Skip to content

IPQS (IPQualityScore): What the Fraud Score Means

Compare IP fraud detection: transparent vs black-box scoring, VPN detection, and pricing.

By Last updated: Nov 8, 2026

0:00
0:00

IPQS (IPQualityScore) is a fraud-prevention company whose fraud score rates an IP address, email, phone number or transaction from 0 to 100. For an IP address, IPQS calls 75 or more suspicious and recommends blocking from 90. This page explains what the score is built from, what IPQS costs, and how an IP-only risk score compares.

My IP Help is our product, so read the comparison with that in mind. We read the IPQS pages cited here on 8 November 2026; prices are as of that date. One difference matters more than the rest: IPQS can score the person and the transaction, while we score only the IP address.

What is IPQS (IPQualityScore)?

IPQS sells fraud detection to businesses: APIs that check IP addresses for proxies, VPNs, Tor and bots, validate emails and phone numbers, fingerprint devices and score transactions. Its home page says it uses “proprietary fraud detection data from our honeypot network” and that customers “benefit from shared insights across 100m+ daily transactions from 3,500+ businesses.” In other words, its scores draw on what its network sees, not only on public lists.

Combination padlock on a laptop keyboard beside two bank cards
Credit: Towfiqu barbhuiya via Unsplash

The IP product is the Proxy and VPN Detection API. Besides the fraud score, its response parameters include yes/no flags for proxy, VPN, active VPN, Tor and crawlers, the connection type, recent_abuse (“recently verified abuse across our network”, such as a chargeback, account takeover or bot attack), abuse_velocity (high, medium, low or none) and bot_status. Two more, frequent_abuser and high_risk_attacks, are marked as Enterprise data points.

What the IPQS fraud score means

IPQS defines fraud_score as “the overall fraud score of the user based on the IP, user agent, language, and any other optionally passed variables.” So it isn’t a fixed property of the address: the same IP can score differently with a different browser, language or transaction details attached. Its documentation describes the levels like this:

Fraud score (as IPQS writes it)IPQS labelIPQS description
< 75Suspicious“Has had previous reputation issues or is using a low risk proxy/VPN.”
< 85High Risk“Has suspicious behavior signals.”
< 90Frequent Abusive Behavior“Has demonstrated frequent abusive behavior over the past 24-72 hours.”

IPQS’s table prints “<” where its text means “at least”. Its labels put High Risk at 85, while its guidance says to flag or block from 90.

Its advice on acting on the number is plain. “Fraud Scores >= 75 are suspicious, but not necessarily fraudulent.” And: “We recommend flagging or blocking traffic with Fraud Scores >= 90, but you may find it beneficial to use a higher or lower threshold.” It adds that a score of 75 or more is “likely to be a proxy, VPN, or TOR connection” and “could indicate that users are protecting their privacy online”.

Two request settings change the result. strictness sets how deep the check goes; higher values “may provide a higher false-positive rate”, and IPQS recommends 0 or 1 for the most accurate results. allow_public_access_points relaxes some checks for schools, research institutions and some corporations, as the API overview explains. Its email validation API uses the same wording: an email fraud score of 75 or more is “suspicious, but not necessarily fraudulent”.

Why your IP fraud score is high

Say a site blocked you and an IPQS check shows a high score for your address. The usual reasons follow from IPQS’s own definitions. You may be on a VPN or proxy. The address may be shared (a mobile carrier, a workplace, a public Wi-Fi) and someone else on it misbehaved recently. Or a device on your network is infected and running bots.

Turning the VPN off, or reconnecting to get a new address, often changes the result. Our IP reputation check shows which public lists name your address, which can point to the cause.

