Well, it's happened again--another spectacular crash and burn of an information-technology system that was supposed to be the magical answer to a state human services system's performance and cost woes. This time it is Indiana's 10-year $1.16 billion deal with IBM to pre- and re-qualify clients for health and human services ranging from TANF to Medicaid to food stamps. Two and half years into the deal, Indiana Governor Mitch Daniels flipped off the switch, cancelling the contract and sending IBM packing.

As the story is told in the general media, it's the classic and predictable bad-guys-in-action story line: Governor Mitch Daniels, conservative Republican, looking to do a little union-busting, decides to turn over a significant portion of the work done by the state's Family and Social Services Administration (FSSA) to IBM. As part of the deal, hundreds of former public employees are summarily shifted over to IBM, where they serve as at-will employees, outside of the state's public employee collective bargaining system.

Meanwhile, IBM low-balls the cost of providing the new high-tech screening and requalification system. Adding a little spice to the story line, IBM's prime subcontractor on the job is Affiliated Computer Services, which used to employ Mitch Roob, now state commerce secretary and formerly head of the FSSA, and one of the guys who helped hatch the privatization scheme.

The acute reason that Indiana's partnership with IBM failed is that the system couldn't handle the influx of applicants applying through call centers and computers; complaints of errors and long waits rose. The state warned IBM to make corrections to the system, before finally ending the contract.

But as with all these complicated mega-IT deals and their promises of doing more with less--and fixing every problem vexing government simply by bringing servers, desktops, fiber-optic cable and telephones into the process--there's lots more going on here than first meets the obvious story line.

Yes, Governor Daniels is a Republican with a penchant for privatization. And no, there's not a great deal of affection between him and organized labor. But Daniels doesn't dive into these deals for ideological reasons, or because of the close connections some of his appointees might have with some of the principals on the private sector side. Daniels is a pragmatic guy who wants to figure out ways to make his state more efficient, and he honestly believed he was doing that when he signed off on the privatization plan. And yes, IBM probably did overpromise when it signed the contract, but large corporations don't go into these deals expecting to fail, or to have their internationally recognized logos publicly dragged through the muck.

What went wrong with this deal is what usually goes wrong with these deals. First, no major overhaul of how government does anything should be based on a promise that the job at hand will be done more cheaply. Rather, it should be done on a promise that the job at hand will be done more efficiently, delivering a better outcome. If cost savings result, that's great. But deals where the prime driver is saving money usually will be destined for trouble. And yet negotiations around these projects almost always start with the money piece and not the performance piece.

Second, simply slathering some type of comprehensive new technology over old technology that "evolved" to help public employees administer programs that are already incredibly complicated, rule-bound, siloed, regulated and inter-governmental--not to mention dependent on human behavior and habit--is just asking too much of technology. Technology is a solution, not the answer, which helps explain why IBM's performance was so uneven. An e-mail sent to IBM was not returned prior to publication, but for IBM's part, it has told other publications that high unemployment along with the current awful economy stressed the system.

Third, these deals need to be true partnerships with the required competencies being brought to bear jointly by the public and private sectors, alike. Governments need to dramatically boost their capacity to write RFPs that will get honest responses. And they must fashion contracts that shift away from the premise that you can do a "snap-shot" assessment of an agency's needs and then create a technological fix based on that snap-shot demand--customers and costs are simply too dynamic for that. The private sector, for its part, needs to stop promising the moon and planets, and instead promise to work closely with public officials to achieve certain performance levels, while being honest about what can actually be delivered, how quickly and what that's really going to cost. This appears to be where Indiana is now headed, as officials discuss a "hybrid plan" whereby state employees will work with ACS on fixing system problems.

Finally, simple solutions to delivering human and health services will continue to be elusive as long as we have such complicated, convoluted, rule-addled and fragmented systems. In last month's newsletter, I quoted Georgia's health and human services director, B.J. Walker, who believes that in the future more of those who need public assistance will demand control over how they access and manage that help, which suggests an absolutely vital role for technology. But Walker's observations in the November newsletter got this spot-on response from a reader:

"If we truly want to put more work and responsibility on the customer, then we must simplify our systems and processes in a way that makes it intuitive for customers to use. At the end of the day, 'it's their responsibility' ends up producing a lot of rework for case/financial workers... If we want customers to grasp the self-help model (which is a great concept), then we need to radically transform our business processes."

Some states and localities are figuring this out and are making great progress. What characterizes that progress is that they all have recognized the same thing: That no private sector provider is ever going to be able to sweep in with some wonderful new turn-key system to take over and operate significant parts of human and health services programs without the close cooperation not only of public officials, but also of the not-for-profit provider community that is so central to fulfilling programs and providing services.

Technology is absolutely vital to progress--and the private sector has a very significant role to play in all that. But even more vital is continued public sector control and guidance over how these systems are designed and administered. The trick will always be blending the two realistically, based on honest assessments of need, risks, potential complications and cost.