Most importers are pretty bad when it comes to assessing the probability of quality problems with a given supplier. There are two reasons for this.

The first one is that they might not know the factories that work for them. The solution is not always simple, especially when a trading company is blocking information about the manufacturing facility/facilities it uses. So it is important to avoid this situation from the beginning, by requesting to approve the factory.

The second reason is more subtle. Buyers tend to underestimate the likelihood of quality issues when they start working with a new supplier. And the reverse is also true: once they have noticed something unacceptable, they tend to overestimate the probability of future disasters.

This psychological bias (among others) is explained at length in The Black Swan. The author calls it the problem of inductive knowledge. I am going to explain how it relates to a typical importer’s situation, and what to do about it.

Underestimating the chances of a quality problem

If you haven’t had any problem with a Chinese supplier, you never will, right? Wrong! You’d be very lucky to work with one of the few A-level manufacturers, and even them are not 100% reliable.

Here is a great example from that book:

Consider the turkey that is fed every day. Every single feeding will firm up the bird’s belief that it is the general rule of life to be fed every day by friendly members of the human race “looking out for its best interests,” as a politician would say. On the afternoon of the Wednesday before Thanksgiving, something unexpected will happen to the turkey. It will incur a revision of belief.

The risk here is to find that your shipments are unsellable on your market once they are in your warehouse. Your supplier is not perfect and you should watch him! That’s a painful (and late) discovery.

The chances of a production disaster

This is another psychological bias: after your inspectors have caught an unacceptable production run, that supplier often loses your trust entirely.

Here is another example from the book:

After a [very surprising event], such as September 11, 2001, people expect it to recur when in fact the odds of that happening have arguably been lowered.

The risk for buyers is to commit excessive resources for monitoring production, double-checking quality in depth.

But this is often handled incorrectly: the result is to piss off the current supplier–who is not used to so much policing and resents it. In the end, the buyer/seller relationship turns sour and is often ended abruptly.

The best solution: a reasonable quality control plan from the beginning

First, you should choose the factories that make your products. All quality control firms perform some factory audits. You can then run a ‘legal records check’ with one of the many firms providing that service. Basically, the idea is to choose suppliers on the basis of (1) their capabilities and (2) the risks they represent for you.

Second, you should have regular quality inspections in place. You can send inspectors more often (say, during AND after production) to risky factories, and resort to skip-lot final QC if production is always good.

What do you think?