Building Supplier Redundancy: Lessons from Diversifying Beyond a Single China Source

I’ve spent a lot of my career on factory floors in China, negotiating supplier terms face to face rather than reading about them in a report. That first-hand travel to source and qualify manufacturers built an in-country network my clients still draw on today. For one importer, that meant direct introductions to vetted Chinese suppliers as an alternative to the sourcing route they’d been locked into for years.

So when I talk about supplier redundancy, I want to be clear about what I’m not saying. This isn’t an argument for leaving China. It’s an argument against having exactly one way of getting the thing you sell.

The current environment makes the point better than I can. China’s rare earth export controls now cover 12 of the 17 recognised rare earth elements, and the second wave of measures was suspended only until 10 November 2026 – a deadline that’s now weeks away with no public clarity on whether it will be extended. The EU’s Critical Raw Materials Act has gone as far as introducing mandatory diversification thresholds to prevent exactly the single-country dependency that’s left European industry exposed.

You don’t have to be in rare earths to feel this. Here’s what my experience has taught me about building redundancy that actually works.

Single sourcing is a cash problem before it’s an operational one

I get asked to look at supply chain risk far less often than I get asked to look at cash. In practice they’re the same conversation.

When a sole supplier goes down – a shutdown, a quality failure, a licensing delay, a price increase you have no leverage to refuse – the damage shows up as expedited freight, air shipments instead of sea, emergency buying at spot prices, lost orders and stock you’re funding in the wrong place. I’ve rarely seen that cost quantified in advance.

My advice is to put a number on it before you do anything else. What does four weeks without your primary supplier cost in lost gross profit and emergency logistics? That figure is your diversification budget, and it’s usually far larger than the cost of qualifying a second source.

A second name on a list is not redundancy

This is where most “China plus one” plans quietly fail. A supplier you’ve never ordered from, whose tooling doesn’t exist, whose samples you’ve never tested and whose lead times you’ve never measured, is not a backup. It’s a phone number.

Real redundancy means a supplier who has produced your part, passed your quality checks, and quoted a landed cost you’d genuinely accept. Until you’ve run product through them, you don’t have a second source, you have an intention.

Go and look

My experience of qualifying suppliers is that the gap between the audit pack and the factory floor can be enormous. Capacity claims, sub-contracting arrangements, who else they’re serving and how you’d rank in a shortage. None of that comes through reliably on a video call.

I spent three weeks in Mexico assessing a manufacturing partner for a North American electronics distributor, and that groundwork is what allowed us to cut delivery lead times from up to 45 days down to a maximum of 15, while saving the client $1 million in the first year. That result didn’t come from a spreadsheet comparison. It came from being there.

If you can’t go yourself, use someone with an established in-country network who can. What you’re buying is judgement, not a site visit.

Compare landed cost, not unit price

I’ve watched businesses reject a perfectly good alternative supplier over a unit price that was 8% higher, without modelling duty, freight, routing, payment terms, minimum order quantities, tooling amortisation or the working capital tied up in longer lead times.

Routing alone can move the answer. For one global distributor I restructured US shipping routes to go via Panama to Florida rather than across the Pacific to California, reducing transportation costs by 35%. The unit price hadn’t changed at all.

Build a proper landed cost model per supplier, per route, and make the decision on the bottom line.

Check who your alternative supplier buys from

This is the trap I’d most want manufacturers to avoid right now. You can qualify a second supplier in a different country and still have a single point of failure, because both of them are buying the same input from the same tier-two source.

China’s controls reach further than most people assume. They’ve applied to foreign-manufactured products incorporating Chinese rare earth technology or processing know-how, even where no direct Chinese material was present in the finished item. Meanwhile the refining bottleneck sits at an estimated 85–90% of global capacity in China, which is why genuine diversification has to solve processing, not just assembly.

Map your bill of materials two tiers deep. Redundancy at tier one over a shared tier two is expensive theatre.

Get the contract right while everyone is still friendly

One of my clients put it well when he said the value I brought to Chinese sourcing was in the supplier management and, specifically, the contract terms and agreements. That’s deliberate.

Before you place volume with a new source, settle who owns the tooling and where it’s physically stored, what your IP position is, what happens to unpaid stock if the relationship ends, what quality standard triggers rejection, and what notice each side gives. My rule is simple: negotiate the exit at the start, when nobody wants to use it.

Dual-run permanently, even if it costs you a little

The businesses that coped best with the last few years of disruption weren’t the ones with a contingency plan. They were the ones already splitting volume, perhaps 80/20, across two qualified sources.

Yes, you give up some volume discount. What you buy is a live, tested alternative, real price discovery, and negotiating leverage with your incumbent that you simply don’t have when they know you have nowhere else to go. In my experience that leverage alone often recovers the discount you gave up.

Build the timeline backwards

Qualification, sampling, tooling, first article inspection, trial order and ramp-up typically run six to twelve months. If the reason you’re diversifying is a deadline – a tariff change, a licensing expiry, a customer audit – you need to start well before it lands.

I moved 60% of a toy importer’s operations to the Netherlands ahead of Brexit to protect European market access. That worked because we started early. The businesses that waited for certainty before acting had no options left by the time certainty arrived.

Where I’d start

If you’re sitting on a single China source today, I wouldn’t begin with a supplier search. I’d begin with two numbers: what a four-week outage costs you, and what your true landed cost is by supplier and route. Those two figures tell you how much redundancy is worth buying and where the money actually goes.

From there it’s a sequencing problem, and that’s the work I do with manufacturing and distribution clients — diagnose the real exposure, design a plan that protects margin as well as continuity, then help deliver it.

Frequently asked questions

Does supplier redundancy mean leaving China?
No, and I’d be cautious about anyone who says it does. Chinese suppliers remain highly competitive on cost, capability and scale. What you’re removing is the dependency, not the relationship.

How many suppliers should we have per critical part?
Two qualified sources for anything critical, with volume genuinely split so both stay live. For non-critical, lower-value parts, one source plus a mapped alternative is usually proportionate.

What does it cost to qualify a second supplier?
It varies with tooling and testing, but I’d rather frame it against the alternative: model what four weeks of outage costs you in lost margin and emergency freight, and the qualification cost usually looks modest.

How long does it take?
Typically six to twelve months from first approach to production volume, depending on tooling and certification. Start before you need it.

Where should we look beyond China?
It depends on your product, your customers and your duty position. I’ve worked with clients across Mexico, the Netherlands, Chile, North America and the UAE, and the right answer differs every time. Beware of moving assembly while leaving your raw material dependency untouched.

We’re too small to dual source. What now?
Then inventory buffer and a mapped, pre-negotiated alternative are your bridge. It’s not full redundancy, but it buys you time — and time is what turns a crisis into an inconvenience.

This article is general commercial guidance rather than legal or trade advice. Export controls and duty positions change quickly, so confirm current requirements before acting.

Carrying more supplier risk than you’d like, and not sure what it’s costing you?