The study of failure is Japan’s new growth industry, says a May article in The New York Times.  The Japanese government has financed the “failure database project” which catalogs 600 Japanese failures.  There’s a Failure Society, also based in Japan, which publishes articles and runs seminars. 

Innovators know all about failure.  The Doblin Group estimates that more than 90% of all new products fail in the sense that they don’t make back their investment costs.  There’s thus a rich store of failures from which to choose.

The trouble is that it’s difficult to draw lessons for success from most examples of failures.  This is true for the simple reason that finding the things that went wrong doesn’t provide much direction for putting together what will go right. 

For example, here are some lessons drawn from failure, quoting from Japan’s failed supermarket tycoon Kazuo Wada:

"Businesses are most dangerous when they are successful, because people become arrogant and that prevents them from learning."

These kinds of lessons may be true and useful, but they don’t provide much guidance on what to do differently.

There are, however, ways of using failures that can lead to a higher probability of success.  The first step is distinguishing between three different types of failure:

1. Catastrophic Failures – the kind that kill companies and cost billions.  The Iridium Global Communicator was a catastrophic failure for Motorola.  The recent explosion of the space shuttle Columbia was a catastrophic failure for NASA.  Kazuo Wada’s supermarket chain Yaohan over-expanded, took on too much debt, and went bankrupt in 1997.  

2. Experimental Failures – the kind that happen in development, and result in products or services never moving beyond prototype.  Millennium Pharmaceuticals, for example, believes that one of its competitive advantages is its ability to fail new drug products earlier in their development process.   This allows Millennium to run more experiments than its competitors for the same amount of time and money.

3. Pro Forma Failures – These are the kind that you can predict before physical or market prototype. They are failures in a model as opposed to in a product or a service. 

Of these three, pro forma failure is most valuable because its cost is so low – it’s failing in a model, rather than in the real world.  Even more important, pro forma failure works in tandem with pro forma success.  You not only can see what doesn’t work, you can also see what does.

Failure Mode and Effects Analysis (FMEA) is an example of pro forma failure.  Most engineered parts will go through an FMEA analysis to reduce their chance of failure.  The part designs are tested on the computer, rather than in use.  Most of the parts for Boeing’s 777 aircraft, as well as the total aircraft system, were designed and tested on the computer rather than with physical prototypes.

Failure mode analysis can be done for other aspects of innovative systems, such as management and business design.  This allows managers to predict possible failures and to develop appropriate solutions. 

When BMW and DaimlerChrysler built an engine plant together in Brazil, for example, the project managers spent several months identifying management failure modes.  One of these involved imported parts being held up in Brazilian customs.   Brazil has historically managed its balance of trade by controlling the flow of imports into Brazil at its borders.  This meant that some goods could be held in Brazilian Customs for several months.  This kind of long delay for parts shipments would have contributed to a catastrophic failure for the engine plant, which was designed to run lean, with very low parts inventories.

Anticipating this potential failure mode, the engine team negotiated a fast track status for imported parts to ensure they made it quickly to the Brazilian factory. 

Just as with FMEA for parts design, the management failure modes analysis enabled the team to fix a potential problem before it contributed to a catastrophic failure. 

For more information:

1. The New York Times article, titled Learning From Loss Japan's Growth Industry: The Study of Failure, is in 18 May 03 issue.  Here’s a link: http://query.nytimes.com/gst/abstract.html?res=F30F12FC3E5A0C7B8DDDAC0894DB404482
2. The Japanese Society for the Study of Failure is here: www.shippai.org
3. The Doblin Group is here: www.doblingroup.com
4. Dartmouth Professor Sydney Finkelstein book on executive failure, Why Smart Executives Fail, was published in June 2003.  The book focuses on catastrophic failure.  Here’s a link to his book on Amazon:  http://www.amazon.com/exec/obidos/ASIN/1591840104/qid%3D1058025268/sr%3D11-1/ref%3Dsr%5F11%5F1/102-0422139-3650534