A loss is a signal that was watched too late.
Early Warning reads financial and behavioural signals continuously against the live portfolio, ranks what has actually changed, and turns it into an intervention someone owns — with a date, an action and an outcome recorded against the asset.
Quarterly review is an accounting rhythm, not a risk one.
Businesses deteriorate on their own schedule. By the time a periodic review notices, the options that were cheap — a restructure, a limit reduction, a conversation — have usually expired, and what is left is a provision.
Everything that moved, down to the one thing to do today.
Detecting a threshold breach is the easy part. The narrowing below is the hard part — and the two boxes at the bottom are what the process leaves behind, which is the part a securitisation investor eventually asks about.
What comes in
Utilisation, payment timing, cash-flow volatility, sector context and filing changes are tracked as a live set rather than a periodic snapshot.
Signals are scored against the exposure they sit on, so a small change on a large limit outranks a large change on a small one.
Assets move on and off watch under stated rules, with the reason and the reviewer recorded each time.
Every escalation produces a task, a deadline and an outcome written back to the credit record.
And two things the process leaves behind
Sector, counterparty and vintage concentration are visible against limits before they become a covenant conversation.
The signal history behind any decision is preserved — which is exactly what a securitisation investor and an auditor both ask for.
Signal to outcome, with nothing lost in between.
Behavioural, transactional and external feeds update against every live exposure.
Rules and thresholds fire where a pattern has genuinely changed, not merely moved.
Alerts are weighted by exposure and severity so a team of five can work a portfolio of thousands.
The escalation becomes an owned task with a deadline inside Credit Lifecycle.
The outcome returns to the asset and to the next model calibration.
The control layer over a live book.
It reads the record Credit Lifecycle maintains, uses the views Scoring Engines produce, and supplies the performance evidence a pool is later tranched against.
Point it at a portfolio you already hold.
Over an existing book
Early Warning can monitor a portfolio serviced elsewhere, through defined interfaces, without moving the servicing.
Your risk appetite
Signals, weights and escalation rules are configured by the risk team and versioned like any other policy.
Into your queues
Escalations can be raised into the institution's own task and case systems rather than a second inbox.
The ones that come up first.
How is this different from a rules engine we already have?
The difference is ranking and ownership. Detecting a threshold breach is the easy part; deciding which of two hundred breaches a small team should work today, and recording what was done, is the part that changes realised loss.
Does it need transactional data?
It works on repayment behaviour and financial updates alone, and gets substantially better with transactional feeds. Most deployments start with the former.
What does this have to do with securitisation?
A pool is tranched against expected loss. Evidence that deterioration was caught early and acted on is what lets a structure carry less enhancement for the same rating proxy.
Is this the same as FiveEye?
No. FiveEye monitors financial-crime risk — screening, AML, transaction monitoring. Early Warning monitors credit risk. They run on the same platform and answer to different regulators.
Bring the portfolio and the last four provisions.
The useful question is whether the signals were there before the loss was.