IP Reputation: What It Is, How to Check Yours and How to Fix It
What IP reputation means for email and for websites, who keeps the scores, why a home or mobile IP gets flagged, how to check yours and which fixes work.
In this guide
IP reputation is a score or listing that mail servers, websites and fraud systems keep about an IP address, built from what that address has done before: spam, attacks, proxy use, or nothing at all. There is no single reputation. Each list has its own evidence and its own fix, and consumer addresses inherit a score from whoever used them last.
Test it now
Cloudflare edgeRead at the Cloudflare edge from this request. The homepage adds your provider, network type and the VPN, DNS and WebRTC checks.
Show detailsEmail reputation and web reputation are different things
The phrase covers two different worlds that share a name. Email reputation decides whether mail from an address reaches an inbox. Web reputation decides whether a website lets a visitor from that address through, challenges them, or blocks them. The evidence, the operators and the fixes differ, and the reason a person searches for "ip reputation" is usually that one of the two has gone wrong for them without any obvious cause. This guide explains who keeps each kind of score, why an ordinary address ends up with a bad one, how to see what yours looks like, and what actually changes it.
Who keeps IP reputation scores
Nobody runs a central registry. Reputation lives in many independent systems, and an address can be clean in all but one.
Public DNS blocklists. Operators such as Spamhaus publish lists that mail servers query for every incoming connection. The lists differ in what they claim. The Spamhaus PBL is not an accusation at all: it marks "end-user IP address ranges from which email should never be sent directly", which covers most home and mobile connections by policy. The Spamhaus XBL is different: it lists addresses "exhibiting signs of compromise", meaning Spamhaus has evidence that a device using the address is infected or hijacked.
Crowd-sourced abuse databases. AbuseIPDB collects reports from administrators about attacks, spam and fraud, and publishes a "Confidence of Abuse" rating from 0 to 100 that its FAQ describes as "a very conservative value" based on user reports. Many firewalls and hosting panels consult it automatically.
Mail providers' private scores. Gmail, Microsoft and the large mailbox providers keep their own reputation for every sending address and domain, visible only to the sender through tools such as Postmaster Tools. Google's sender guidelines tell bulk senders to keep the spam rate reported there below 0.3 percent and to have valid forward and reverse DNS for sending addresses; the score itself is never published.
Anonymizer and hosting databases. Geolocation and fraud vendors classify addresses by what kind of network they belong to. MaxMind's Anonymous IP database has separate flags for VPN providers, hosting providers, public proxies, Tor exit nodes and suspected residential proxies. These are not about behavior; they are about the address's neighbors. A website that buys this data treats every address in a flagged range the same way.
Websites' own history. Any large site keeps a memory of what each address did: failed logins, scraping, chargebacks, account creation bursts. This is the reputation you cannot look up anywhere, and it is the one behind most "unusual traffic" and "temporarily blocked" messages. The your IP has been temporarily blocked guide covers that case in detail.
One widely cited score has quietly gone away. Cloudflare's per-request threat score, which many site owners used in firewall rules, is documented as always 0 now; the company's bot detection has moved to other signals. If a guide tells you to check your "Cloudflare threat score", it is out of date.
Why a normal home or mobile IP gets a bad reputation
Most people who look this up have done nothing wrong. Their address has.
You inherited it. Consumer addresses are dynamic. The ISP hands you an address that someone else used last week, and any listing earned then comes with it. Mobile networks make this worse: carrier-grade NAT, defined in RFC 6598 with the 100.64.0.0/10 shared range, puts hundreds or thousands of phones behind one public address at the same time. One infected phone, or one person running a scraper, scores the address for everyone behind it. The CGNAT guide explains how to tell whether you are behind one.
A device on your network is compromised. An XBL-type listing means something using your address contacted a sinkhole, sent spam or probed other networks. On a home connection the usual suspects are an old router with remote management exposed, a smart TV or camera, or a PC with a browser extension that quietly relays traffic. Residential proxy networks recruit exactly these devices, which is why a "residential proxy" flag can appear on a household address whose owner never installed a proxy.
Your ISP's range looks like hosting. Some ISPs, especially small or business-oriented ones, register their ranges in ways that geolocation vendors classify as datacenter or hosting. Every customer then looks like a server. The residential vs datacenter IP guide shows how that classification is made and why it is sometimes wrong.
You use a VPN, relay or proxy. This is the one case where the flag is accurate. A VPN exit address is shared by thousands of users and is listed in every anonymizer database, so sites that distrust anonymizers distrust you. Apple's iCloud Private Relay produces the same effect on Safari traffic for people who never thought of themselves as VPN users; the iCloud Private Relay guide explains how to recognize it.
You send email from the wrong place. A home address on the PBL is not a mistake to be corrected; it is the policy. Mail must go through your provider's authenticated submission server, and a device on your network that tries to deliver mail directly on port 25 is treated as a spam source by design.
How to check your IP reputation
Because there is no single score, check the three views a website or mail server would take. Each tool below reports the address of the connection you open it from, so run them from the network you are asking about, and run them again from a phone on mobile data to compare.