IPQS pricing and free tier

The IPQS plans page offers “1,000 Free Lookups” to start, with a free account. The plan data on the same page lists these packages (8 November 2026):

PackageMonthlyYearly
Free$0–
Startup$139.99$1,199.88
SMB Basic$599.99$5,999.88
SMB+$1,149.99$11,988.00
EnterpriseBy arrangementBy arrangement

The higher packages add more custom rules, alerts, “Premium Data Points for IP, Email, Phone, URL, Device” and features such as Fraud Fusion and residential proxy data. The packages cover all IPQS products, not the IP API alone, and the page also offers a demo for larger needs.

How My IP Help scores an IP

Our risk score answers a narrower question: what is this IP address, and which lists name it? It is computed from lists we load and refresh on a schedule, not from customer traffic, and it doesn’t take a user agent, an email or a transaction. Points add up, capped at 100: Low is 0 to 29, Medium 30 to 69, High 70 to 100. Every signal that adds points is shown with its source.

SignalPointsSource
Tor exit node75Tor Project exit list
In a hijacked or criminal netblock70Spamhaus DROP (© The Spamhaus Project)
On public blocklists20 for one list, up to 60IPsum (aggregates 30+ lists)
On FireHOL’s attacker lists45, 30, 20 or 10 (levels 1 to 4)FireHOL
Public proxy45Public proxy lists
Listed as a proxy (if the public proxy lists don’t already count it)40IP2Proxy LITE
VPN service30VPN provider ranges
Datacenter or cloud15Published cloud and datacenter ranges
iCloud Private Relay5Apple’s relay ranges

The bands are not the same scale as IPQS’s. Our High starts at 70 and a VPN alone scores 30, Medium; IPQS’s levels start at 75. Compare the reasons, not the numbers.

Through the IP risk API, every plan gets the score, its band and the main reason (risk_reason); the full list of signals with their points (risk_signals) comes with Pro and up. The free Starter plan includes 1,000 lookups a month, and Pro costs $29 a month for 100,000.

The API leaves out two of the lists the website uses: IPsum and FireHOL levels 1 and 2. Both include DShield data, whose terms don’t allow use in an API for now. So an address can score lower through the API than on the website.

Feature comparison

Feature (as of November 2026)My IP HelpIPQS
What is scoredThe IP addressThe IP with the user agent, language and optional user or transaction data; also emails, phones and devices
Score0–100, Low / Medium / High (High from 70)0–100; suspicious from 75, blocking recommended from 90
Built fromPublic and licensed lists, each named on the resultIts own network: honeypots and data shared across its customers
Reasons shownEvery signal with its points and sourceFlags and fields such as recent_abuse and abuse_velocity; the formula isn’t published
Free web checkYes, no signupYes (IP fraud score checker)
Free API1,000 lookups a month: score, main reason, location, network1,000 free lookups with a free account
Paid APIFrom $29 a month for 100,000 lookups, with every signalPackages from $139.99 a month, all products
Email, phone, device, transaction scoringNoYes
Bulk checkPaste up to 100 IPs; first 5 free; reports from $1.99CSV uploads on paid packages
Location and ownerMaxMind and DB-IP location, ISP, ASN, registry recordCountry, region, city, ISP and ASN fields
Network tools26 free tools, such as DNS lookup and the port checkerFraud lookup tools (IP, email, phone, URL); no DNS, WHOIS, port or traceroute tools

Key differences

The IP, or the person and the transaction

IPQS is built to decide whether a sign-up, login or payment is fraud, and it can weigh the browser, the email, the phone number and the order alongside the IP. We score only the IP address: whether it is a VPN, proxy, Tor exit or datacenter address and which blocklists name it. That never makes a verdict about the person behind it, and it won’t catch a fraudster on a clean home connection. If you need transaction-level fraud scoring, IPQS does more than we do.

Network data vs public lists

IPQS sees abuse across its customers and honeypots, so it can flag an address that misbehaved somewhere in the last few days before any public list does, and it says so through fields like recent_abuse. Our lists are public or licensed and refreshed on a schedule; they are slower to catch something new, but you can see exactly why an address scored what it did, and check the source yourself.

