The Authenticity Score
Authenticity score / noun / — a measure of whether a user's claimed identity looks real. An authenticity score evaluates the profile data a user submits (email address, name, username, organization, and anything else they provide) for the telltale shortcuts of a fabricated identity: disposable email domains, keyboard-mashed names, generated-looking profiles, and emails that don't plausibly belong to the person they claim to be.
Dregs computes an Authenticity score from 0 to 100 for every user in your application, continuously, as one of its four identity scores. It's the dimension that powers fake account detection: fraudsters can rent clean devices and residential proxies, but they often fail to fake a convincing identity.
What the Authenticity Score Measures
Every signal feeding the Authenticity score answers some version of the same question: did a real person put real information into this account, or did someone (or something) fabricate just enough to get through the form?
How Dregs Computes the Authenticity Score
Dregs runs a pipeline of analyzers over each identity. Each analyzer inspects one signal and produces observations with a value from 0.0 (suspicious) to 1.0 (legitimate), a confidence, and a human-readable explanation. The Authenticity score is the weighted aggregate of every observation in the Authenticity category, scaled to 0–100. The observation list is preserved, so you can always see exactly why a user scored the way they did.
{
"category": "AUTHENTICITY",
"id": "authenticity.disposable-email",
"label": "Disposable Email",
"value": 0.0,
"confidence": 0.95,
"explanation": "Email domain is a known disposable email provider",
"metadata": { ... }
}Scores update as identities change. If a user edits their profile from a throwaway alias to a corporate address, their Authenticity score recovers on the next scoring pass; if a plausible-looking account later reveals fabricated details, it drops.
Fake Account Detection, Continuously
Point-in-time fake account detection (a check at signup, then nothing) misses everything that happens after the form. Dregs scores authenticity continuously, so an account that slipped through registration still surfaces when the evidence accumulates. And because the score is per-identity rather than per-signup-attempt, it composes with the other three dimensions: the classic fake-account signature is low Authenticity plus low Uniqueness (the same device keeps making accounts), often with low Humanity when the signups are scripted.
See it applied in the fake signups and free trial abuse use cases, or try the free disposable email checker, which runs the same email-risk dataset this score uses in production.
Authenticity Scoring vs. Email Verification
Email verification tools answer the deliverability question: can this inbox receive mail? The Authenticity score answers the intent question: does this identity look fabricated? A perfectly deliverable Mailinator inbox passes verification and fails authenticity; both checks have their place, but only one of them is fraud detection.
More Than an Email Risk Score
Plenty of tools stop at an email risk score: rate the address, return a number. That's a useful input, and email risk is one of the signals Dregs weighs. But an address is a single field. The Authenticity score evaluates the whole claimed identity, so a pristine email attached to a keyboard-mashed name still gets flagged, while a quirky address on an otherwise coherent profile doesn't sink an honest signup.
The Other Three Scores
Authenticity is one important score, but Dregs scores every identity's Humanity, Uniqueness, and Behavior as well.
Humanity
Is a real human controlling the browser, or automation?
Uniqueness
Is this the user's only account, or one of many?
Behavior
Does their usage look like a legitimate customer journey?
All four scores are computed by the same identity scoring pipeline, on every plan, from the events you're already sending.
Frequently Asked Questions
Q: What is an authenticity score?
A: An authenticity score is a continuous measure of whether a user's claimed identity looks genuine. In Dregs it's a 0–100 value computed from the identity data a user submits (email address, name, username, organization) evaluated for signals like disposable email domains, keyboard-mashed or generated-looking names, and whether the email plausibly corresponds to the stated name. A score near 100 means the profile data is consistent with a real person; a score near 0 means it was almost certainly fabricated.
Q: Is an authenticity score the same as a user trust score or identity trust score?
A: They're similar. User trust score and identity trust score are umbrella terms different vendors use for a per-user risk number. Dregs's Authenticity score is deliberately narrower: it measures one specific thing, whether the claimed identity data looks real, and leaves bot-ness, duplicate accounts, and usage patterns to the Humanity, Uniqueness, and Behavior scores. Keeping the dimensions separate is what makes each one actionable.
Q: How is authenticity scoring different from email verification?
A: Email verification services check deliverability: whether an inbox exists and can receive mail. Authenticity scoring checks intent: whether the identity behind the address looks fabricated. A real, deliverable inbox at a disposable provider passes verification and fails authenticity. A typo'd address at a corporate domain fails verification but may score fine on authenticity. Most teams that care about fraud want both.
Q: Can the Authenticity score do fake account detection on its own?
A: It's the single strongest signal for it, because fake accounts often cut corners on identity data: throwaway inboxes, mashed keys, mismatched names. But Dregs detects fake accounts with other scores too: a fabricated profile (low Authenticity) operated by automation (low Humanity) from a device already tied to other accounts (low Uniqueness) is a much more confident verdict than any single dimension. You can badge or alert on whichever combination fits your abuse pattern.
Q: What data does Dregs need to compute an Authenticity score?
A: Whatever identity data your application already collects, typically email plus whatever profile fields you have (name, username, organization). Dregs maps your field names to canonical concepts, so nombre or FirstName both work. There's no minimum: an email alone produces a useful score, and additional fields sharpen it.
Q: Does a low Authenticity score always mean fraud?
A: Not always. Some legitimate users are just privacy-conscious and sign up with alias addresses or minimal profiles. That's exactly why Dregs reports Authenticity as one of four scores instead of a single verdict, and why every score comes with the underlying observations explaining what was found. Treat a low score as a signal to look closer or to gate risky features, not as an automatic ban.
Authenticity scoring for every signup, automatically.
Drop the Dregs tracking script into your application and every identity gets a continuous Authenticity score (plus Humanity, Uniqueness, and Behavior) from the very first event.
Schedule a Demo