're considering, using the how to choose a proxy provider guide for the rest of the decision.

Run the Checklist on NodeProxies

You can run all nine tests on a small plan: NodeProxies sells residential from 100 MB at $0.65/GB, with a $3 minimum deposit and no KYC. The pool is 90M+ IPs across 195+ countries, with per-request rotation by default and sticky sessions via -session-{id}-time-{minutes}. Register, read the residential proxies page and compare tiers on the pricing page.

FAQ

How do I check if a residential proxy is really residential?

Look up the IP's organisation. Residential IPs belong to consumer internet providers; hosting companies indicate datacenter IPs.

What success rate should residential proxies have?

It depends heavily on the target. Measure on your own site at your real concurrency and compare providers on the same test.

How much bandwidth does testing a proxy use?

IP lookups are tiny; a few hundred requests to a normal page usually use well under 1 GB. A 100 MB to 1 GB plan is enough.

Why do residential sticky sessions sometimes change IP early?

The IP belongs to a real device that can go offline. Occasional early changes are normal; frequent ones are a quality warning.

NodeProxies

Ready to get started?

90M+ residential, mobile & datacenter proxies. Datacenter from $0.30/GB, residential from $0.42/GB. No contracts.

Create free account