Price and volume

IPQS’s paid packages start at $139.99 a month and cover its whole product range. Our Pro plan costs $29 a month for 100,000 IP lookups with every signal, and the web lookups are free without an account. For IP-only checks at moderate volume we cost less; for a full fraud stack, compare IPQS with the tools you would otherwise combine.

When to use each tool

Choose IPQS when:

  • you need to score sign-ups, logins or payments, not just the IP
  • you also want email, phone or device checks from the same provider
  • you want signals from a shared fraud network, such as recent abuse seen at other businesses
  • you need many custom rules and tuning (strictness, thresholds) on a large volume

Choose My IP Help when:

  • you want an IP risk score with every reason and source shown
  • you need VPN, proxy, Tor, datacenter and blocklist checks at a lower price
  • you also need the address’s location, owner and network tools
  • you want to check a list of up to 100 addresses at once and export the results
Open padlock among scattered keyboard keys under red and green light
Credit: FlyD via Unsplash

Frequently Asked Questions

What is IPQS?

IPQS, short for IPQualityScore, is a fraud-prevention company. Its APIs check IP addresses for proxies, VPNs, Tor and bots, validate emails and phone numbers, fingerprint devices and score transactions, each with a fraud score from 0 to 100 drawn from its own network and honeypots.

What does the IPQS fraud score mean?

It is IPQS’s estimate of the risk behind a request, from 0 to 100, based on the IP, user agent, language and any other data you send. IPQS calls 75 or more suspicious, though not necessarily fraudulent, and recommends flagging or blocking at 90 or more.

What is a good IP quality score?

IPQS doesn’t define a good score; it defines when to worry. From 75 it calls an IP suspicious, often a proxy, VPN or Tor connection, and it recommends flagging or blocking from 90. A lower score isn’t a guarantee that the visitor is legitimate.

Why is my IP fraud score high?

Most often because you are on a VPN or proxy, or your address is shared with others, such as a mobile carrier, an office or public Wi-Fi, and someone else on it misbehaved recently. An infected device on your network can do it too. Turning the VPN off or getting a new address usually helps.

Is IPQS free?

Partly. IPQS offers 1,000 free lookups with a free account and a free web checker. Its plans page lists paid packages from $139.99 a month for Startup to $1,149.99 for SMB+, with Enterprise by arrangement, as of November 2026.

Does My IP Help score the person or the IP?

Only the IP address. Our score says what the address is (a VPN, proxy, Tor exit or datacenter address) and which blocklists name it, with the source of every point. It doesn’t score the person, an email or a transaction, and it makes no fraud verdict about anyone.

Why do IPQS and My IP Help give different scores?

They measure different things on different scales. IPQS uses its own network’s data and the details of the request; we use public and licensed lists about the address alone. Our High band starts at 70; IPQS recommends blocking from 90. Compare the reasons each gives, not the numbers.

Is there a free alternative to IPQS?

For IP checks, yes. My IP Help scores any IP from 0 to 100 for free with no signup and shows every reason, and its API includes the score, main reason, location and network free for 1,000 lookups a month. For email, phone or transaction scoring it is no alternative.

Does a high fraud score mean the user is a fraudster?

No. IPQS itself says a score of 75 or more is “not necessarily fraudulent” and may just mean someone protecting their privacy with a VPN. Treat a high score as a reason to ask for more, such as a captcha or a verification step, rather than as proof.

Can I use both tools together?

Yes. A typical setup uses IPQS for the decision on a payment or sign-up and our lookup to see why an address is flagged: which list names it, who owns the network and where it is. If both flag an address, you have a score and the reasons to act on it.

Try it yourself

Score an address with the IP reputation check, test your own connection with the VPN and proxy detector, or open any address on the IP lookup. For lists, use the bulk IP lookup; developers can start with the IP risk API and the API docs. For a score built from reports instead, see how AbuseIPDB compares.

Create your free account

Get access to IP lookup tools, bulk reports, and more. Free forever.