IP Reputation Check: How to Detect Risky and Malicious IP Addresses
Introduction
An IP address looks like a simple label, but it carries a lot of context. It can tell you the country and city the traffic is routed from, which network owns the address, and whether that address is tied to a VPN, a proxy, Tor, a hosting provider, or a known threat list. That context is what people mean when they talk about an IP reputation check.
The catch is that reputation only helps if you check it at the moment of the request. Attackers rotate addresses quickly, especially when they hide behind residential connections.
GreyNoise's 2026 report "The Invisible Army" found that, across 4 billion sessions observed over 90 days, 39% of unique IPs targeting the edge came from home internet connections, and 78% of those residential IPs appeared in only one or two sessions before vanishing, too quickly for traditional reputation feeds to flag them. Source: GreyNoise, The Invisible Army: Residential Proxy Abuse in Internet-Scale Attack Traffic
This guide explains what IP reputation is, how it differs from IP geolocation and broader IP intelligence, and how to check IP addresses for risk in signup, login, and checkout flows. Greip supplies the live IP, network, and security signals. Your rules decide when to allow, challenge, or block.
What Is IP Reputation?
IP reputation is a judgment about how risky an IP address looks, based on what is known about that address and the network behind it. A reputation check typically looks for signs of abuse, anonymization, or infrastructure that fraudsters use often, then turns those signs into a score or a risk level.
It is not the same thing as knowing who the person is. An IP address identifies a connection, not a customer. Two honest users can share a mobile network or a coffee shop Wi-Fi address. One attacker can appear to come from hundreds of home connections in a single hour.
That is why a useful IP reputation check is current and specific. It should tell you what is true about this address right now, not only whether it appeared on a public blacklist last month.
IP Reputation vs IP Intelligence vs IP Geolocation
These terms get mixed together, but they answer different questions.
IP geolocation answers "where does this address appear to be?" It estimates the country, region, city, postal code, coordinates, and timezone associated with the IP. That estimate comes from network routing and registration data, not from a GPS reading on the user's device.
IP intelligence is the broader set of facts you can look up for an address. Alongside location, that usually includes the ISP, the Autonomous System Number (ASN), the organization that owns the range, and security flags such as VPN, proxy, Tor, hosting, and relay.
IP reputation is the risk layer on top of that intelligence. It compresses the signals into a score, a risk band, and a recommended action so your application can decide what to do next.
You need all three. Geolocation without reputation will treat a clean home connection in the same country the same way it treats a proxy that happens to egress there. Reputation without location will miss a login that jumps continents in minutes. Intelligence without a decision policy leaves your team staring at flags with no next step.
Why a Static IP Blacklist Is Not Enough
Blacklists still matter. If an address is on a threat list, or on a list you maintain yourself, that is a strong reason to block it. The problem is everything that never stays listed long enough to be useful.
Residential proxy networks and short-lived VPN exits let attackers borrow ordinary ISP addresses, use them once or twice, and move on. By the time a feed-based reputation system catalogs the address, the next request already comes from somewhere else. The GreyNoise research above is a clear example of that rotation.
A live IP reputation check solves a different job. Instead of asking "have I seen this exact address before?", it asks "what kind of network is this, and what risk signals are present on this request?" Hosting ASNs, VPN and proxy services, Tor exits, and bot-like clients can be detected even when the specific IP is new.
Keep your own blacklists for addresses and ranges you already trust or distrust, and combine them with live checks for VPN, proxy, Tor, hosting, and reputation scores. One source is not enough.
What Data Goes Into an IP Reputation Check
A practical check uses a few types of IP data, each for a different reason.
Geolocation data
Greip's IP Location Intelligence returns location details such as country, city, region, postal code, and timezone. These fields help you spot a new country on a login, a shipping address that does not match the card or the IP, or an impossible jump between two sessions.
Treat the location as a network estimate. It is accurate enough for fraud rules, but it is not a street-level GPS fix.
Network and ASN data
Network Intelligence maps an IP to its ASN, the organization that owns the range, and the network type. In Greip's IP Lookup sample, a DigitalOcean address is returned with asn.number set to AS14061, asn.org set to DigitalOcean, LLC, and asn.type set to hosting. That type is often more useful than the IP itself, because many fraud attempts come from cloud and hosting networks rather than home broadband.
This is network registration data, not a full personal WHOIS record. It tells you who operates the range, not the name of the individual sitting at a laptop.
Security and reputation signals
Greip's Proxy & VPN Detection adds the risk layer: VPN, proxy, Tor, hosting, relay, bot, and blacklist flags, plus an IP reputation score. These signals are how you distinguish a normal ISP customer from traffic that is being masked or launched from a data center.
