A few months ago, I was in a conversation with a senior IT executive at a Fortune 500 company with a substantial European presence. He had just wrapped up an RFP to select offshore IT providers to handle project work that was conducted inhouse. The project spend was substantial – in the range of €150 million as best as they could tell. He had an interesting problem on his hands - "Now that I have these providers in place, how do I get my managers to send them work?" In other words, how should a manager recognize that a project had the characteristics to be outsourced?

Large sums were riding on getting this right – the average Indian offshore rates are a third of the European or North American rates. This was an area on which I had a point of view. I had begun my career at a major application outsourcing shop, one of the Tata companies, and had stayed in the industry for several years. The bane of the IT outsourcing industry in the early days was figuring out what types of work could be outsourced. There were plenty of failures on both sides – suppliers overcommitting and buyers trying to sweep a mess under the carpet by outsourcing it. We had our own "Top 5" reasons for why outsourcing projects failed:

1. Outsourcing a bad/ineffective process to a partner - outsourcing to shift a problem to someone else

2. Poor project scope definition (this is why managers find it easier to just hire contract workers –project scoping takes discipline)

3. Excessive focus on cost reduction – heart surgery by the lowest bidder?

4. Vendor mismatch due to over representation of capabilities

5. Internal resistance from managers who oppose outsourcing/offshore

Few companies have embraced outsourcing as completely as General Electric. In 2000, they talked about a 70/70/70 model – 70% of all GE work would be outsourced, of that 70% would go offshore, of that 70% would go to India. GE and others came up with tightly defined and repeatable criteria to help their managers make the right outsourcing decisions at the project level. The unit eventually took in outside investors and began serving other companies besides GE.

Back to my conversation with the IT executive. The €150 million was being spent by hundreds of managers making thousands of independent decisions. They were the front line. There was no central coordination. The executive wistfully spoke of wanting something that was as easy to use as Quicken or Turbotax. It was a good analogy – those tools have taken a numbingly complex subject (the tax code) and made it accessible to mere humans. They did it by creating an elegant decision tree that guides users through only the things that are relevant. Companies will have to solve this problem for outsourcing if they are to see the full benefits of the global service delivery model.