Let's define a few terms up front. Outsourcing means that you take a business function and pay an entity not directly a part of your company to do it. It could be anything from buying a part that goes into your product to someone answering your customer support phone calls.

Boeing says it will outsource 70% of its upcoming 7E7 Dreamliner plane. Boeing is not in the business of making comfy seats, in fact describing any airline seat as comfy is a stretch, or vomit-resistant carpeting. They are not in the blue chemical toilet business either.

It is a little more of a stretch to say a company is not involved in the generation of bills for its customers, but many large companies do this. Customer support is a real stretch, in fact it is basically saying that you don’t want to deal with the only source of income you have. Technical support outsourcing is just plain stupid.

Outsourcing your tech support is more or less the kiss of death for an organisation. Dell recently came to this conclusion. It outsourced its bread and butter corporate support to India, and customer satisfaction went into the toilet. It backpedaled pretty quickly, which is a pretty good indicator that it was hurting sales. Dell, as far as I have heard, has not pulled back consumer support.

Other companies, for their own reasons, do not fix outsourced support problems, even when hit with the proverbial baseball bat with "duh" written on it in block letters. HP is an example of this. I personally used to buy a lot of HP machines, and was very happy with the support. In fact it was the reason I kept buying HP. Then it outsourced its support, and it went to the smelly brown stuff. I no longer buy HP gear, IBM is my brand of choice, not that it won me over exactly - HP went out of the way to alienate me.

If a function is not performed inside your company, geography be damned, it is outsourced. India does not define outsourcing, and neither does Canada nor China. If you have a branch office in Bangalore, and you lay off the people to have a 3rd party do it, it is outsourcing. If that contract takes jobs back from India to California, a block from your corporate HQ, it is still outsourcing. The line is drawn between "you" and "not you".

Geography can play a role. If you outsource to a country where your native language is not the same as theirs, there will be problems. They range from stupid xenophobic reactionary arguments that have no basis in fact to real problems. It can be hard to understand someone with a real thick accent, and if you are under stress to get the server back up and now, that doesn’t help. This is not a problem relating to the country: I have had problems dealing with thick southern accents as well as Indian ones. Other times, I don’t have a clue about the location of the call centre, local or not.

There is one universal thread that ties most outsourcing efforts together, service universally suffers, and so does the "customer", internal or external. In the world of US corporations that can’t see past the end of the quarter, short-term savings are a good thing. When people don’t replace the widget they bought from you with another widget from you, but instead go to your competitor, you will probably panic. In the meantime, you are saving real dollars. In fact, by the time you lose customers to the competition, you have cashed out, bought your island, and retired. Worse yet, you go into politics. Hi Carly.

Both my personal experience and anecdotal evidence indicates that outsourcing is not paying off. Cost savings are nowhere near where they should be, or worse yet, were promised to be. Customers are not happy, and people are not buying your product a second time. Don’t believe me? Think about how happy you are when you finally get through the phone maze and are greeted with a foreign accent. Have you ever said to yourself "Boy, I am glad this person doesn’t natively speak my language. This call will be so much easier".

OK, so we all know that outsourcing is bad for people who buy your products. What most people don’t seem to realize is that it is worse for the company that does the outsourcing. There are two reasons for this, neither one of which is obvious. First is an expansion of the "Institutional Memory" concept I wrote about earlier. If you outsource, you effectively destroy the ability to promote from within. The other is that you place your only contact with the customer in the hands of people who have no economic justification to care.

Long term, the first is the worse of the two, there is simply no way to "fix" the problem once you realise what has happened. Customers can be brought back - look at the US auto industry in the late 70s and early 80s. Memory and experience can’t be bought at any price.

As I said last time, I have a friend who worked his way up from dishwasher to top rung executive at a multi-million dollar organization. He can tell you what every person in that organisation does, and at a pinch, probably do a fairly credible job at covering their duties. This is good, if there is a problem, he understands what the situation is without a half hour brief, and is never portrayed as a remote and removed executive.

Now, lets look at the other side of this. Say you are a large computer manufacturer that decides to outsource your technical support. Why have your own monkeys answering the phones for $15 an hour when you can have an outsourcer pay the same people half that to read from the same database? If you only look at the bottom line, logic would dictate that you could save a lot of money by outsourcing.

If you have 1,000 people in your call centre, the cost savings of moving the entire place from $30K a year to $15K amounts to over $15 million a year, enough to buy half a Gulfstream V! Short term, that is a powerful incentive, and a lot of people jump at it. If you get bids on a two year contract, in a mere 24 months, you can bid it again, and maybe save a million or two more. Wow, good deal, right?

Not when you look at the big picture. If you look at the 1,000 people in the call center, it would be safe to assume that many of them are not great workers. Some are good workers, and a few are outstanding people. Those outstanding ones are the ones you want to give more responsibility to, and have them move up the corporate ladder. Some of these people will stand out immediately, others will take time to develop and mature.

If you have an in house call centre, you can take these people, mentor them, tutor them, train them, and have them do better for you. Level one techs that are really good can take a class and become level two and level three techs. They have seen all the problems, and know how to deal with them from the basic to the complex. If you put them on another track, they can train in new employees, or manage the call centre. After a few years, and a few training courses, they have the potential to do very well in corporate management, certainly better than a MBA candidate who can’t tell you the difference between the servers you make and the box they come in.

When you outsource the call centre, this chain of command and ability to promote from within is irrevocably shattered. Those one in 100 people that HR would kill for get washed away in the next bidding cycle. There is no bottom up knowledge and personnel transfer, the bright ones simply go away.

The upside is that your company can concentrate on its so called core competencies, like making computers. The bad thing is that you have to do so with people that make it through the automated resume scanners in HR. Will there be bright ones that make it through? Undoubtedly. Will those people, on average, be of the same quality that the call centre weeding out process produces? Not a chance in hell. When you outsource, you lose the knowledge, skills and background these people build.

If you want to make a computer, do you hire people off the street, or people who have made the last three generations of your computers? What if the people off the street will cost you 50% less? Would any sane manager pick randomly from the labour gene pool to replace a tested team that has worked together for years simply to save a few dollars?

In America at least, the answer is a resounding yes, especially if you are in high tech. That corporate gene pool is being diluted at an alarming rate, and no one seems to care. Some companies really do get it, and Intel is a good example. Why did Intel take the Alpha design team and keep them together to design Tukwila? It probably would have been cheaper to hire a bunch of fresh grads to do the same work.

This type of oursourcing distances the top tier from the bottom, something that a company cannot afford to do. Steve Jobs can probably sit down and tell you what most tiers of Apple are up to. More importantly, he most likely could have a reasonably intelligent conversation with anyone in the company about what they are doing. At the very least, he can hold his own, no detachment there, in the corporate sense.

When you draw your next server executive from a toothpaste company because everyone else is in the Canadian outsourcing outfit, you lose, your company loses, and you pay more for less. Free tubes of whitening tartar control gel is not a substitute for uptime, and you can bet your company will not be in the running for the next round of bidding. Outsourcing causes this, directly, and no amount of minty fresh breath will save you here.