Your Backup Supplier Is a Phone Number, Not a Plan

Ask a founder who their backup supplier is and you will usually get a name. Ask when that factory last made something for them, and the room goes quiet.

That silence is the whole problem. A factory you have never ordered from is not a backup. It is a contact you feel better for having.

Do you need a backup supplier?

If one factory makes more than half of what you sell, and losing them for a quarter would do real damage, then yes. But the only version worth having is a second factory that has already made your product at small volume. A name in a spreadsheet does nothing on the day you need it.

Almost everyone knows this and almost nobody has done it. The Allianz Risk Barometer 2026 puts business interruption, which includes supply chain disruption, third on its global list of business risks, named by 29 percent of respondents. Forty nine percent say they are renegotiating and diversifying their supply chains. Thirty five percent are looking at nearshoring. Three percent describe their own supply chain as very resilient.

Three percent. That is the distance between intention and a purchase order.

Your product is not a drawing

It is a drawing plus everything a factory learned while making it, and most of that learning never made it onto paper.

• The tooling sits in one building, and depending on what you signed, it may not be yours to move.
• The tolerances that actually matter got sorted out over three sample rounds and a phone call, not in the tech pack.
• Someone on that line knows the step your product fails at, and they quietly fix it without telling you.
• Your material and finish come from a supplier the factory found, not one you chose.

All of that has to be rebuilt somewhere else, and the first run at a new factory is always the worst run you will ever get from them. Source Haus has the execution detail, including who actually owns the mould and how to move factories without losing your tooling.

Why splitting your order usually makes things worse

The standard advice is to dual source. Run two factories, split the volume, sleep better. That advice was written for companies buying hundreds of thousands of units, and at that scale it is cheap insurance.

At your volume it is not insurance. It is a tax.

Split a run of 3,000 into two lots of 1,500 and three things happen at once. You drop under the price break on both halves. Your quality control work doubles, because two factories means two rounds of samples, two inspections and two sets of problems. And you go from being a decent customer at one factory to a small customer at two.

That last one costs the most and never appears in the spreadsheet. The factory that answers on a Sunday, holds a line for you in November and absorbs a small mistake without a fight is doing it because you matter to them. Two B grade relationships protect you less than one A grade relationship, and they cost you more.

The three honest options

There are only three, and the right one depends on how much cash you have and how much a stockout actually costs you.

Qualify a reserve, properly

Pick one alternate factory, send the brief, pay for samples and place one small real order. Not a trial. A run you will sell. What you are buying is proof they can make it, a price you can trust, and a relationship that will take your call.

Expect it to cost you. Samples, a tooling conversation, a first run at a price you would not normally accept, and a slice of your own attention. That is the premium, and it is worth knowing the number before you decide against it. Source Haus has the questions to ask in how to vet a factory before you pay a deposit.

A local maker can also work as a reserve for short runs even when it could never carry your volume. That is one of the few situations where manufacturing in Australia stacks up commercially rather than sentimentally.

Buy time with stock instead

If your product is stable and does not date, extra stock is a simpler backup than a second factory. Eight weeks of cover buys you most of the time you would need to move. It costs cash rather than attention, which for some founders is the easier trade.

It is also the option that quietly kills businesses when it happens by accident. Stock is only a strategy if you know what you are holding and why.

Accept the risk, in writing

This is a legitimate choice and most small brands make it. The difference between accepting a risk and ignoring one is whether you have written down what would make you act.

What should make you move

Concentration on its own is not a crisis. These are the signs it is turning into one.

• One factory makes more than half of what you sell, and you have no second price to sanity check against.
• The business has changed hands, moved site, or lost the person who ran your line.
• Quality is drifting run to run rather than failing outright, which is harder to raise and much easier to ignore.
• Replies have slowed and you have started making excuses for them.
• You have become a rounding error in their order book, or they have become uncomfortably dependent on yours.

The calendar does half the work for you

Chinese New Year falls on Saturday 6 February 2027. Maersk guidance for the 2026 holiday described production slowing for two to three weeks before the break, with capacity returning gradually afterwards. Trying to qualify a new factory in January, while they are clearing every order on the floor before the shutdown, is the worst timing available to you. Source Haus has the order dates in the Chinese New Year 2027 shutdown guide.

Nobody qualifies a reserve factory in a calm week. It almost always happens straight after something has gone wrong, which is the most expensive moment to start.

What I would do

Not everything needs a backup. Most ranges have one or two products carrying the business and a long tail that does not.

Pick the product that would hurt most to lose. Qualify one alternate on that product only, with one small paid run. Leave the rest where they are. That is samples, a tooling conversation and a few months of attention, not a second supply chain.

Go and see them if you can. You learn more about whether a factory can carry you in an hour on their floor than in a year of email, which is most of the reason I still take founders to factories in China rather than sending them a list of names.

Sixteen years in, the thing I would tell a younger version of myself is not to go and find a backup. It is to notice earlier when one relationship has quietly become the entire business.

Curious how exposed your current manufacturing arrangements leave you? Explore Source Haus.