← Glossary / Residential Proxies

What is Residential Proxies?

Residential proxies are exit nodes whose IP addresses are assigned by real Internet Service Providers to real home broadband subscribers — not to data centres. When a request leaves through one, the target server sees a Comcast, Jio, or BT address, not an AWS or Hetzner range. For scrapers, that single fact is often the difference between a clean 200 and an instant block.

IdentityASNIP ReputationRotationISP
// 02 — definitions

Why the IP
address matters.

Every IP belongs to an Autonomous System with a reputation score. The gap between a residential ASN and a data-centre ASN is the gap between trusted and guilty-until-proven-innocent.

Ask a DataFlirt engineer →

TL;DR

Residential IPs carry the trust of real households. Data-centre IPs are trivially blocklisted by ASN. At scale, pool quality, rotation strategy, and sticky-session management matter far more than raw IP count. The best proxy is one your target has never seen flagged.

01Definition & how the trust works

Every IP address belongs to an Autonomous System (ASN) — a network operated by a single organisation. ISPs like Comcast, Jio, and BT have residential ASNs. AWS, Google Cloud, and Hetzner have data-centre ASNs.

Anti-bot systems maintain constantly updated ASN reputation lists. Data-centre ASNs are treated as guilty until proven innocent. Residential ASNs inherit the trust of the millions of real users who share that address space. A residential proxy routes your traffic through a device sitting on a real ISP connection — so to the target, your request looks like it came from someone's home broadband.

02How proxy pools are structured

A proxy pool is a managed set of exit nodes with metadata: IP, ASN, city, reputation score, last-used timestamp, and active session count. A well-managed pool:

  • Distributes load across multiple /24 subnets — subnet-level blocklisting is common
  • Tracks per-IP velocity and enforces cooldown windows
  • Flags IPs that return anomalous responses (CAPTCHAs, 403s, honeypot HTML)
  • Prefers IPs with geo consistency — city matches the pipeline's target region

Raw pool size is marketing. What matters is clean IP rate — the fraction of your pool that passes a live reputation check right now.

03Rotation strategies
  • Per-request rotation — new IP every request. Maximum anonymity, but breaks session-aware targets and looks unnatural for multi-page flows.
  • Sticky sessions — same IP for the lifetime of one logical task. Matches real user behaviour. Rotate only when the task completes or the IP is blocked.
  • Time-based rotation — rotate every N minutes regardless of session state. Useful for long-running crawls where velocity per IP is the primary risk.

DataFlirt uses sticky-by-default with automatic fallback rotation on block signal detection.

04Where residential proxies still fail

A residential IP is necessary but not sufficient. You can still be caught by:

  • Browser fingerprint mismatch — residential IP + headless Chromium fingerprint = obvious bot
  • Shared pool burnout — providers selling the same IPs to 50 customers accumulate block history fast
  • WebRTC IP leak — unpatched headless browsers can leak the real machine IP via WebRTC, bypassing the proxy entirely
  • Velocity on shared carrier NAT — mobile NAT IPs are watched harder because legitimate abuse also hides behind them

The full identity stack needs TLS fingerprint, browser fingerprint, and residential IP to be coherent with each other — not just any one of the three.

05Legality and responsible use

Residential proxy pools sourced from opt-in SDK networks are legal to operate in India, the US, and the EU when device owners have consented — reputable providers publish their sourcing methodology. Pools built on compromised devices or malware are not legal, and the distinction matters when evaluating providers.

On the usage side, the hiQ v. LinkedIn precedent applies: accessing publicly available data through a proxy is lawful. Bypassing authentication, harvesting personal data, or violating a site's ToS in ways that cause demonstrable harm are separate questions that require counsel.

// 03 — the signals

How a target
scores your IP.

Anti-bot systems don't just check whether an IP is residential — they score it on velocity, subnet history, and geo consistency. The three models below are what DataFlirt's pool manager optimises against.

Block probability by ASN type = P(block) = f(ASN reputation · subnet history · request velocity)
DC ASNs carry near-zero prior trust. Residential ASNs start at ~0.92. DataFlirt measurement fleet, 2026
Sticky session viability window = Tsticky = min(ISP DHCP lease, target session TTL, classifier cooldown)
Rotating mid-session triggers anomaly detection. Hold until the lease boundary. Internal proxy SLO
Pool diversity index = PDI = unique ASNs × unique /24 subnets / total IPs
PDI > 0.4 prevents subnet-level blocklisting. Our active pool: 0.67. DataFlirt fleet planner
// 04 — proxy selection in action

From request to
residential exit node.

