
Track overview
About this track
Banking's first seven courses build the bank as a balance sheet and a set of businesses: credit prices the assets, funding the liabilities, capital the cushion under held risk, distribution the estate where the bank meets its customers, markets the trading book, payments the flow of funds, and advice other people's plans. Every one of those decisions ends in a price: a rate, a spread, a threshold, a payout shape, and the answer is money. This course is about the part of the bank that cannot be priced. The penalty is imposed by a third party, it is discontinuous, it is levied on your process as much as on your results, and the outcome you are actually trying to produce, criminal money stopped, is not observable to you at all. Every number available to the function is a proxy, and every proxy is gameable in the direction that makes the real problem worse: the average risk rating in the board pack, the high-risk volume the regulator asks for, the alerts cleared, the reports filed, and even the honest measure of what you missed inside your own book all improve when the bank simply banks fewer people. The learner sits in a seat no other Banking course occupies. Every other one owns a P&L line; this one owns the licence to operate, and answers to a supervisor who is never in the room and who judges the pattern of what was accepted rather than an outcome anybody can prove. The course trains three things that do not come naturally. A risk rating is a judgement made with the information available, so where visibility is thin the rating is not merely uncertain: it is biased downward, because the facts that would have raised it are exactly the facts that could not be obtained. Refusing business relocates flow rather than removing it: most of what a bank exits returns through the parts of its book it can see least well, and the rest leaves the regulated system entirely, so measured risk falls while actual risk rises and no dashboard shows it. And an appetite is not a filter you tune: it is a frontier you draw and must be able to defend, stated as a rule you could apply to a customer you have not met. The throughline: draw the line against what you can see rather than what you are rated, let the standard rise with the risk, and price the fact that exiting is not the same as stopping.
How each module works
Every module in this course runs the same three steps, from the first lesson to the debrief. The loop is what makes Decyde a place to practise, not just to read.
Step 1
Read
The lessons. Read the situation, the evidence and the shape of the trade-off before you commit.
Step 2
Size
The simulation. Put numbers on both sides of the trade-off, then make the call against the engine.
Step 3
Defend
The debrief and report. See what your call bought and why, so you can defend it or change it.
The same loop in every module, and a new decision each time. Judgment compounds.
Modules · In order
4 modules in this track
Each module is one decision: short readings, the decision itself, then a debrief and a replay in a shifted world.
01/04
The Appetite Line
Keystone Bank's board paper is four pages long and every number in it is true: average risk rating down, high-risk relationships down eleven per cent, suspicious activity reports up, alerts cleared inside the service level. It cannot answer the only question that matters, is less criminal money moving through this bank than last year, because that number does not exist anywhere in the building and never will. If you could observe it you would have stopped it. This is a seat no other Banking course occupies. Every other one owns a P&L line; this one owns the licence to operate, the penalty is set by somebody else and lands on your process as much as your results, and there is no rate that compensates you for a consent order. So the question is not what you charge but what you will take at all. The book, $19.45bn of annual payment volume, is laid out on a 5 x 5 map: inherent financial-crime risk across, and visibility, how much of what the customer actually does you can see, up. You set an appetite as a five-segment staircase across it, and it cannot go backwards: a line demanding less visibility of riskier business is not an appetite, it is a list of exceptions. Three questions run against the instincts the industry trains. What does a rating mean where you cannot see? What happens to the flow you refuse? And which of the numbers this function reports can you trust, when every one of them improves as the bank banks fewer people? Then a peer is fined, nothing about the book changes, and the whole line has to be re-derived. This opens Financial Crime & Compliance.