Device context is separate
Some lookups can also return device details such as device type, operating system, and browser. In Greip, that data comes from the request's user agent when you include the device module, not from the IP address itself. The IP does not reveal the phone model or the hardware specifications. Keep those signals in their own bucket so you do not over-claim what IP data can do.
If you compare device and IP signals, use both. A mobile user agent arriving from a hosting ASN, or a desktop browser that changes too often for one account, is more interesting than either signal alone.
How to Check IP Reputation in Practice
The most direct way to check IP reputation with Greip is the IP Scoring lookup at GET /score/ip/{ip}. It returns a composite score from 0 to 100, where a higher number means higher risk, a risk_level of low, medium, or high, and a recommendation of accept, review, or reject.
Here is the documented sample response:
{
"success": true,
"data": {
"ip": "12.12.12.12",
"score": 35,
"risk_level": "medium",
"recommendation": "review",
"rules": [
{
"id": "IP_VPN_OR_PROXY",
"description": "IP address is associated with VPN, proxy, or anonymization service."
},
{
"id": "IP_HOSTING",
"description": "IP address belongs to hosting or cloud provider ASN."
}
],
"security": {
"vpn_or_proxy": true,
"vpn_or_proxy_type": "Socks",
"tor": false,
"bot": false,
"relay": false,
"relay_country": null,
"relay_city": null,
"hosting": true,
"is_blacklisted": false
}
}
}In this sample, the address is not on a blacklist and is not a Tor exit, but it is both a proxy and a hosting IP. The score is 35, the risk level is medium, and the recommendation is to review. That is a good illustration of why a single flag is not a verdict. You would usually challenge this traffic, not approve it blindly and not block every hosting IP in sight.
The rules array lists only the signals that fired. In production logs, those rule ids make reviews much easier. The documented ids are IP_BLACKLISTED, IP_TOR, IP_VPN_OR_PROXY, IP_RELAY, IP_BOT, and IP_HOSTING.
If you already look up the IP for location, you can also pass params=security on IP Lookup or IP Geolocation and read the same security object there. Use IP Scoring when you want the composite score and recommendation in one response.
VPN, Proxy, Tor, Relay, and Hosting Signals
Not every anonymized connection is an attack, and not every hosting IP is a bot. The flags still change how much trust you should give the request.
vpn_or_proxy means the address is associated with a VPN, proxy, or other anonymization service. The vpn_or_proxy_type field tells you the type when it is known, such as VPN or Socks in the documented samples. This is common in account abuse, card testing, and multi-accounting, but privacy-minded customers use VPNs too.
tor means the request came through a known Tor exit node. For payments and account creation, that is usually high risk.
hosting means the address belongs to a hosting or cloud provider ASN. Real customers occasionally come from cloud networks, for example a company VPN hosted in a data center, but most retail checkouts do not.
bot means the client looks automated. Combine it with velocity and the rest of the security object before you decide, and do not block the good bots you actually want, such as your own monitors.
relay means the connection matches a private relay pattern, such as Apple's Private Relay. Many of those users are genuine. A blanket decline on relay will create false positives.
is_blacklisted means the IP is on a Greip threat list or on one of your custom blacklists. That is one of the few flags that usually justifies an immediate block.
A solid starting policy, which Greip also outlines in its checkout guide, is:
- Block when
is_blacklisted,tor, orbotis true - Challenge when
vpn_or_proxyorhostingis true - Allow or use only a light challenge when
relayis true and nothing else is risky - Allow when the security object is clean and the IP reputation score is low
Treat those as inputs to your own policy, not as hard-coded declines for every product.
Do not block every VPN, proxy, or hosting IP. Challenge first on those signals, especially for returning customers, and reserve hard blocks for Tor, confirmed bots, blacklisted addresses, and repeat failures.
Using Geolocation and ASN With Reputation
Reputation tells you about the network's risk. Geolocation and ASN tell you whether that network makes sense for the user in front of you.
Compare the IP country with the account's usual country, the billing country, and the shipping country. A first-time customer paying with a card issued in one country, placing an order from a hosting IP in another, and shipping to a third is a much stronger case for review than any one of those facts alone.
Watch for impossible travel. If the same account was in one city an hour ago and now appears on another continent, step up verification. Greip gives you the location fields. Your system stores the history and measures the gap.
Use the ASN when the IP itself is new. A fresh address from a consumer ISP is different from a fresh address on a cloud range that has already shown up in your card-testing logs. Network Intelligence is useful here because fraud rings often share a small set of ASNs even when they rotate IPs.
Where to Apply IP Reputation Checks
Run the check at the points where a bad IP can cost you money or access.
Signup and trial starts. New accounts created from VPNs, proxies, Tor, or hosting networks are a common source of promo abuse and fake accounts. A medium score should trigger extra verification. A blacklist or Tor hit should stop the signup.
