Last updated: 28 July 2026. This is structural sourcing guidance from a trading company — the patterns below are what we see in day-to-day work, not a claim about any particular factory or a substitute for your own commercial and legal judgement.
Single sourcing is rarely a decision. It is a default that sets in because the first good factory made looking for a second one feel like wasted effort. This page makes the case structurally: what one supplier costs you while everything is still working, when a second source is the wrong answer, and what actually holds when it isn't.
The four costs
Four costs. Three of them apply even when nothing goes wrong.
- No capacity headroom. Win a large order and you are limited by one factory's line time. Growth stops being something you decide and becomes something you have to ask permission for.
- No price leverage. A supplier who knows they are your only option has no commercial reason to hold their price, and every reason to test it. This one is quiet and continuous, and it never announces itself.
- No quality pressure. The same dynamic drives quality fade. A factory with no competitor for your volume has less reason to hold spec, and the drift is gradual enough to look like nothing.
- No fallback. The one everybody thinks of, and the rarest of the four. But when it lands it does not land on an order — it lands on a whole season.
When not to
Second sourcing is wrong more often than the internet admits.
The argument for a backup is easy to make and easy to overapply. There are four situations where adding a second factory costs more than the exposure it removes, and recognising them is most of the discipline.
- Skip it when tooling costs are high and would have to be duplicated to make the second factory real.
- Skip it when splitting the volume drops both factories below workable MOQs and raises your unit price on everything you buy.
- Skip it when the product needs deep process knowledge that takes a year to build — you would not have a second source, you would have a second learning curve.
- Skip it when you have one SKU, stable volume and a genuinely good supplier. That is a relationship to invest in, not one to hedge.
A second source you never order from decays. Prices go stale, your contact leaves, and their capacity quietly gets committed to somebody who does place orders.
A dormant backup is a document, not a supplier. If it is not being kept current, it will be discovered to be out of date on precisely the week you needed it — which is the same as not having one, minus the effort you spent.
What actually works
Four moves that hold, without doubling your workload.
The useful version of second sourcing is not a mirror image of your existing supply chain. It is a set of narrower, cheaper positions that keep an option alive where an option is worth having.
- Split the volume, not the product. An 80/20 split keeps the second factory warm, current and motivated without fragmenting your buying power at the first.
- Second-source the risky SKUs only. The highest-volume item, the seasonal item, and the one with a single sole-source component inside it. The rest can stay where they are.
- Qualify without ordering. Audit, sample, price and document a factory so it is activatable in weeks rather than months. Cheaper than a live second source, and most of the protection.
- Separate the geography. Two factories in the same industrial park share a grid, a labour pool and a weather event. That is not a second source; that is the same source twice.
The three positions, side by side
- Live split Both factories in production. Most protection, most management, highest ongoing cost.
- Qualified backup Audited, sampled, priced, documented. No live orders, so it has to be refreshed to stay real.
- Single source, eyes open One factory, with the exposure written down and the trigger for changing it agreed in advance.
The third position is a legitimate answer, and it is the one most often reached by companies that have actually done the arithmetic. What makes it defensible is not the choice itself — it is having made it deliberately, with the tooling question, the MOQ maths and the geography written down. That is also what makes it a decision you can put in front of a board. If you want the exposure narrowed rather than removed, inspection before shipment and firmer procurement control do real work on the same risk without a second factory.
What we do
Tell us what you buy, and from how many factories.
We assess your current supply base for concentration risk, identify which SKUs actually justify a second source, and then either qualify a backup or run a live split — whichever the arithmetic supports. That assessment is free, and it includes the case for changing nothing where that is the honest conclusion.
Where a second source is warranted, the search itself is supplier research, running two factories against one specification is procurement management, and restructuring the whole base rather than one product line is supply chain optimization. If the pressure you are managing is geographic rather than commercial, China plus one covers the version of this argument that crosses a border.
A single factory visit is US$268 and you can book it outright — a supplier verification visit or a pre-shipment inspection, whichever of the two you need. Everything else beyond the assessment is quoted per request, before any work starts. And if your current supplier has already stopped replying, this is not your page — start with what to do when a supplier goes silent, then come back to the structure afterwards.