Kavra LabMulti-accounting in plain terms
Most online services assume one account equals one customer. Their welcome offers, free trials, betting limits, votes and bans all depend on it. Multi-accounting breaks that assumption: one actor appears as many customers, so everything the rules give "once per person" can be taken again and again.
Not every second account is abuse. A household sharing a laptop, a customer with a personal and a work account, or someone who forgot their old login are normal. It becomes multi-accounting in the fraud sense when the extra accounts exist to take value or dodge rules, and when the actor hides that the accounts are connected. Intent and concealment, not the number of accounts, are what separate a household from a fraud ring.
Why multi-accounting matters
The obvious cost is the value handed out many times: bonuses, credits, discounts and referral payouts that were budgeted for new customers but went to one actor. The quieter cost is bad data. Signup, activation and retention numbers look healthier than they are, so marketing keeps funding offers that mainly feed abusers.
Multi-accounting also undermines safety tools. Bans stop working when a banned user returns under a new name, self-exclusion in gambling fails when a player opens a fresh account, and limits on tickets or bets mean little when one person holds twenty accounts. That is why the check belongs at signup and again before value leaves, not only in a monthly report.
Common forms of multi-accounting
| Form | Where you see it |
|---|---|
| Bonus abuse | Welcome bonuses, free bets, referral rewards |
| Gnoming | Sportsbooks and casinos, accounts held in other people's names |
| Free-trial and credit farming | SaaS and AI products |
| Ban evasion | Marketplaces, communities, games |
| Limit bypassing | Ticket caps, betting limits, purchase limits on drops |
| Vote and review manipulation | Rankings, contests, marketplaces |
How it is done and how it is caught
At small scale, a person uses a second email and a different browser. At scale, operators run dozens or hundreds of accounts from one machine with an antidetect browser, a different residential proxy for each account, and scripts or emulators to create and farm them. Each account looks like a separate person on a separate device. Some operators skip the tech and pay real people for their identities instead.
What gives them away is what they share underneath: the real device behind the spoofed profiles, the proxy network, identical behavior and shared payout details. Kavra links accounts on that evidence, treats a rotating fingerprint on the same actor as one actor with many rotations, and flags the cluster before the reward is paid. The full playbook, with warning signs, industry examples and a prevention checklist, is on the multi-accounting detection page.