What Is a Phone Number Fraud Score? How Risk Scoring Works
A phone number fraud score turns a dozen quiet signals about a number into one risk rating. Here's what goes into it, how to read it, where to set your cutoffs, and the mistakes that cause good customers to get flagged.
On this page
A phone number fraud score is a single number that estimates how likely a phone number is to be tied to fraud, based on signals like the line type (mobile, landline or VoIP), the carrier, how long the number has been active, recent changes to the line, and how the number has been used. In our lookups it runs from 0 to 100, and higher means riskier. A long-held mobile number on a major carrier usually lands in the single digits. A freshly activated app-based number that’s shown up on a pile of sign-ups lands near the top.
It’s a probability, not a verdict. Most of getting value out of one comes down to that distinction.
What a fraud score is really answering
The score answers one narrow question: does this number look like the numbers fraudsters tend to use?
It doesn’t answer whether the person in front of you is honest. It doesn’t confirm who owns the number. And it says nothing about whether they’ll pay rent on time. Those are different questions with different tools (an identity match, and a credit file).
The phone is such a useful signal because phone numbers have quietly become the skeleton key for modern identity. Banks text one-time codes to them. Apps use them as usernames. And fraudsters need lots of them, cheaply, and they need them to be disposable. That need leaves fingerprints: new numbers, internet-based numbers, numbers that hop carriers, numbers that show up on many applications under different names.
The Federal Reserve has called synthetic identity payments fraud, where real details like a Social Security number get mixed with a made-up name, address or birth date, a “fast-growing but little-understood problem” that can slip past standard identity and credit screening (Federal Reserve on synthetic identity fraud). Those made-up identities still need working contact details, and fraudsters reuse them. One number tied to six different names is a pattern. A fraud score is a way to catch patterns like that without anyone eyeballing a spreadsheet.
The signals that move a phone number fraud score
Every provider’s model is different and nobody publishes their exact weights, ours included. The inputs, though, come from the same small family across the industry.
Line type
Mobile, landline, or VoIP. And within VoIP, the important split is between fixed VoIP (a business phone system or a cable company home line, tied to a physical address and an account) and non-fixed VoIP (numbers you can create in an app in two minutes with an email address). Non-fixed VoIP numbers are cheap and disposable, which is exactly why fraud skews toward them. We wrote up the differences in mobile vs. landline vs. VoIP.
Carrier and account type
Major postpaid carriers verify customers, run credit checks of their own, and bill monthly. Prepaid plans and smaller resellers often require much less. None of that makes a prepaid customer suspicious (millions of people prefer prepaid), but it shifts the base rate. Our carrier lookup guide covers what carrier data does and doesn’t tell you.
Tenure
How long has this number been active with this owner? A number that’s been in continuous use for eight years is expensive to fake. A number activated last Tuesday is not. Tenure is one of the strongest single signals, and it’s also the one that most often trips up honest people who just switched phones.
Recent changes to the line
A recent port to a new carrier or a recent SIM change can be totally normal, or it can be the fingerprints of a SIM swap, where a criminal talks a carrier into moving the victim’s number to a device they control. The FCC adopted rules in late 2023 requiring wireless carriers to use secure methods to authenticate customers before executing a SIM change (47 CFR § 64.2010), which tells you how big the problem had become.
Usage and velocity
How often has this number appeared in identity checks recently, and under how many different names? A number seen once in two years is quiet. A number checked forty times this week across different identities is loud.
Consistency with the identity
Does the caller ID name roughly fit the person? Does the number’s name match the name on a credit file? Mismatches are common and often innocent (family plans, old CNAM records), but they still feed the picture.
Reading a score: three fictional numbers
Here’s how three lookups might come back, and how we’d think about each one.
(555) 555-0118, score 6. Mobile, major carrier, held for years, no recent port, name on the line matches the applicant. This is the boring, good result. Move on to whatever you were actually trying to decide.
(555) 555-0163, score 47. Mobile, prepaid, activated about two months ago, no other flags. Middle of the road. Lots of real people look like this: someone who just moved, someone who lost a phone, someone who switched to a cheaper plan. Ask one more question. “Is this the best number to reach you? Do you have a second one?” Or confirm the person matches the credit file on the number.
