Skip to main content
User base audit

One face, two payouts.This week's audit catches both accounts.

Your existing accounts never got a face check. Run one now, no ID, and the ones sharing a face come back named.

The problem

The duplicates already in your user base

None of these three accounts failed a check at signup. There wasn't one yet.

  • acct_1042
    Joined before the check
    Face on fileNone
    ActiveYes

    The account you already approved

    A panelist or annotator who joined before you had a check is still active, and nothing in your records says they're also someone else.

  • Owner
    Now
    Different person

    The login that changed hands

    A vetted account gets sold or rented out, and the work keeps coming in under a name that isn't doing it anymore.

  • Payout, acct_2201Payout, acct_4490Same face

    The same face, two payouts

    One person collects under two accounts because nothing ties either one to the other until someone compares them by hand.

How it works

Find the duplicates, then decide

  1. 01

    Passive signals score each account on its next visit

    Device and network signals score each account as it comes back: VPN, data center and Tor traffic, and IPs already on an abuse list, without asking anyone anything.

  2. 02

    Flagged accounts take an anonymous face check

    For the accounts the signals flag, or a list you choose, an emailed invite asks for a 3 second selfie with no ID. Whoever verified first keeps that face, and a later account with it comes back named as a match.

  3. 03

    Duplicates come back named, ready to act on

    Each flag names the account it matches, often enough to act on without opening anything, and the export lines up with your own account IDs.

Risk signals
IPIP addressResidential
TYTypingSteady
NWNetworkData center
acct_1042
VPNNo
Data center IPYes
On abuse listNo

A solution for every scenario

Let's talk trust

Start with a free pilot: see what the check catches on your own users, with no dev work to begin.

With tools like VerifyYou, AI becomes a force for good. It helps prevent AI-driven fraud and can be done with much lower friction than what we built over the years.
Wes MichaelFounder of Rare Patient Voice

Built by humans from

  • Quora
  • Pinterest
  • Square
  • Reddit
  • Gusto
  • Cohere Health
  • Google
  • Accenture
Security

Privacy and security in every decision

We keep a face code, not the photo, and hold it under SOC 2 Type II and ISO 27001:2022 compliant controls.

  • No images stored. VerifyYou does not store images.

  • No ID or documents. HumanCheck anonymously confirms there is a real, live, unique person, with no name or ID.

  • A face code, not a photo. We treat it as sensitive biometric data.

  • AICPA SOC 2SOC 2 Type II
  • Intercert ISO/IEC 27001 certificationISO 27001:2022
  • iBetaISO/IEC 30107-3 PAD Level 1 & 2
  • GDPRGDPR compliant
Visit our Trust Center

Common questions

Won't this lock out paying customers?

Not if you set it to Allow and flag, which is what an audit wants. Everyone passes, the duplicates are recorded, and you decide in your own system rather than at the door.

How do results tie back to our accounts?

Without code, the invite goes to the email on each account, so the face ties to that account. With code, pass your own user ID and duplicates are reported against it.

When should we run it?

On the clock you already work to. If payouts go out on a Tuesday, an audit that lands on Wednesday cost you the week. Email invites are in beta with a daily sending cap, so send a first batch early.

What if someone never opens the invite?

Treat it as a reason to follow up rather than a pass. An audit only measures the people who take the check, so an account that never responds stays unverified in your own system, same as before the audit ran.

Can it find an account that shares a login, not just a duplicate signup?

Yes, over time. If two different people take turns behind one login, each selfie ties to a different face, and the mismatch shows up the next time you run the check.

Do you store photos of our users?

No. We keep a face code, not your users' images.

How much work is it to add?

It's two API calls: one to start the check and one to read the result.

Do we have to replace our KYC or bot tools?

No. It sits on top of your stack, so you keep your KYC and bot tools.

What does it cost?

It's 50 to 100 times cheaper than KYC. You pay per check, and nobody uploads an ID document.

Ready to get started?

VerifyYou scales trust: passive signals on every session, a face check only when risk calls for it.

Get startedTry it
The VerifyYou dashboard on a laptop, showing verification volume and threat signalsAn anonymous HumanCheck running on a phone

We use cookies. Privacy