Consolidation is still rational—until policy makes it irrational overnight
Procurement teams didn’t start consolidating suppliers because they enjoyed dependency. They did it because it worked: fewer suppliers to manage, better volume pricing, cleaner specs, and simpler quality control. The problem is that today’s biggest supply shocks are often political choices, not operational accidents. A supplier can be “perfect” on OTIF and still become unusable because a tariff moves, a license is denied, a shipping lane becomes risky, or an input is suddenly classified as strategic.
KPMG’s recent supply chain outlook and similar 2026 guidance across the market point to the same uncomfortable reality: regulatory pressure, national security screening, and regionalization are not temporary. The old playbook—single global source plus a safety stock buffer—doesn’t match a world where the rules of trade can change faster than your next contract renewal.
Tariffs and trade controls don’t just raise cost—they scramble your sourcing math
Tariffs are often treated as a line-item surcharge. That’s a mistake. They change relative competitiveness between regions and can flip the “best value” supplier without any underlying change in manufacturing efficiency. If your selection model assumes stable duty regimes, it will mis-rank suppliers and encourage false certainty.
A practical way to think about it: you are no longer buying only a part or a service; you’re buying an import pathway. That pathway includes classification, country-of-origin rules, transshipment risk, documentation quality, and the supplier’s ability to provide evidence when customs scrutiny rises. Two suppliers with identical unit prices can have radically different landed-cost volatility.
A working model: “Tariff-adjusted total cost” with volatility bands
Instead of one landed-cost number, build a range. Use your best estimate plus a downside case that reflects plausible policy moves during the contract term. You don’t need a crystal ball; you need a disciplined habit of asking what happens if duties rise, exemptions end, or a product code interpretation tightens.
Calculate base landed cost (unit price + freight + duties + brokerage + compliance admin).
Add a “policy variance band” (e.g., a range of duty outcomes) and score suppliers on exposure, not just average cost.
Separate controllable cost (supplier efficiency) from non-controllable cost (policy). Don’t punish suppliers for the latter; design around it.
Require origin and classification documentation as a contractual deliverable, not a nice-to-have.
Nearshoring isn’t a moral stance; it’s a lead-time and jurisdiction decision
Nearshoring is often sold as a resilience upgrade. Sometimes it is. Sometimes it’s an expensive detour that creates new bottlenecks (limited capacity, thinner sub-tier ecosystems, higher labor cost, or weaker tooling depth). The right question is narrower: where do you want your lead time to sit, and under which legal and regulatory regime do you want disputes, inspections, and enforcement to happen?
Regional supplier ecosystems matter more than the final assembler. A nearshore plant that still depends on farshore subcomponents can reduce transit time while keeping geopolitical exposure. Procurement needs visibility beyond tier 1, at least for the few inputs that can stop production.
A blunt test: “Can the region reproduce the part without heroic effort?”
If the answer is no—because the region lacks specialty materials, certification labs, or tooling vendors—then your nearshore option is a dual-dependency disguised as diversification. That can still be worth it, but it should be priced as a risk-reduction move, not as a clean replacement.
Designing a multi-sourcing portfolio (without blowing up cost)
Multi-sourcing fails when it’s treated as “add a second supplier.” That approach often creates two half-committed suppliers, inconsistent quality, and no one willing to invest. Multi-sourcing works when it’s treated like a portfolio with roles: one supplier optimized for cost, another for continuity, sometimes a third for surge capacity or regional compliance.
This is where procurement has to be opinionated: paying more for resilience is acceptable only when the business agrees on what it’s buying. If the organization wants uninterrupted supply, then redundancy is the product. If the organization wants lowest cost, then it must accept concentrated risk. Many teams pretend they can have both without trade-offs, and that’s how you end up with brittle supply at premium pricing.
Three allocation patterns that show up in real categories
80/20 split (cost leader / continuity supplier): Use when switching costs are moderate and you can validate equivalence. Keeps the primary supplier motivated while funding a viable alternative.
Regional split (Americas / EMEA / APAC): Use when logistics reliability and regulatory alignment matter as much as unit cost; reduces border-crossing exposure and shortens recovery time.
Capability split (high-mix / high-volume): Use when the disruption risk is tied to complexity; a flexible supplier covers engineering changes and low runs while the scale supplier drives unit economics.
Contracting for fragmentation: terms that matter more than price
When supply chains fragment, contracts need to anticipate forced changes. The most common procurement frustration is getting blamed for shortages while being locked into terms that assume stable trade. If you can’t renegotiate quickly when policy shifts, your “savings” are theoretical.
Regulatory change clauses: define what happens if tariffs, sanctions, export controls, or licensing requirements change mid-term (price adjustment mechanics, re-sourcing rights, and documentation duties).
Dual-tooling and IP provisions: clarify who owns tooling, who can replicate it, and how quickly a second site can be qualified without legal friction.
Allocation and priority rules: during shortages, specify how capacity is allocated and what data the supplier must share (ATP, backlog, sub-tier constraints).
Audit rights for origin and compliance: origin claims and restricted-party screening should be auditable, not trust-based.
A procurement decision framework: cost-competitive, diversified, and explainable
Executives don’t need procurement to predict geopolitics. They need procurement to show its work: what you bought, what risks you accepted, and what options you preserved. The best teams I’ve seen use a simple scoring model that’s easy to defend when conditions change.
The “4D” scorecard
Duties & controls: exposure to tariff shifts, licensing, restricted inputs, and documentation failure.
Distance & time: lead time, transit variability, and recovery time after a disruption.
Depth of ecosystem: availability of local sub-tier suppliers, labs, tooling, and alternate logistics routes.
Dependence: concentration risk (single plant, single raw material, single lane, single engineering owner).
Use the scorecard to justify when you will pay a premium for redundancy and when you won’t. Then tie it to a concrete action: a second qualified source, a regional buffer, a redesigned spec, or a different incoterm strategy. Multi-sourcing isn’t a slogan; it’s a set of options you can actually exercise under pressure.