Duplicate Account Detection

One person, five accounts? They're probably gaming your referral program, manipulating votes, stacking free resources, or evading a ban you already handed out. Multiple account abuse is one of the hardest abuse patterns to catch because each account looks perfectly normal on its own.

Dregs uses a pipeline of AI-assisted analyzers to connect the dots across duplicate accounts automatically.

The Multiple Account Problem

Duplicate accounts in all sorts of applications. Referral programs get gamed when the same person refers themselves. Community features lose credibility when one person controls multiple votes or reviews. Leaderboards become meaningless when the top spots are all the same player. Per-user resource limits get bypassed by simply creating more users.

The core challenge is that each individual account looks legitimate. The abuser uses a real email address, fills in a plausible name, and behaves like a normal user on each account. The problem only becomes visible when you zoom out and look across accounts — and most systems don't do that.

The conventional defenses like bot prevention aren't equipped for this.

Multiple account abusers are typically real people making real use of your product. As such, they pass all the normal checks designed to stop bots and throwaway signups. The only reliable way to catch them is by finding patterns and recognizing the person behind the accounts.

What Duplicate Accounts Cost Your Business

Multiple account creation doesn't just waste resources — it actively degrades the experience for your legitimate users and undermines the features you've built.

How Dregs Detects Duplicate Accounts

Dregs doesn't rely on a single signal. It combines device fingerprinting, cross-account relationship discovery, identity analysis, and behavioral patterns to catch multi-accounters from multiple angles at once.

Uniqueness Score

Device fingerprint sharing is the single strongest signal for duplicate accounts. When two accounts use the same device, the Uniqueness score for each drops sharply. Dregs does this automatically — no cookies required, just persistent hardware and browser characteristics that survive incognito mode, cache clearing, and new logins.

Identity Relationships

Dregs automatically discovers relationships between identities that share devices, IP addresses, sessions, or behavioral patterns. When someone creates their second account, Dregs links it to the first one instantly. By the time they create a fifth account, you have a complete map of the entire cluster, and the user's scores drop accordingly.

Authenticity Score

People are creatures of habit, even when they're trying to be sneaky. The Authenticity score detects similar name patterns, email structures, and identity data across accounts. Multiple accounts with similar name and email conventions? That's a pattern Dregs can pick up on... not to mention randomized or keyboard-mashed entries.

Behavior Score

Duplicate accounts controlled by the same person tend to behave the same way — the same active hours, similar navigation patterns, and the same features used in the same order. The Behavior score looks at these overlapping patterns even when the accounts use different devices, filling in gaps where fingerprinting alone falls short.

Example: Catching a Multi-Accounter

Here's an example of how Dregs could help you catch an abuser with multiple accounts in your application:

No manual investigation needed... the duplicate is caught and linked within seconds of signup.

Responding to Duplicate Accounts

Detection is only half the story. How you respond depends on your product and what the multi-accounter is doing. Dregs gives you the data to automate whichever approach fits.

With Dregs webhooks, any of these responses can be fully automated. Your application receives scores, badges, and relationship data in real time and acts on them without human intervention — whether it's midnight, a holiday, or the middle of a product launch.