Short answer: An early warning system is a short list of signals (not news) checked on a regular schedule, with a predefined threshold for each signal and one person responsible for raising the flag. It beats an on-call advisor when the organization has continuous exposure: exports to several markets, an international supply chain or long-term investments. An on-call advisor suits a one-time decision.
The longer view
A good system contains 10 to 15 signals, not a hundred. Each signal is chosen because it precedes a specific scenario: a change in marine insurance premiums on a given route, a regulator's announcement of an import review, or a shift in a major power's official language toward the region. Each signal gets a threshold ("if X crosses Y, we convene the executive team") and a review frequency. Ownership matters most: someone has to read the signals every week and report, even when nothing happens.
Most systems fail for two reasons: they are built as news archives rather than signal lists, and they lack a mechanism that forces someone to decide when a signal crosses its threshold. A system that does not produce decisions is a newspaper.
This is a reading through the seven layers of The Blue Pumpkin, the method of Dr. Bella Barda-Bareket.
What to do tomorrow morning
Choose ten signals that precede the three scenarios that matter most to the organization, assign each a threshold and an owner, and allocate a twenty-minute review in every executive meeting.