Fraud Risk Score API

Q: How is the fraud risk score calculated?

Each signal contributes a weighted amount to a 0-100 total: email quality (disposable domains, unroutable addresses, role accounts), network signals (VPN, proxy, Tor, datacenter, bad ISPs), geolocation mismatch graded by distance, appearances on abuse blocklists, and payment signals when Stripe is connected. Weights reflect how strongly each signal correlates with real fraud, so one weak signal does not condemn a legitimate user.

Q: What is an account risk score?

An account risk score is a single number that summarises how likely a signup, login, or order is to be fraudulent, based on everything observable at that moment: the email address, the IP, the location, and past behaviour. It replaces a pile of separate true-or-false checks with one value you can set thresholds against.

Q: What score should I block at?

It depends on the cost of being wrong in each direction. Most teams start by blocking above 80, reviewing 50-80, and allowing below 50, then tighten at high-stakes endpoints like checkout and loosen at low-stakes ones like newsletter signup. Because Fidro returns the contributing reasons, you can audit the boundary cases and adjust with evidence.

Q: Can I see why a request scored high?

Yes. Every response includes the list of signals that fired and their contribution, for example disposable email domain, datacenter IP, and country mismatch. That makes manual review fast and lets you explain decisions to affected users.

Q: Is a risk score better than separate fraud checks?

For making a decision, yes. Separate checks leave the hardest part, combining them, to you. A weighted score encodes how signals compound: a VPN plus a fresh disposable email plus a mismatched country is far riskier than any one of those alone, and the score reflects that without custom logic on your side.

Send an email address and an IP, get back a single 0-100 risk score with an allow, review or block recommendation and the reasons behind it. No custom scoring logic to write, no thresholds to guess.

What is Fidro?

Fidro is a fraud detection API that combines email validation, IP intelligence, geolocation analysis, and Stripe chargeback prevention in a single API call. Built for developers at startups and growing companies, Fidro returns a risk score with a clear recommendation — allow, review, or block — so you can stop fraud without building your own scoring logic. Fidro offers a free plan with 1,000 validations per month, no credit card required.

The Problem

Raw fraud signals do not make decisions. A VPN flag, a disposable email check, and a geolocation mismatch are three booleans from three vendors, and someone on your team still has to decide how they combine, what outweighs what, and where the block threshold sits. Most teams either hard-code brittle rules or wave everything through.

The Fidro Solution

Fidro calculates the score for you. Every check contributes a weighted amount: disposable or unroutable email, VPN, proxy, Tor or datacenter IP, location mismatch graded by distance, blocklist appearances, and payment signals when Stripe is connected. The result is one number, one recommendation, and a readable list of which signals fired, so review queues see the why, not just the what.

Key Features

Weighted, Not Boolean

A VPN alone scores differently from a VPN plus a throwaway email plus a country mismatch. Signals compound the way real fraud does.

Allow / Review / Block

The score maps to a clear recommendation, with thresholds you can tune per endpoint: stricter at checkout, looser at newsletter signup.

Reasons Included

Every response lists the signals that contributed, so manual review takes seconds and support can explain a rejection.

0-100

One score per check

<200ms

Average response

200

Free checks/mo

Learn more about Fidro

Explore our features, see how we compare, or try Fidro for free.

All features API documentation Free email checker Free IP checker Compare Fidro

Frequently Asked Questions

How is the fraud risk score calculated?

Each signal contributes a weighted amount to a 0-100 total: email quality (disposable domains, unroutable addresses, role accounts), network signals (VPN, proxy, Tor, datacenter, bad ISPs), geolocation mismatch graded by distance, appearances on abuse blocklists, and payment signals when Stripe is connected. Weights reflect how strongly each signal correlates with real fraud, so one weak signal does not condemn a legitimate user.

What is an account risk score?

An account risk score is a single number that summarises how likely a signup, login, or order is to be fraudulent, based on everything observable at that moment: the email address, the IP, the location, and past behaviour. It replaces a pile of separate true-or-false checks with one value you can set thresholds against.

What score should I block at?

It depends on the cost of being wrong in each direction. Most teams start by blocking above 80, reviewing 50-80, and allowing below 50, then tighten at high-stakes endpoints like checkout and loosen at low-stakes ones like newsletter signup. Because Fidro returns the contributing reasons, you can audit the boundary cases and adjust with evidence.

Can I see why a request scored high?

Yes. Every response includes the list of signals that fired and their contribution, for example disposable email domain, datacenter IP, and country mismatch. That makes manual review fast and lets you explain decisions to affected users.

Is a risk score better than separate fraud checks?

For making a decision, yes. Separate checks leave the hardest part, combining them, to you. A weighted score encodes how signals compound: a VPN plus a fresh disposable email plus a mismatched country is far riskier than any one of those alone, and the score reflects that without custom logic on your side.

Start catching bad signups in the next 5 minutes

Create your account, grab your API key, and send your first request. Free plan with 200 validations/month. No credit card. Cancel anytime.

Start free