A-book
Operating model where client flow is hedged or passed to liquidity venues rather than internalized as principal risk.
Shared language for brokerage systems discussions.
Operating model where client flow is hedged or passed to liquidity venues rather than internalized as principal risk.
Operating model where the brokerage takes the other side of client flow as principal, increasing risk management requirements.
Internal systems for account administration, permissions, operational exceptions, and control workflows.
Connectivity layer that can route or transform order/market data between platforms and liquidity sources.
Contract for difference — a derivative product commonly offered by multi-asset brokerages.
Process from application through required checks to an eligible funded trading state.
Capability allowing clients to mirror selected strategy providers under defined risk constraints.
Operational team intervening in pricing, hedging, or trade handling according to the firm’s model.
Partner who refers clients under commercial terms tracked in partner systems.
Know Your Customer — identity verification processes often integrated via specialized providers.
Anti-Money Laundering controls and screening processes adjacent to onboarding and monitoring.
Venue or counterparty supplying pricing and/or execution pathways.
Allocated multi-account management structures with precise operational definitions required before software commitments.
Straight-through processing patterns that reduce manual trade handling between client and venue.
Client portal for onboarding status, funding actions, and account controls.
Delivery model emphasizing coordinated stack assembly toward operational readiness.
Brand-led deployment on a platform with explicit customization boundaries.
Customer lifecycle system tailored to brokerage conversion, retention, and partner context.
Consolidated view of exposures used by risk and dealing teams.
Layer coordinating deposit/withdrawal providers, statuses, and reconciliation handoffs.
Agreed interface, data fields, error handling, and ownership between systems.
Intensified support window immediately after go-live.
User acceptance testing against agreed business scenarios before production.
Controlled switch from legacy to new systems during a migration.
People, permissions, runbooks, and monitoring prepared for production use.
Testable conditions that define whether a delivered capability is done.