Where there is a Starbucks there’s a (future) subway station, another rule in Shenzhen, and they seem to be expanding still. Well the store density is not even close to what it is in the US, where they reached a saturation point a few years ago.

So I thought about this a bit: why Starbucks, why now? There are plenty cool coffee stores around here, with plush couches, nice atmosphere, good music etc. I decided to ask: we know it is good quality coffee, yes expensive, but good.  I think that was pretty much it.

My attraction? Same as in the US. Nobody bugging you for more drinks, cool atmosphere, feels like your own little living room, study room, Jazz playing in the background, and yes, people watching (as they do in Europe on all these patios….).

Some of these factories have very large showrooms with sample products; providing good inspiration, designers have access to all production materials and realistic prototypes are made in a day or two.

The advantage is speed; the disadvantage can be substantial cost creep. Some customers would consider this a marketing cost, being there first and win. In a traditional bidding environment the price would be set in stone, some cost creep would only happen because of design and material changes.  

A very common trick used by supplier is the “Minimum Order Quantity” or “Minimum Color Quantity” numbers: MOQ or MCQ. These numbers can change with a new material, or is not communicated until a time close to production. When the customer can’t meet the MOQ/MCQ the unit price will increase, which is hard to negotiate down from thereon.

My feeling is that bidding should always be done with multiple suppliers, no matter the urgency, and that the tooling cost/upfront investment should be kept in check to avoid getting locked into a single supplier. I am surprised to see how many suppliers use the customer’s own tooling as leverage against them.

 The early Manhattan High-rises such as the Chrysler building were beautifully constructed, with many elegant details.  The labor to material cost ratio was 20:80. As this ratio flipped over the century to 80:20, the buildings became boxier in shape, unless awarded by wealthy institutions.  In China this ratio seems to be 20:80, possibly explaining the surge in building complexity.

Look at the latest opera house in Guangzhou, designed by famous Zaha Adid, known for her hard to build structures, even with today’s advanced design and planning tools. The construction was challenging but successful and on time.  Other great examples are: Guangzhou’s twisted TV tower, the 2010 Asian games buildings and the Shenzhen Universiade games venues for 2011.

Slower expanding rich markets don’t provide enough growth for large corporations such as GE. High end products on the other side are not competitive in emerging markets. The threat of new giants from these regions forces companies such as GE to innovate inside these markets, at much lower cost point. This reverse innovation is the opposite of the globalization approach that many manufacturers tried without much success.

As example is mentioned a PC based portable Ultrasound device, now selling at US$15,000. Originally these bulky devices were priced at $100k and then dropped to $30-$40k for portable versions.

The Chinese development team decoupled computer and software from the Ultrasound device, using a laptop and advanced software, replacing a large (expensive) box with custom hardware. This device has become a hit especially in rural areas, with not as high a performance as the origal 100% custom built model but “good enough” for use in remote areas. The 50% solution at a 15% price.

It’s interesting to see how intense competition drives aggressive development. Big brands thrive on market share, huge R&D budgets and manufacturing facilities, other brands compete by differentiation.

TCL, according to their news flash, is the first company to commercialize 3D TV

Proper QC however is more than a “look and feel” inspection. QC should balance all measures taken against product specs and warranty (long term effects). Nobody wants large volumes of returns, support calls (at $20 each) or worse: recalls.

QC can also save the client a considerable amount of money by lowering reject rates on molded parts for instance. Depending on part complexity this requires true in depth injection molding and tooling knowledge.

Larger volume production is a great exercise in statistics: from source parts to final assembly. Troubleshooting is ambiguous and can lead to multiple paths, each one possibly spawning new (undetected) issues.

This all asks for a very methodical approach, extending from the DFMEA done upfront, DOE on the source part creation and fish bone diagrams etc..

QC should be an extension of upfront engineering, by engineers in the right discipline who can communicate directly (without filters). I have seen too many occasions where other visitors claimed to have “solved the problem already”, whereas the real problems only came out in the next few days.

Let’s not take a short cut to solve a short cut….