Login and account recovery. This is where credential stuffing and account takeover attempts show up. A new country, a hosting ASN, or a proxy on an account that usually comes from a home ISP is a reason to require a second factor or to delay sensitive changes.
Checkout and payouts. Score the IP before you capture a payment or send funds. Challenge VPN and hosting traffic with a step-up check. Block Tor, bots, and blacklisted addresses.
Account changes. Re-check the IP when someone changes an email address, phone number, password, or payout destination. Attackers often log in with stolen credentials and then move quickly to lock the owner out.
Once you have a score, the next step is deciding what to do with it. Our guide to IP risk scores explains how to read the 0 to 100 result and apply accept, review, or reject actions at signup and checkout without blocking honest VPN users.
Pass a stable user_id on these lookups so you can trace risky events back to one account in the Greip dashboard.
Common Mistakes When You Rely on IP Data Alone
The most common mistake is treating a reputation score as a yes-or-no identity check. It is a risk signal. It should sit next to email, phone, payment, device, and velocity checks, not replace them.
The second is over-blocking. If you decline every VPN or every data center IP, you will also decline remote workers, privacy-conscious shoppers, and some corporate networks. That shows up as false declines and lost revenue.
The third is trusting geolocation as a precise location. It will not give you a street address, and it will be wrong when the user is on a VPN, a relay, or a mobile gateway far from home.
The fourth is relying only on a downloaded blacklist. As the research above shows, many abusive residential IPs never last long enough to be listed. Live intelligence has to fill that gap.
The fifth is reading device details as if they came from the IP. Browser and operating system data come from the user agent. They are useful, but they are easy to spoof and they are not IP intelligence.
IP Reputation Best Practices
Build a simple, consistent habit around these checks.
- Score the IP in real time on signup, login, checkout, and sensitive account changes
- Use the score, the risk level, and the fired rules together, not the score in isolation
- Store country, ASN, and security flags per user so you can compare this request with the last trusted one
- Challenge VPN, proxy, and hosting traffic; block Tor, bots, and blacklisted IPs unless you have a strong reason not to
- Allow for shared mobile networks and honest travelers, especially on accounts with a clean history
- Review false positives and adjust thresholds instead of tightening every rule after one incident
- Combine IP intelligence with email, phone, and payment signals so one noisy IP does not decide the outcome alone
Frequently Asked Questions
What is IP reputation?
IP reputation is an assessment of how risky an IP address looks, based on live intelligence such as abuse history, VPN or proxy use, Tor, hosting networks, and blacklist status. It helps you decide whether to allow, challenge, or block a request.
How do you check IP reputation?
Look up the IP at the time of the request and read its reputation score, risk level, and security flags. With Greip, the IP Scoring lookup returns a score from 0 to 100, a risk band, a recommendation, and the rules that fired, along with VPN, proxy, Tor, hosting, relay, bot, and blacklist signals.
What does an IP reputation score mean?
In Greip, the IP reputation score is a number from 0 to 100, and a higher score means higher risk. Use it with the risk_level and recommendation fields rather than picking a single cutoff on your own. A medium score with VPN and hosting flags, like the documented sample of 35, is usually a review or challenge, not an automatic decline.
What is the difference between IP reputation and IP geolocation?
IP geolocation estimates where an address is routed from, such as the country, city, and timezone. IP reputation estimates how risky that address is. You typically need both: location tells you whether the request fits the user, and reputation tells you whether the network itself looks abusive or anonymized.
Can an IP reputation check detect VPNs, proxies, and malicious IP addresses?
Yes. A live check can flag VPN and proxy services, Tor exits, hosting networks, bots, private relays, and addresses on a threat list or your own blacklist. Those flags will not catch every malicious IP, especially short-lived residential proxies, so you should also watch velocity, account history, and other identity signals.
Does a VPN or hosting IP always mean fraud?
No. Plenty of real customers use a VPN, and some businesses reach you from cloud or hosting networks. Those signals are a reason to add friction, not a reason to block everyone. Hard blocks are a better fit for Tor, confirmed bots, blacklisted addresses, and repeated high-risk behavior.
Conclusion
IP data becomes useful when you treat it as intelligence, not as a static list of bad addresses. Geolocation, ASN, and security flags tell you where a request comes from and what kind of network sent it. The reputation score turns those facts into an action you can apply in real time.
Start with a live IP reputation check on signup, login, and checkout. Challenge anonymized and hosting traffic, block the clearest threats, and keep your rules flexible enough for honest customers. Greip gives you the IP location, network, and reputation signals. Your policy decides what happens next.
Get started
Start protecting your business today
Our service is trusted by thousands of businesses worldwide.
- 1,000 requests during trial
- Cancel anytime