Periodic review sets a date and checks the file when it arrives. The date has no relationship to the customer, so a change occurring the month after a review waits until the next one to be noticed, and a customer whose circumstances have not moved at all is reviewed anyway.
Perpetual KYC inverts that. The trigger is the change, and the review happens because something happened.
Why the diagram has no gaps
The superannuation rail on this site is mostly empty, because that page argues about a relationship where almost nothing occurs for decades. This one is the opposite reading of the same instrument: every span is occupied, because there is no waiting.
That is the whole distinction. A periodic model spends most of its time in a gap it created. A continuous one has no gap to be exposed in.
The step that decides whether this works
Monitoring continuously is straightforward. The difficulty is what happens next, because a model that treats every signal as a review request produces more work than the cycle it replaced and collapses within a quarter.
The materiality assessment is what prevents that. Most signals resolve there: noise, duplicates, changes with no bearing on risk. Only what survives becomes a refresh. Get that step right and total review volume falls, because effort stops being spent on customers whose circumstances have not moved.
The change that matters most is usually invisible
The single change most likely to alter a customer's risk is a change in who owns or controls it, and it rarely arrives as a customer notification. It shows up in a registry, in a filing, or not at all.
That makes ownership change detection the highest-value trigger in the set, and the one most likely to be absent from a programme that only watches names against lists.
What a supervisor asks under this model
Under a periodic model the question is easy: when was this customer last reviewed, and the answer is a date. Under a continuous model there is no such date, so the question becomes what coverage existed and what happened when things changed.
That is answerable, but only if the trigger history is retrievable per customer. Continuous coverage that cannot be evidenced looks, from the outside, exactly like no coverage.
For the underlying capability, see PEP and sanctions screening and transaction monitoring. For the ownership data the best trigger depends on, see Know Your Business. For what happens at the start of the relationship, see customer onboarding.