What does China plus one actually mean?

Keeping your existing Chinese supply base while establishing a second source outside China, usually in Southeast Asia, so tariff exposure and single-country risk are reduced. The phrase describes a sourcing outcome. The work is mostly transfer and qualification: making a second site produce parts your existing inspection criteria will accept.

Why do China plus one moves cost more than expected?

Because the quotation compares unit prices, and the real cost is in qualification. Tooling that moves does not arrive at the same process window: machine tonnage, shot profile, die temperature and alloy chemistry differ, and defects appear that never occurred at the original site. Someone has to own that gap. On most brands, nobody is assigned to it, so it surfaces as rejected shipments months later.

Should the second source replace the Chinese supplier?

Usually not at first. A second source that has never run volume is not a replacement; it is an option you are paying to keep open. The useful sequence is to dual-source a small number of parts, run both sites against the same drawing and inspection plan, and compare real data before shifting volume.

How long does qualifying a second source take?

For moulded or cast parts, plan in months rather than weeks: tooling transfer or new tooling, first articles, dimensional and functional validation, then a pilot run at production speed. The schedule is decided by how many correction loops the process needs, which is why an honest first-article review matters more than an optimistic timeline.

China Plus One Is a Management Problem

Where I am writing this from

I am building one. In 2024 we bought roughly twenty thousand square metres in Semarang, Indonesia, and established PT SunOn Industries Indonesia. We broke ground on 30 April 2025. As of September 2026 construction is in its final stage, with handover expected before year-end. I am responsible for the project from land acquisition through construction to production and operations, and I am on site at least one week a month.

So this page is not a market overview. It is what the exercise looks like from inside it, while it is still unfinished.

Groundbreaking ceremony for PT SunOn Industries Indonesia, April 2025 — Groundbreaking, Semarang, 30 April 2025.

Said plainly, because you would find out anyway. That plant belongs to SunOn Manufacturer Group, where I am a partner. It is not a condition of working with me and it is not what I am selling on this page. If your second source should be somewhere else, or someone else, that is the answer I will give you — and the reason you should ask a supplier's own view of "where to go next" one more question than you otherwise would. How I handle that conflict →

Why the arithmetic disappoints

A China+1 decision is usually justified on landed cost: the tariff line goes away, the labour rate is lower, the unit price comes down. That comparison is correct and incomplete, because it prices the part and ignores the transfer.

I have watched a supplier change made to save eighty cents a part end up costing a hundred and thirty-four thousand dollars — and that was a move within China, between two factories speaking the same language, with the same die. Crossing a border does not make the same problem smaller.

What transfers with a die is the geometry. What stays behind is the process window that made it work.

Machine tonnage, shot profile, die temperature, alloy chemistry, operator habit — none of it moves in the container. The parts come out dimensionally similar and behave differently. Porosity appears in castings that never had it. A finish that passed for four years starts failing adhesion. Every one of those is discovered by someone, months later, usually by your customer.

The $134,000 supplier switch, in detail →

What the work actually consists of

Four things have to happen, and none of them is buying a building.

1 · Decide what is being duplicated, not what is being moved

Moving production means one site. Dual sourcing means two sites producing to one specification. These need completely different contracts, tooling budgets and inspection plans, and brands routinely start the first while describing it as the second. Decide which one you are doing before anyone quotes.

2 · Write down the process window, not just the drawing

The drawing travels well. The knowledge that makes the drawing producible usually lives in two or three people at your existing supplier and has never been written down. Capturing it — parameters, known failure modes, what the current factory does that is not on any document — is the single highest-return week in the whole project, and the one nobody budgets.

3 · Agree the acceptance criterion before first articles

A second source will produce a first-article report with deviations. Some of them matter and most do not. If the criterion is agreed beforehand — which dimensions are functional, measured from which datums — the review takes a day. If it is not, you get a hundred non-conformances on a part that assembles perfectly and a month of argument about whose measurement is right.

Why every part can pass and nothing fits →

4 · Run both, on real volume, before you shift

Qualification data from a sample run tells you the site can make the part once. It does not tell you the yield at speed, which is the number the business case depends on. Keep the Chinese source running and compare real production data across both. It costs more for two or three quarters. It costs very much less than discovering the difference after you have moved the volume.

Who does this inside your company today?

That is the actual question, and it is usually the reason the project stalls rather than fails. The tariff analysis is done by finance. The site visit is done by whoever was travelling. The qualification is done by an engineer who already has a full job and is twelve time zones from both factories.

China+1 does not fail on strategy. It fails because it is nobody's day job.

That is the thing I do. Not the building — the department that runs both sides of it: supplier management across the whole bill of materials, inspection at both sites against one standard, the correction loops, and the decision of when the second source is actually ready.

What a monthly retainer covers, and what it costs →

An honest position on timing

The Indonesia plant is not finished. Handover is expected at the end of 2026, and until it runs volume I will describe it as what it is: a site under construction that I am personally responsible for. If your decision needs a qualified second source producing this quarter, that is a different conversation and I will say so on the call rather than after the contract.

What is available today is the part that has to happen first anyway — deciding what you are duplicating, writing down the process window, and setting the acceptance criteria. That work is worth doing regardless of which country you land in, and it does not depend on my plant.

A second factory is the easy half. A second supply chain is the part that fails.