Methodology
Not everything that changes becomes a trend. The sequence is how we decide what has.
01 — Event
A discrete occurrence that can be pointed at. Something happened, and a source can be named.
An event is the smallest unit we will use. It is local, dated, and checkable. A speech, a price, a rule, a product, a protest, a shutdown. If it cannot be sourced, it does not enter the sequence.
02 — Signal
An event that may matter beyond itself: a shift in behaviour, cost, rule, or expectation.
Most events stay events. A signal is an event with a direction in it — someone is paying, banning, copying, or abandoning something they did not before. Detection is a judgement, and it should be a narrow one.
03 — Pattern
Related signals showing up in more than one place, sector, or moment. Correspondence, not a conclusion.
A pattern is repetition with a reason to compare. Two cities repricing the same kind of space, or several regulators reaching for the same lever, are not yet a trend. They are a reason to keep looking.
04 — Trend
A pattern consistent enough to describe a direction of change. A description of what is happening, not a forecast of what must.
We use the word sparingly. A trend is a pattern that has held long enough, and across enough contexts, to be named without pretending it is universal. It can stall, reverse, or stay confined.
05 — Implication
A conditional reading of what the trend could change, and for whom. Open to being wrong.
Implications are not predictions. They say what would follow if the direction holds, who would feel it, and what would count as evidence that the reading was off. Certainty is not the product.
Evidence before prediction.
Signals before narratives
The sequence starts with what can be pointed at. The account comes after, and only if the evidence still holds.
Context before conclusions
A change in one market is not a change in the world until it has been set against others.
Implications, not certainty
We say what could follow, and where that reading might fail. A brief that cannot be wrong is not analysis.
Sources remain visible
A reader should be able to see where an observation came from, rather than being asked to trust the tone.
Uncertainty stays visible
Where the record is thin, or the comparison is weak, the brief should say so in the same place as the claim.