Most supply chains being restructured right now are not being restructured. They are being reconcentrated.
Every week there are announcements. Manufacturing relocating. Sourcing shifting. Dual source strategies under consideration for three years suddenly have budgets and timelines.
The pressure is real. But the logic driving many of these decisions is the same logic that created the concentration risk in the first place: optimise for the current constraint.
The last concentration, 20–25 years back, was built around cost. Supply chains moved to China.
This one is being built around tariff exposure. Supply chains are moving to Vietnam, Mexico.
The destination has changed. The mistake has not.
And tariffs are only one force. Protectionism and physical disruption risk are pushing the same decisions toward a new single point dependency.
The question not being asked with enough rigour: what does this supply chain need to do across a range of futures, not just the one where today’s tariff structure holds?
Because these forces do not stabilise. Most companies are treating trade policy and geopolitical disruption as one-time design constraints rather than ongoing risks to model.
The companies handling this well are the ones that slowed down long enough to ask a different question.
Not “where should we source to avoid the tariff” — that is a point-in-time answer to a structural problem.
But “what sourcing architecture lets us shift allocation as the environment changes, and what does each option cost in lead time, working capital, and service exposure?”
That is a scenario question. It requires a model, not a destination.
The companies that had already built probabilistic scenario models — ones that could run a tariff shock through cost structure, inventory position, and service commitments by SKU — had a material advantage in 2026. Not because they predicted tariffs. Because they built the mechanism to respond when the cost structure moves.
Most boards are funding infrastructure right now: relocation, dual source acceleration, geographic diversification. Necessary, but not resilience unless paired with the decision architecture to use it.
Decision architecture is trigger points and allocation rules before the next shock arrives, not debates after it hits.
A new supplier in a new location solves the tariff problem of Q1 2026.
Resilience is not a map of where your suppliers are. It is the capability to make a better decision than your competitors when the map changes.
How much of your current restructuring is infrastructure investment? How much is decision architecture?