Most organizations experience their operating environment as something that happens to them, they react to events after they occur. A working indicator framework changes that relationship. It doesn't predict the chaos itself. Instead, it notices the early warning signs (the "pressure drop") while there's still time to do something about it.
What a framework is for
An indicator framework is a standing (always-ready) answer to a single question: What would we expect to see change before the risks we care about actually happen, and are we deliberately watching those things? It doesn't depend on analysts, data feeds, or software. What it needs is decisions made once, then reviewed regularly about what matters and what "movement" looks like.
Six steps
- 1. Name what must not surprise you.
Five to ten priority risks, written in operational language: loss of a critical supplier region, a targeted campaign against the organization, regulatory posture hardening in a key market, disruption to a dependency everyone forgets until it fails. If the list is longer than ten, it is a risk register, not a watch list and it will not be watched. - 2. Define leading observables.
For each risk, ask what would plausibly move first, and choose observables that are checkable in the real world: filings, permits, appointments, procurement patterns, local reporting, sector chatter, pricing anomalies. The discipline is to prefer leading over lagging an indicator that confirms the event has happened is a newspaper. - 3. Map the sources.
For every observable, note where it would actually be seen: which publication, register, platform, relationship or internal report. Unmapped indicators are wishes. This step also reveals honest gaps observables nobody can see, which are themselves valuable knowledge about blindness. - 4. Set thresholds with attached actions.
For each indicator set, define what amber and red mean, and bind each level to a pre-agreed response: brief leadership, activate a contingency review, open the crisis structure. A threshold that triggers nothing but concern is furniture. - 5. Fix cadence and ownership.
One named responsible. A known process rhythm, weekly or fortnightly for most environments, daily in elevated periods. A one-page summary: what changed, what it means, what we recommend. - 6. Review and let indicators decay.
Quarterly, ask three questions: did the framework notice what happened; what tripped that should not have; what has quietly stopped being diagnostic. Indicators age. A framework that never retires an indicator is accumulating noise with a clear conscience.
An example scenario
Consider a manufacturer dependent on a supplier cluster in one foreign region. The specific risk they're watching for is losing access to that entire supplier cluster for more than thirty days.
Early warning signs to watch:
Rather than waiting for the disruption to happen, you monitor a few leading indicators in that region:
- Notices of strikes or labor action in the area.
- How long export permits are taking to process.
- Local news (in the local language) about power or water rationing.
- Whether the suppliers themselves are changing their hiring or payment behavior.
- Changes in insurance and freight costs for that shipping corridor.
How to react:
- Amber level:
When two of these indicators start moving, that's the trigger to launch a sourcing review and brief the relevant people. - Red level:
When there's clear evidence that disruption is imminent, you activate the alternative-supply plan you've already built in advance.
The framework's product is not foresight. It is a shorter distance between the first sign and the first decision.
Three traps
Vanity metrics: indicators chosen because they are easy to count rather than diagnostic (media mention volume is the classic offender).
Hindsight indicators: sets rebuilt after every event trying to optimize the entire framework. Dashboard theatre: investment in visualisation; a beautiful display of undiagnostic data is a screensaver.
The common cure for all is the same review discipline: does movement in this indicator actually change what we believe or do? If not, it goes.
Sense-making frameworks such as Cynefin make a broader point worth keeping in view: in genuinely complex environments, indicators inform judgement rather than replace it. The framework tells leadership where to look and when to lean in. The leaning is still theirs.