Maskbreak vs SEON

Weighing more than two tools? Our guide to SEON alternatives covers the wider field, and our Arkose Labs vs SEON post compares SEON’s signals with a challenge-based approach.

SEON pricing in September 2026

SEON publishes one price. Starter is $699 a month for 2,500 fraud checks, 10 users and 50 custom rules, with platform and API access, implementation help by ticket and email, and basic monitoring and standard reporting. Everything else sits in Premium, listed as “Custom pricing for your needs”: unlimited API calls, users and rules, plus Case Management and AML Compliance, a dedicated implementation team, 24/7 support and Managed Risk Services. Both plans’ buttons read “Speak with an expert”, so there is no self-serve checkout.

Do the arithmetic before the demo. 2,500 checks a month is about 83 a day, and $699 across 2,500 checks comes to roughly $0.28 each. That rate is our division, not a SEON price: SEON publishes no per-check rate, no overage fee and no annual price, so ask what happens in the month a bot run doubles your volume. The units are not like for like, either. A SEON fraud check can return email, phone, IP and device enrichment in one response; a Maskbreak check covers the visit and nothing else.

Trying SEON as a developer: SEON’s API docs limit trial accounts to 2 requests a second (20 after the trial), and the Email API to 120 a minute during the trial, which is enough to wire up a staging form. Ask SEON how many API checks the trial includes; the trial page lists only the 500 manual ones.

What is SEON?

SEON Technologies Ltd. was founded in 2017 by Tamás Kádár (CEO) and Bence Jendruszák (COO), who, by SEON’s own account, started it after fraud attacks threatened a cryptocurrency startup they were building. It began in Budapest. Its press releases are now datelined Austin, Texas, and it lists offices in Austin, London, Budapest, Singapore and Jakarta. It has raised $187M, most recently an $80M Series C on 16 September 2025 led by Sixth Street Growth, and it reported more than 80% ARR growth in 2025. SEON says it works with more than 5,000 businesses; customers it names include Revolut, Plaid, Nubank and Entain. The industries it lists are fintech, payments, BNPL and lending, betting and gaming, financial services, digital banks, marketplaces and retail.

If you last looked at SEON a few years ago, you probably remember an email and social-profile lookup. That picture is out of date. Its products now include Digital Footprint and Device Intelligence, identity verification (documents, liveness, proof of address), AML screening and transaction monitoring, a Scoring Engine that mixes rules with machine learning, and a Case Management queue that fraud and AML teams share. The AML side grew out of Complytron, a Hungarian AML API provider SEON bought in early 2023, as tech.eu reported. Developers can call the unified Fraud API, or the email, phone, IP, BIN and AML APIs on their own.

How SEON works, and how Maskbreak differs

The short version: SEON asks who the person is and how they arrive; Maskbreak asks only how they arrive. SEON starts from identifiers. Hand its Email API an address and it returns the domain’s age and registrar, whether it is disposable, known breaches, SEON’s consortium fraud history for it, and how many platforms it is registered on, grouped into categories such as social media and AI platforms. The Phone API does the same for a number and adds the carrier, SIM-swap and porting history. An address created last week for a bonus claim usually has a thin footprint, and that thinness is the signal.

SEON checks the visit too. Its JavaScript agent returns an encrypted session payload from seon.getSession() that your backend passes to the Fraud API; the iOS and Android SDKs work the same way. SEON documents residential and mobile proxy detection from browser data plus TCP/IP and TLS fingerprints, and writes that reputation-based IP checking is “not adequate for the detection of residential or mobile proxies”. All of it feeds the Scoring Engine: default rules SEON pre-configures (you can switch them on or off and change their weight, but not their parameters), your own custom rules, and an ML score from 0 to 100. The answer is a state of APPROVE, REVIEW or DECLINE, a fraud_score, and applied_rules, which lists each rule that fired and what it added.

Maskbreak starts and stops at the visit. It does not look up who the person is. A browser snippet collects two tokens when the form is submitted; your server sends them to POST /v1/evaluate with its API key and gets one decision back from a default policy, with the reasons. Here is a fake browser (an antidetect browser that spoofs its fingerprint) behind a residential proxy, trying to open an account. Example response, selected fields:

{
  "decision": "block",
  "risk_score": 93,
  "reasons": ["proxy_detected", "antidetect_browser"],
  "network": {
    "vpn": false,
    "proxy": true,
    "datacenter": false,
    "tor": false,
    "service": "<proxy service, when known>"
  },
  "device": {
    "antidetect": true,
    "automation": false,
    "emulator": false,
    "tampering_score": 0.8
  }
}

Both reasons block under the default policy. A VPN on its own comes back as review with vpn_detected, and the service named when known, because plenty of honest people use one. A cloud-server address alone stays allow, with datacenter_asn still listed. The default policy applies from the first check; when you want a different action, the console’s Rules tab sets one per signal, from VPN and disposable email to multi-account devices, plus one for a VPN and a fake browser seen together. Every response reports its own processing time in evaluated_in_ms. The quickstart begins in watch mode: log each decision, let everyone through, and switch to enforce once the Events log looks right.

Where each one earns its place

Signup with a fresh identity. A bonus hunter opens account number twelve with a new webmail address and a prepaid SIM. SEON’s email and phone checks are built for this moment: a mailbox registered on no other platform, a number with no history behind it. Maskbreak sees none of that. Its identity side is deliberately thin: send the email with the check and a disposable domain moves allow to review, and that is all. What it does see is the visit: the residential proxy, the fake browser and, if you pass an accountId, the same device already tied to other accounts on your site (multi_account_device, see multi-accounting). The iGaming bonus-abuse playbook works that attack through from the network side.

Login under account takeover. Here the email and phone belong to the real customer, so their footprint looks healthy. It is the victim’s history, not the attacker’s. What gives the attacker away is the visit: a residential proxy, a fake browser, automation replaying stolen passwords. Both vendors check that part, SEON through its device and IP modules and Maskbreak with its whole product. If account takeover is your main problem, you can test the visit evidence free before deciding whether you also need the identity half.

Checkout and money movement. SEON’s Fraud API can add BIN data to a payment check, and its AML products screen payments and monitor transactions (Premium). Maskbreak scores no orders, cards or transfers. At checkout it tells you whether the visit comes through a proxy, Tor or a fake browser, which helps against card testing, and leaves payment risk to your payment provider’s own tools.

Maskbreak vs SEON, feature by feature

When Maskbreak is a SEON alternative, and when it is not

SEON vs Sardine

Sardine turns up on the same shortlists, mostly at banks and fintechs. It calls itself an “Agentic Financial Crime Platform for Fraud Prevention & AML” and, like SEON, pairs device intelligence and behaviour biometrics with KYC and AML screening. Its range runs further into money movement: bank account validation, credit underwriting, payment and bank-transaction fraud, and card-issuing fraud. Both vendors name Nubank as a customer.

On price, SEON publishes a starting point and Sardine does not: sardine.ai/pricing returned a 404 when we checked on 24 September 2026. Sardine was founded in 2020 and raised a $70M Series C led by Activant Capital in February 2025, taking it to $145M; fintech.global reports a $25M extension in May 2026 that brought the total to $170M. If you are choosing between the two, compare them on the money-movement products you actually need and ask both for a quote at the same monthly volume. Maskbreak competes with neither on AML or KYC. It is a visit check a smaller team can run before it needs a financial-crime platform.