Key takeaways

  • Three modules: Client and IB Screener, Dealing Report, Infrastructure and Identity.
  • Every pattern is re-tested across linked logins, devices and the whole IB chain.
  • Runs on your existing MT4/MT5 stack — no platform migration.
  • White-labelled on your domain with tenant-isolated data.

The problem it addresses

Your least profitable clients are usually not the ones losing money. They are the ones extracting value with minimal risk exposed, and in every report you currently run they look like ordinary volume.

A-FBI exists because that pattern is invisible at the account level and obvious at the entity level. The work is joining the data well enough to see entities.

Module 1: Client and IB Screener

Point the screener at an email address. It returns every trading login behind that client, every account sharing its device signature, the full IB chain above and below, and a pass or fail verdict against a library of twelve red flags.

Crucially it shows the arithmetic. A verdict without the numbers behind it is not usable in a dispute, and disputes are where this data earns its keep.

  • All logins linked to one client identity
  • Accounts sharing a device signature
  • Full upstream and downstream IB chain
  • Pass/fail across twelve red flags with visible arithmetic
  • Net position of the client and of the chain

Module 2: Dealing Report

The Dealing Report answers where the book is bleeding, now. Realised and floating client PnL by hour and by day, stop-out compensation paid per hour, low-liquidity symbol performance with the winning accounts named, and top clients by equity and live exposure.

It is designed to be read by a dealing desk rather than a data team — the output is the finding, not a dataset to query.

Module 3: Infrastructure and Identity

This layer turns apparently unrelated clients into single entities. It detects VPS and hosting IPs, VPN and data-centre ranges, device signature reuse across different email addresses, Expert Advisor fingerprints, geolocation against registered country, and reused KYC documents.

Without it the other two modules produce lists. With it they produce entities, which is what you can act on.

What it catches

The pattern library covers the abuse types that recur across retail brokerages.

  • Internal hedging across linked accounts
  • Latency and HFT abuse
  • IB rebate farming and commission loopholes
  • Bonus and promotion abuse
  • Low-liquidity and thin-window trading
  • Copy-trading loops
  • Funding and payout abuse, measured as an ABC ratio
  • Coordinated multi-account identities

How it runs

Four stages. Ingest from MT4/MT5, CRM, IB panel, bonus engine, PSP and session logs. Match on email, name, country, IP, device signature, payment method and KYC documents. Score across trading, funding and geolocation risk into a weighted damage index. Act — alert, freeze, hold a payout, void trades, flag a device or an IP.

Every pattern is re-run across the client's other logins, shared device signatures and the whole IB network before a verdict is issued. Detecting one account is not detection.

Deployment

A-FBI is a data-integration project, not a migration. It runs against the platform you already operate, white-labelled on your domain, with tenant-isolated data, configurable thresholds and exportable reports.

Typical deployment is days. The usual first engagement is a screen of your existing book against a defined period.

Want this mapped to your brokerage?

Book a call →

Frequently asked questions

Does A-FBI require the EnterTrade platform?

No. It runs against existing MT4/MT5, CRM, IB and payment data.

Can thresholds be tuned to our risk appetite?

Yes. Thresholds and the weighted damage index are configurable per tenant.

What is the ABC ratio?

A funding-abuse measure comparing deposit, trading and payout behaviour to surface accounts cycling funds rather than trading them.