Skip to content

Fraud detection API with a score you can explain

Send any mix of email, phone and IP. Get a 0 to 100 fraud score, a risk level and the exact rules that fired. Each check costs 3 credits.

3 credits / checkValidateCSV bulkMCP
POST/v1/fraud-detection
{
  "success": true,
  "data": {
    "score": 72,
    "level": "high",
    "rules_fired": [
      {
        "rule": "voip_number",
        "signal": "phone",
        "detail": "Line type is VoIP."
      },
      {
        "rule": "datacenter_ip",
        "signal": "ip",
        "detail": "IP belongs to a hosting provider."
      },
      {
        "rule": "geo_mismatch",
        "signal": "phone_ip",
        "detail": "Phone country is US, IP country is NL."
      }
    ],
    "signals": {
      "email": {
        "address": "jordan.avery@example.com",
        "valid": true,
        "disposable": false,
        "role_account": false
      },
      "phone": {
        "phone_number": "+15125550142",
        "valid": true,
        "line_type": "voip",
        "country": "US"
      },
      "ip": {
        "address": "203.0.113.42",
        "country": "NL",
        "vpn": false,
        "proxy": false,
        "tor": false,
        "hosting": true
      }
    },
    "inputs_checked": [
      "email",
      "phone",
      "ip"
    ]
  },
  "meta": {
    "credits_used": 3,
    "credits_remaining": 999,
    "cached": false,
    "request_id": "req_8f3a2c1e9b7d"
  }
}
[01/Response]

What you get back

The fields most teams branch on. The full response is in the sample on this page and in the API reference.

scoreFraud score0 to 100. Higher means more rules fired, or weightier ones did.
levelRisk levellow, medium or high: a banded reading of the score for simple branching.
rules_firedRules firedEach rule that contributed, with the signal it came from and a plain-language detail.
signals.emailEmail signalsValidity, disposable-domain and role-account findings for the email you sent.
signals.phone.line_typePhone line typemobile, landline or voip. VoIP numbers are a common trait of throwaway accounts.
signals.ip.hostingDatacenter IPTrue when the address belongs to a hosting provider instead of a consumer network.
inputs_checkedInputs checkedWhich of email, phone and IP were present and evaluated in this check.
[02/Capabilities]

What Fraud Detection does

01

Rules, not a black box

The score comes from named rules, and the response lists every one that fired. When a customer asks why they were flagged, you can point to the rule instead of guessing at a model's reasoning.

02

Any mix of inputs

Email, phone and IP are all optional. Send one, two or all three, and the check runs the rules that apply to what you provided. More inputs give the cross-signal rules something to compare.

03

Cross-signal checks

Some rules look at how the inputs relate to each other. A phone number from one country arriving with an IP address in another fires a geo mismatch rule that neither lookup would raise alone.

04

The underlying signals included

Next to the score you get the findings for each input: disposable email, VoIP line type, datacenter IP and more. Log them, and you can tune your own thresholds later without re-running anything.

[03/Try it]

See the response before you sign up

Fill in the request and run it. The dashboard has the same form, plus history, bulk upload and live data.

POST /v1/fraud-detection3 credits / check

IPv4 or IPv6

Preview mode: this page returns a sample response so you can see the shape of the data. Create a key to run it on live data.

Response

Fill in the field and run the lookup to see what comes back.

[04/Integrate]

One POST request

Send JSON with a Bearer key. The same key works for all 79 products.

curl -X POST https://api.1lookup.dev/v1/fraud-detection \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"email":"jane.doe@example.com","phone_number":"+15125550142","ip":"203.0.113.42"}'
Endpoint
POST /v1/fraud-detection
Base URL
https://api.1lookup.dev
Request
email?phone_number?ip?
Price
3 credits / check
Coverage
Global
Bulk
CSV upload and bulk API
Cache
Repeats within 7 days are free
Also via
Dashboard and MCP server (run_lookup)
[05/Use cases]

What teams use it for

Score orders before fulfillment

Run the billing email, phone and checkout IP through one check. Ship low scores automatically and hold high ones for a manual look.

Filter lead form submissions

Score each inbound lead and keep disposable emails and datacenter traffic out of the sales queue.

Review promo and referral abuse

Check accounts that claim a signup bonus. Clusters of VoIP numbers and hosting IPs tend to show up as the same set of fired rules, over and over.

Back-test a threshold

Upload historical records as a CSV, then compare scores and fired rules with the outcomes you already know before you decide where to block.

The score is a risk signal, not proof of fraud. Legitimate users on VPNs or VoIP numbers can score high, so keep a review path for borderline results.

[06/Cost]

3 credits / check

Credits come with your plan and work on every product. Requests that fail on our side are never charged. Beyond your plan, extra credits are pay as you go.

See pricing
PlanPer 1,000 creditsPer 1,000 checks
Starter $29/mo$4.83$14.50
Growth $99/mo$3.96$11.88
Pro $299/mo$3.32$9.97
Scale $799/mo$2.66$7.99
[07/FAQ]

Fraud Detection questions

How is the fraud score calculated?

It is rule-based. Each input is tested against rules such as disposable email, VoIP number, datacenter IP and a mismatch between phone country and IP country, and the rules that fire set the 0 to 100 score. Every fired rule is named in the response.

Do I have to send email, phone and IP together?

No. All three fields are optional and any combination works, as long as you send at least one. Cross-signal rules such as geo mismatch only run when both of the inputs they compare are present.

What does a fraud check cost?

3 credits per check, whether you send one input or all three. An identical request repeated within 7 days comes from cache and is not charged.

What fraud score should I block at?

That depends on your tolerance for false positives. Start by logging scores without acting on them, compare them with known good and bad outcomes, then choose your thresholds for review and block. The level field gives you a default banding to begin with.

What can Fraud Detection not see?

It evaluates the contact details and IP you send. It does not see payment data, devices or behavior on your site, and a careful fraudster with a clean email, a mobile number and a residential IP will score low. Use it as one layer of several.

[ Validate ][ Enrich ]
[Get started]

Run Fraud Detection on your own data

Starter is $29 a month for 6,000 credits, enough for 2,000 checks a month.

[ Screen ][ Scrape ]