---
agentTools:
  projectIndex: https://docs.defastra.com/llms.txt
---

# Fight fraud by going deeper

An introduction to our fraud prevention tools ***Deep Email Check*** and ***Deep Phone Check***

[block:callout]
{
  "type": "info",
  "body": "**This article is an overview of how our fraud detection works.**\n\nFor a more technical approach to *why Defastra performs better than other anti-fraud solutions*, we encourage you to read our various [technical papers about disposable phone numbers & temporary emails](https://defastra.com/reports).",
  "title": ""
}
[/block]

[block:api-header]
{
  "title": "Overview"
}
[/block]

Our approach to fraud prevention is **simple** but delves much deeper than other solutions, providing real-time, accurate, and detailed results.

Fighting fraud with us is transparent and straightforward:

1. Simply submit a single phone number or email address using our [manual tools](https://docs.defastra.com/docs/perform-manual-fraud-checks), [batch processing](https://docs.defastra.com/docs/what-is-a-batch) or [API](https://docs.defastra.com/reference/authentication)
2. We gather real-time social & digital data about the email or phone submitted, also called data enrichment
3. Our service runs a range of deep anti-fraud checks (click to see precisely what check we perform [against phone numbers](https://docs.defastra.com/docs/deep-phone-check-signals) and [against email addresses](https://docs.defastra.com/docs/deep-email-check-signals)
4. We return you all the data we gathered (from social media, data breaches, etc.) along with a `risk score` & `risk level`, and an advanced yet easy-to-understand fraud report, including what risk or trust signals were discovered

In the next few sections of this article, we will go into detail about:

* How to interpret the risk level & risk score we generated
* What risk & trust signals are uncovered
* And what raw data will you also get

[block:api-header]
{
  "title": "Risk level & risk score"
}
[/block]

Each deep check will return a risk score and a risk level.

* The risk score is a number between **0** *(safe)* and **100** *(fraudulent)*.
* The risk level is our appreciation of the risk score.

There are 4 risk levels. Below you can see rough definitions of these risk levels, and our recommendations for each:

[block:parameters]
{
  "data": {
    "h-0": "Risk level",
    "0-0": "<span style=\"color:green\">low</span>",
    "h-1": "Description",
    "h-2": "Recommendation",
    "1-0": "<span style=\"color:orange\">medium</span>",
    "2-0": "<span style=\"color:red\">high</span >",
    "3-0": "<span style=\"color:red;font-weight:bold\">extreme</span>",
    "0-1": "No risk signals were uncovered and the phone or email appears to be safe.",
    "0-2": "Allow",
    "1-1": "Some light risk signals were triggered.",
    "1-2": "Allow but flag",
    "2-2": "Allow and flag or **block** ",
    "3-2": "**Block** ",
    "2-1": "Serious risk signals were triggered, and we highly recommend security-sensitive businesses to **block**.",
    "3-1": "The phone or email is fraudulent and triggered many definitive risk signals, always **block**.",
    "h-3": "Risk score range",
    "0-3": "0",
    "1-3": "1 - 20",
    "2-3": "21 - 40",
    "3-3": "40 - 100"
  },
  "cols": 4,
  "rows": 4
}
[/block]

[block:api-header]
{
  "title": "Risk & trust signals explanation"
}
[/block]

The strength of our ***Deep Check*** tools lies in our carefully crafted fraud checks backed by AI, ratio surveillance, and years of analyzing fraud schemes.

Whenever you use our products, a list of checks will be run against your submitted element to find trust & risk signals, and uncover synthetical identities.

Our solution is **fully transparent**: each request will detail how the risk score was generated, along with all the signals we uncovered.

**A signal can be:**

* <strong style="color:red">Risk</strong>: we uncovered a risky pattern
* <strong style="color:green">Trust</strong>: we uncovered a trustworthy pattern

Each signal will also contain a `detail` property, giving you easy-to-read details about what we uncovered.

Here is an example of how a ***Deep Email Check*** result is displayed in our dashboard:

[block:image]
{
  "images": [
    {
      "image": [
        "https://files.readme.io/92b3591-de1.png",
        "de1.png",
        2181,
        1926,
        "#000000"
      ],
      "caption": "Example of ***Deep Email Check*** result",
      "border": true
    }
  ]
}
[/block]

**To deep-dive into what makes our Deep Check tools so efficient in uncovering fraudulent patterns and what powers them, check out our detailed Fraud checks list**:

* [***Deep Email Check*** signals](https://docs.defastra.com/docs/deep-email-check-signals)
* [***Deep Phone Check*** signals](https://docs.defastra.com/docs/deep-phone-check-signals)

[block:callout]
{
  "type": "info",
  "title": "Note for API users",
  "body": "You can head to the API reference of the [***Deep Phone Check***](https://docs.defastra.com/reference/deep-phone-check) or [***Deep Email Check***](https://docs.defastra.com/reference/deep-email-check) products and click on the **example response result** to see the ```JSON``` representation of a typical response."
}
[/block]

[block:callout]
{
  "type": "success",
  "title": "Deep, yet simple",
  "body": "Each fraud check we perform drills deeply into the data associated with the phone number or email address provided, but delivers a simple and easily readable sentence that clearly says what happened."
}
[/block]

[block:api-header]
{
  "title": "Data enrichment"
}
[/block]

In addition to our risk analysis, we will also return exclusive social and digital data about the phone or email checked.

To find out what data we return, check our \*\*[Up-to-date data list](https://docs.defastra.com/docs/full-data-list) \*\*.