Skip to main content

Outsourcing: The Good, the Bad, and the Ugly – Lessons from Nortel Networks

Jeff Townley, former Chief Procurement Officer at Nortel Networks, provided some fantastic insights in my MBA class this evening.  In so doing, he shared some great insights about “Outsourcing:  The Good, The Bad, and the Ugly.”

The GOOD – “ODM’s were a wonderful way to outsource, and we had a great experience with them, as they are always good at common hardware. The ODM’s in Taiwan have a single tech park where they all started. Almost every one of the heads of these ODM’s all went to school together and started this Tech park. Nortel decided to put their phones as well as other data access products with the ODM’s – and they would help you by adding some design value. You didn’t have to add designers, and they leveraged the fact that they designed similar products for other people, could do it quicker, and would leverage that, and would get their money back through manufacturing in a low cost environment. At the time – it was all in Taiwan, and over time most of them opened manufacturing locations in China in order to compete on cost.

Clean sheeting was an activity Jeff led on all of Nortel’s projects, and which involved taking all of the transformation costs, where it was being built, and put it all down on paper. “We would go negotiate with other suppliers to try and move it. It didn’t make it easy to move, but having to learn a lot of experiences, we learned to move things pretty well. Nortel was on the first to create an outsource system house in Boston which was moved to Penang. PCB’s were outsourced in the 1990’s, and final assemblies not long after, but the system house where it was all consolidated, putting it through your systems test and shipping it – was all outsourced to Penang.”

The BAD – “Back in the 90s a lot was going to Mexico. Nortel had a plant in Calgary which produced phones that were moved to Mexico  This was Solectron, that was later bought by Flex. All of a sudden I recognized there was a problem – you are not keeping up with demand. I couldn’t get a good story out of them. Eventually we figured out that when Nortel had the PO’s on the suppliers for the long leadtime items – and they would get put back to the front of the queue. Component suppleirs put it at the front of the queue and it resulted in a 20 week leadtime! This was a 10B company – and a $50M hole in our fourth quarter. I got a team of people and we killed ourselves to get the orders out and only missed our revenue target by $2M, which ended up being noise on a bottom line.”


The UGLY – “Nortel went to Chapter 11 in 2008. One reason was they blew the whistle on themselves for financials – that maybe wasn’t needed. They weren’t doing anything that others weren’t also doing.  It is my feeling and the feeling of many others, that Nortel should have sold off some divisions and streamlined their business in the 2005-2007 timeframe rather then trying to maintain such a broad portfolio for the Enterprise and Carrier markets”. Nortel had an opportunity to sell the optical division to Corning for $100B – but decided not to do it! Years later, when they declared chapter 11 – they left Flex with $400M of inventory. Needless to say, the relationship got really bad.  They put everything on stop ship – would ship nothing. What good would that do us? We have to continue to ship to get paid by customers. But we wrote them a check for tens of millions. Nortel ended up paying 97 cents on the dollar for many of the claims when they sold their IP through the Google consortium, and paid off not all, but a number of their bills.

Today Flex has software that helps them understand when components aren’t turning fast enough. That was surely developed on the foot of getting stuck with Nortel’s inventory! Nortel was growing too quickly, and World Com and others realized they had a lot of “dark fiber” in the ground, and the internet was not being used.  Today, Netflix is 37% of Internet traffic on any given night! I guarantee you that if that had happened in 1999 – the fiber would not have been dark. And there is finally a push for more bandwidth. But it’s about 15 years too late!