An architectural council set up as part of the outsourcing deal and led by an Entergy CTO and technical leads provided by SAIC did a good job of setting standards but, according to Smith, it never played a role in bringing viable emerging trends to the business. On the occasions that SAIC did present Smith with ideas, "they'd say: Here's something new, what do you think about it?" says Smith. "The next question was: How much are you willing to pay? That's not want I want to hear from my value-add buddy. They didn't understand my business.

"I didn't expect them to live at the nuclear plant," he says, "but they could spend a week there." Smith included SAIC employees in all his own meetings, but the further he went up SAIC's chain of command, the less they knew. The one innovation for which the business was most hungry - wireless capability - "turned into discussion of SLAs and how to restructure them", says Smith. "They wanted to set it up so each wireless point was treated the same as a server or router." (Entergy implemented the wireless project with limited help from SAIC.)

"The further we got downstream from the deal, the more we got over our illusions about what the relationship with SAIC might be in the area of innovation," says Smith. "We couldn't figure out how to measure innovation and ultimately we had to decide whether we wanted them to manage the IT utility or innovate, because we weren't hitting either."

Willingness to innovate only decreases over time, says BAE's Fecteau. It's an open secret that the beginning of the outsourcing relationship is a money-losing proposition for outsourcers; they make it up on the back end. If you try to renegotiate to improve or expand services, the provider puts in a less experienced and cheaper team, says Fecteau. "They can still deliver exactly to the SLA, but there is no thought about improving or innovating," he says. To get that, "you have to build a new kind of contract".