Note7: shipping new batch 2 weeks later seems absurd, but was it strategy?

Phil Baker wrote an excellent piece analyzing how a problem like the Note7 can happen.

The cost of the recall itself is estimated at $3 billion, but the fact that it is now a federal crime to carry a Note7 on a US flight has evaporated many more billions from Samsung's brand and stock market value.

Trying to prevent a massive recall like this is always a top concern, but this case highlights the tricky issue of how much money and especially time one can afford to invest in testing. Smartphones probably have shorter product life cycles than any other electronic product, so every week spent on testing means an opportunity loss of tens of millions.

Actually I feel you can’t really blame Samsung too much for the first recall, 35 defects on 300 million phones shipped per year is a very low percentage, so statistically there always is a fair chance that such a PPM (Parts Per Million) level problem does not come out in the tests in time to stop shipments.

But resuming shipping in 2 weeks looks extremely amateurish. We don’t know yet what really happened, but I wouldn’t be surprised if top management was breathing very hard down the necks of the production engineers: ”solve this issue TODAY or else..”. That may well have caused the premature conclusion that it was all the fault of a battery supplier, whereas it now looks like it may actually have been a defective power regulator chip.

Doing a proper failure analysis alone can in some bad cases take months, where you set-up all sorts of experiments and then wait for them to run, trying to find the true root cause. Fixing a problem means that something in the design was changed, so unfortunately the next step is that all the validation testing will have to be done again, to make sure that this “fix” does not have some unforeseen consequences. For an issue this severe they should have taken months.

With Samsung releasing so many models I wouldn’t be surprised if their whole smartphone strategy is based on winning on Time To Market, consciously foregoing much of the testing someone like Apple does, and in that way always offering the “freshest” models. It’s a poker game which until recently has paid off well for them, a PPM level problem causing a defective LCD is easily dealt with, and the gains of shipping earlier far outweigh the cost of those few replacements. But there always was the distinct possibility that they would run into an issue that would eh… blow up in their face.

Keesjan Engelen is founder of www.titoma.com and has been developing electronic products in Taiwan & China for 20 years. He also wrote: China ODM factory designs: 7 pitfalls to avoid