12 lessons · 2 mini-sims · ~30 min
Begin module02/04
The Peer Group
A money-service business on Keystone's book moves $1.45m in an entirely ordinary month. In one peer group that month is invisible and the system would find laundering inside it about nine times in a hundred; in another the same month, unchanged in every particular, is detected once in every eight hundred. Nothing about the customer moved. What moved is who they were being compared to, and telling the system who those people are is a field on a customer record that, in almost every bank, was populated once from the risk-rating field by somebody who thought of it as plumbing. It is most of the decision. Seven customer types have to be arranged into peer groups, which is 877 possible arrangements, and three forces decide what a band ends up being: how alike the group is, how much customers vary anyway, and how confident you can be about where the centre even sits, a term that is inversely proportional to the number of customers, and the reason a group of one has no normal. Two people in the room are both right and want opposite things: Model Risk wants fewer groups because every group is a model with an owner, a document and an annual review, and the head of the alert team wants narrower bands because every alert is a real person spending a real hour. The trap underneath is that a bank holds two orderings of its customers, by how risky they are rated and by how much money they actually move, and only one of them says what a monitoring band should be built on. Every number a monitoring function reports about itself improves as its bands widen, so which dial can you trust? Then a commercial win in an unrelated segment moves one customer type's behaviour and size together: does the obvious repair help, or hurt?
12 lessons · 2 mini-sims · ~30 min
Begin module03/04
The Names in the Report
Ridgeway & Co is a four-partner accountancy practice, the smallest entity on this file, and it is not involved in anything: six weeks of investigation established that. If you name it in this report, its name enters an intelligence file about somebody else's crime, held for five years, read by people at an agency it has never heard of, and it is never told, so it cannot correct it or contest it or know to explain the facility that quietly does not get renewed eighteen months later. That is what naming is: an action with a cost, and the cost lands on somebody who is not in the room. The report on Meridian Trading Co is going in today and that is not the decision. The decision is who else is named, across twelve entities and seventeen links, of which five entities did nothing at all. Legal wants everything in, because omission is the only failure anybody has ever been sanctioned for and they have watched a career end over a name left out, and they are describing the incentives accurately. But the unit receiving the report is not too busy to read it; it is unwilling to act on a diluted one, because acting means six weeks establishing which parts matter. That is a standard rather than a throughput, and it makes WIDTH the failure mode: the report that names everybody can look best on every count a filer tracks. Some cuts cost nothing, and what is left after them is the real decision, where a name may be the only route to part of the network and striking it costs flow. Then a licence is suspended elsewhere in the file, a route opens that does not run through an innocent, and the harder half arrives: the mark is not for adding a name, it is for taking one out.
12 lessons · 2 mini-sims · ~30 min
Begin module04/04
The Lookback
A monitoring rule at Keystone Bank did not fire for three years on a class of customer it should have. The gap is found, the fix is in, and the supervisor wants a lookback across the 154,100 customer-months sitting behind you in twelve cohorts. The first thing to understand is that you cannot simply decline it. If you produce no rate of your own the supervisor has nothing to use, so it applies its own three per cent assumption to the whole estate and charges you on it. The second is stranger: the exercise produces the number it is then charged on. Whatever rate your sample returns is applied to everything you leave closed, weighted by how much of the estate it came from. So a thin lookback can punish you for finding things, which puts the most standard plan in remediation, a small pilot first, in question. The board rewards a method rather than an answer. Which cohorts do you reopen, and how much do you review? Choose a scope and see what it costs, then ask whether the cohorts can be ranked at all. Reopening everything tops every counted number a dashboard shows; is it the right scope? And what does it cost to insist that your scope be statable as segments by years? Then one sentence of the methodology note changes: the supervisor will apply the rate of your worst cohort rather than your pooled rate. Does the method that produced your scope survive the new rule?
12 lessons · 2 mini-sims · ~32 min
Begin moduleAuthored on Decyde Practice loop curated for decision discipline
Vocabulary
What you'll engage
The decision skills you'll practice are paired with the underlying patterns you'll learn to recognize. Both vocabularies compound across the Banking Decisions domain.
Decision skills you'll practice
Credit segment readingBranch performance signalCustomer-tier trade-offsOperational risk weighingPolicy-vs-judgment balanceVolume-vs-quality calls
Patterns you'll see
Practice evidence recorded on Decyde
Every decision practiced builds your Decision Profile
Replays supported
Practice the same decision differently · iteration builds judgment
English
Additional languages on v1.3+ roadmap
What's next · One recommendation
The other Banking Decisions courses each practise a different kind of decision.
Begin practice →FAQ
Questions about this track
Common questions about Financial Crime & Compliance and how this track works.
Related · Other tracks