Here's the pool selection trace for a single request on an e-commerce pricing pipeline. The selector runs in under 8ms and picks the highest-reputation IP that satisfies geo, cooldown, and sticky constraints.

geo-sticky modeIndia poolASN diversity enforced
edge.dataflirt.io — live
CAPTURED
// pipeline: ecommerce-pricing-IN
request.target: "flipkart.com/search?q=headphones"
selector.strategy: "geo-sticky"

// candidate pool after geo filter
pool.size: 1842 // residential IPs · India
pool.asns: ["AS55836 (Jio)", "AS24560 (Airtel)", "AS18101 (VSNL)"]
pool.subnet_diversity: 0.71 // PDI above threshold

// selected exit node
exit.ip: "49.36.xx.xx"
exit.asn: "AS55836 · Reliance Jio" // residential ✓
exit.city: "Mumbai"
exit.last_used: 14m ago // cooldown satisfied
exit.reputation: 0.94

// result
response.status: 200 OK
cf_challenge_issued: false
// 05 — trust signals

What determines
if an IP gets trusted.

Five signals that anti-bot systems check on every IP before the JS challenge even fires. A residential address is necessary — but it's not sufficient on its own.

POOL SIZE ·  ·  ·  ·  ·   380K active IPs
CLEAN RATE ·  ·  ·  ·  ·  97.4% last 24h
UPDATED ·  ·  ·  ·  ·  ·  every 4h
01

ASN classification

ISP vs DC vs hosting · Checked before any JS fires
02

Subnet reputation history

30d block ledger · Rolling abuse signal per /24
03

Request velocity per IP

req/min rate · Spikes flag shared pools
04

Geo header consistency

IP city vs TZ/Accept · Mismatch = proxy signal
05

WebRTC local IP leak

RFC1918 vs exit IP · Headless giveaway if unpatched
// 06 — how DataFlirt sources its pool

Opt-in, verified,

not compromised devices.

Our residential pool is built on partner ISP integrations and an opt-in SDK network. Every IP passes a liveness check every 4 hours and carries a reputation score derived from our own measurement fleet's observed block rates across 300+ targets. No malware. No grey-market installs.

Active session snapshot

A live proxy assignment for one request on a pricing pipeline.

exit.ip 49.36.xx.xx
exit.asn AS55836 · Jioresidential
exit.city Mumbai · IN
exit.reputation 0.94
session.mode geo-sticky
cooldown.satisfied yes · 14m ago
classifier.score 0.03 · trusted

Stay ahead of the pipeline

Data engineering
intel, weekly.

Anti-bot shifts, scraping infrastructure updates, dataset delivery patterns, and business outcomes from our pipelines. Short, technical, no fluff.

// 07 — FAQ

Common
questions.

About proxy types, pool quality, rotation strategy, and how DataFlirt manages IP identity across long-running pipelines.

Ask us directly →
What's the difference between residential, mobile, and data-centre proxies? +
Data-centre IPs come from cloud and hosting ASNs — trivially blocklisted in bulk. Residential IPs are assigned by ISPs to home subscribers and carry real browsing history. Mobile IPs (3G/4G/5G carrier NAT) are the hardest to block because millions of users share a single IP — but they're expensive and have inconsistent latency. Most production pipelines use residential with mobile fallback for the hardest targets.
How many IPs do I actually need for a pipeline? +
It's not about count, it's about velocity per IP. A single residential IP can sustain roughly 1–3 requests per minute to a cautious target. For 10,000 pages per hour you need ~60–160 IPs in rotation — but they must span multiple /24 subnets and ASNs. A pool of 10,000 IPs all from one provider is worse than 200 well-distributed ones.
What is a sticky session and when should I use it? +
A sticky session pins all requests in a logical workflow to the same exit IP. Use it whenever the target tracks session state: login flows, checkout funnels, paginated searches, or sites that fingerprint IP consistency across page loads. Rotating mid-session is one of the strongest bot signals.
Can I target a specific city or ISP? +
Yes — geo-targeting down to city level is available on all plans. ISP-specific routing (e.g. only Jio IPs for a pipeline) is available on Growth and above. Carrier-NAT mobile IPs with specific MCC/MNC codes are an Enterprise feature.
What happens when a residential IP gets blocklisted mid-pipeline? +
Our edge detects a block signal (non-2xx, CAPTCHA trigger, or HTML hash anomaly) within 1–2 requests and marks the IP for cooldown. The session rotates to a fresh IP and, if sticky mode is active, we replay the session initialisation steps before resuming pagination. Block rates across our active pipelines averaged 0.4% last month.
Is using residential proxies legal? +
Pools sourced from opt-in SDK networks are legal to operate in India, the US, and the EU when device owners have consented. On the usage side, the hiQ v. LinkedIn precedent covers accessing publicly available data through a proxy. Bypassing authentication or harvesting personal data are separate questions — consult counsel for your specific use case.
$ dataflirt scope --new-project --target=residential-proxies READY

Tell us what
to extract.
We do the rest.

20-minute scoping call. Pilot dataset within the week. Production within two. Whether you need a one-off catalogue dump or a continuous feed across millions of records — we scope, build, and operate the pipeline.

hello@dataflirt.com  ·  Bengaluru  ·  IST  ·  typical reply < 4h