(555) 555-0191, score 92. Non-fixed VoIP, created recently, seen on many recent checks under different names, no caller ID name. That’s a strong signal. Don’t accuse anyone of anything, but don’t hand over keys, ship goods on terms, or release a deposit based on a phone call from that number. Verify through a channel you control.
Notice that the middle case is where judgment lives. The top and bottom mostly decide themselves.
Where to set your thresholds
There’s no universal cutoff. The right one depends on what a mistake costs you in each direction: letting fraud through versus annoying or losing a good customer.
| Situation | Reasonable approach |
|---|---|
| Landlord screening a rental inquiry | Low scores: proceed. Middle: verify identity before the showing. High: require an in-person ID check and a callback to a different number. |
| Lender or broker taking a loan lead | Route high scores to manual review before any money moves. Never auto-decline on the phone score alone. |
| Business opening net terms | Middle or high: ask for trade references and start with a small first order. |
| Selling a car or item to a stranger | High: insist on meeting at a bank or police-station lot and verify payment with the bank directly. |
| Website or app sign-up | High: step up verification (ID upload, card check) instead of blocking outright. |
Two rules hold everywhere. First, use the score to decide how much more to verify, not to make the final decision. Second, write your thresholds down and apply them the same way to everyone. Ad hoc judgment about who “seems off” is how unfair treatment creeps in.
The mistakes we see most
Treating VoIP as automatically bad. Plenty of small businesses, remote workers and people in rural areas use VoIP numbers that are perfectly legitimate. Fixed VoIP with a long tenure is usually fine. The score already accounts for this, so don’t stack an extra penalty on top.
Over-trusting a low score. A clean number doesn’t mean a clean caller. Scammers spoof trusted numbers all the time, and a spoofed call borrows the real number’s good reputation. If a “customer” calls from a familiar number asking you to change payment instructions, hang up and call back on the number you already have on file.
Ignoring the score because the credit looks good. A strong credit file on a high-risk number can be someone using a stolen identity: the credit belongs to the victim, the phone belongs to the thief. That’s exactly the scenario that pairing the two catches. See identity verification by phone number for how to put the checks together.
Using a free lookup and calling it a fraud check. Free reverse lookups often show stale names and nothing about line risk. Our rundown of what a reverse phone lookup actually shows explains where those names come from and why they drift.
Fraud score plus credit: why we return both
Every lookup we run returns the phone details (line type, carrier, caller ID name and the fraud score) alongside the person’s name and current address from their TransUnion credit file and their FICO® Score 4 with a credit profile summary. The phone side tells you whether the line looks risky. The credit side tells you who the number belongs to on paper and how they handle money. Each covers a blind spot in the other. Our how it works page walks through a full result.
The credit portion is a soft pull, so the person isn’t notified and their score isn’t affected, and it requires a permissible purpose under the FCRA. The simplest is the person’s written OK. If you’re running numbers at volume, from a sign-up flow or a CRM, the API returns the fraud score as a field you can route on. Typical setups flag anything above a cutoff for a human to review.
For the bigger picture of how fake and stolen identities get built, start with synthetic identity fraud. And if a renter, buyer or caller is pushing you to skip steps because they’re “in a hurry,” that urgency is its own signal. It’s in nearly every scam script we’ve read about, and it shows up again and again in the CFPB’s guides to common frauds and scams.
Frequently asked questions
Is a high phone number fraud score proof that someone is a scammer?
No. It means the number shares traits with numbers that have been involved in fraud. Plenty of honest people use prepaid or app-based numbers. Treat a high score as a reason to verify more, not as a verdict.
Can a fraud score detect a spoofed call?
Not by itself. A fraud score rates the number. If a caller spoofs a clean, long-held number, the score reflects the real owner of that number. Spoofing is caught by call authentication on the carrier side and by calling the person back on a number you already trust.
Does checking a number's fraud score affect anyone's credit?
No. The phone details and fraud score are about the phone line, not the person's credit file. In our lookups the credit portion is a soft pull, which also doesn't affect scores.