- Blocklists. Open the blacklist check. It queries your address against ten public DNS blocklists and names each one that lists you. A PBL-only listing on a home connection is normal and affects nothing but direct mail sending. An XBL or abuse-type listing means a device problem, and the why is my IP blacklisted guide walks through finding it.
- Anonymizer and network type. Open the proxy check. It tells you whether the address appears on published VPN, proxy, Tor or relay lists and whether the announcing network is a consumer ISP, a carrier, a business or a hosting provider. If you are not using a VPN and the result says hosting or VPN, the problem is your ISP's range or a relay you did not know was on.
- The fraud-system view. Open the IP fraud score. It combines the published signals above (anonymizer ranges, blocklists, ASN type) into an estimate of how a fraud-screening system would rate the address. It is an estimate from public data, not a verdict and not any vendor's private score, but a high value on a plain home connection is a sign that one of the causes in the previous section applies.
- The address itself. Open the IP lookup and note the ASN, the registry country and the reverse DNS name. Reverse DNS that reads like a server name, or a registry country that differs from where you are, is often what tips a site's rules.
To check a sender rather than yourself, find the sending address in the message's headers first. Gmail's full-header view shows the Received lines; the address in the last Received line before your provider's own servers is the one the provider judged. Look that address up with the same tools.
How to fix a bad IP reputation
The fix depends entirely on which list flagged you, which is why the checks above come first.
For a compromised-device listing, the listing is a symptom. Disconnect devices one at a time, update router firmware, remove browser extensions you do not recognize, and run security software on PCs. Most behavior-based lists expire a listing automatically once the activity stops; the operator's page says how long that takes and whether a manual request is possible. Requesting removal before the device is clean leads to an immediate relisting.
For an inherited dynamic address, get a new one. Power the router off for longer than your ISP's lease time, often several hours, and check whether the address changed. If it did not, your ISP may issue static-like addresses, and the only route is the ISP's support desk. The change your IP address guide covers the options by connection type.
For a wrong network classification, the fix is not yours to make, but you can trigger it. Vendors accept corrections from network operators: MaxMind, for instance, asks operators to join GeoIP Exchange and publish a geofeed rather than file one-off corrections. Tell your ISP which vendor misclassifies the range and where the correction form is; small ISPs often do not know their range is flagged.
For email you send, the reputation belongs to the sending system, not to you as a person. If you send from a home connection, stop and use your provider's authenticated submission server. If you run a mail server or a bulk-sending domain, follow the receiving providers' published requirements: authentication (SPF, DKIM, DMARC), matching forward and reverse DNS, and a spam complaint rate below the thresholds in Google's sender guidelines. Delisting requests are granted on evidence that the cause is fixed, and every major list says so on its removal page.
For a VPN or relay flag, decide whether you want the flag. Turning the VPN off for the site that blocks you is the only fix; no reputation process will clear an address that is, in fact, a shared exit.
What does not work
Reputation is attached to the address and the network, so changing your browser, clearing cookies, or reinstalling the operating system does nothing for an IP listing. Paid "IP reputation repair" services can only file the same removal requests you can file yourself, and they cannot change a classification that the list operator considers correct. Switching to a proxy to escape a flag replaces one shared address with another that is, by definition, on the anonymizer lists.
Finally, keep the two worlds separate when you read a result. A home address on the PBL, a mobile address behind CGNAT that one site rate-limits, and a server address that Gmail is deferring are three different problems with three different owners. The tools tell you which one you have; the fix follows from that.
Sources
- Spamhaus, Policy Blocklist (PBL)
- Spamhaus, Exploits Blocklist (XBL)
- AbuseIPDB, Frequently Asked Questions (Confidence of Abuse rating)
- Gmail Help, Email sender guidelines (spam rate thresholds, PTR records, Postmaster Tools)
- MaxMind Developer Portal, GeoIP Anonymous IP Databases (field definitions)
- MaxMind, Submit a data correction (GeoIP Exchange and geofeeds)
- Cloudflare Ruleset Engine docs, cf.threat_score field (threat score now always 0)
- RFC 6598, IANA-Reserved IPv4 Prefix for Shared Address Space (IETF, April 2012)
- Gmail Help, Trace an email with its full header
Questions
What is IP reputation?
A judgment about an IP address based on its past behavior, kept separately by mail providers, blocklist operators, fraud-detection vendors and large websites. Each one scores the address from its own evidence, such as spam reports, attack logs or proxy usage, and uses the score to accept, filter, challenge or block traffic from it.
What's my IP reputation?
There is no single answer because every list is separate. Check the address against public DNS blocklists, against anonymizer and hosting ranges, and against the fraud signals a website would see. CheckMyIP's blacklist check, proxy check and IP fraud score show those three views for your current address.
How do I fix IP reputation?
Find the list that flagged you and follow its removal process, because each one has its own. For a compromised-device listing, clean the device first. For a dynamic home address, a new lease from your ISP often clears it. For a sender address, fix authentication and complaint rates before asking for delisting.
How can I check the IP reputation of a sender?
Open the message's full headers and read the Received lines to find the address that handed the message to your provider, then look that address up on the public blocklists and reputation checkers. Postmaster tools from the receiving provider show reputation only to the domain owner, not to recipients.
Back to your own address: what is my IP address

