the state and federal level, and a very specific contract with the HHSC.

"The challenge from our perspective was that the contract vehicle was very detailed and prescriptive," he said. "It was very inflexible."

The HHSC built the call centers into its plans with the federal government so it wouldn't have to start a separate approval process for the call centers, according to the report.

"We spent a lot of money on the application, so we just didn't want to throw it away," said Gary Gumbert, the HHSC's CIO. "Why go out and have to go through certification again? It's a waste of money."

 
Deadline Pressure
The first call centers were scheduled to open in late January 2006 - a pace that became a problem when TIERS wouldn't cooperate with another program slated to power the call centers, according to Flood's report. Instead of pushing back the schedule to allow more time for getting the programs working together, managers stuck with the original timetable and had call-center workers enter enrollment data into TIERS and SAVERR concurrently.

By that time, however, hundreds of HHSC employees who were proficient in SAVERR had taken new jobs because their offices were being replaced by call centers staffed by the contractor's personnel. The low cost of operating and staffing phone bays rather than lots of small local offices was one of the call center plan's most attractive parts. The Texas Legislature hoped to save hundreds of millions of dollars from the switch.

In practice, however, losing that much institutional memory may have come at a heavy cost. "The amount of training was different for the contractor staff, who were brought on in great numbers," Flood said, "but had almost no background knowledge."

Gumbert said getting staff - new and old - comfortable with the new program, when some had been using SAVERR for 30 years, has been a huge challenge, but an unavoidable one.

"I wish I could make it easy, but it's just painful for the current staff," he said, adding that the agency is hoping to set up new training labs to help bring employees up to speed.

 
Things You Can't Plan For
Any large program like TIERS runs within some margin of error, McCurley said. For instance, SAVERR had an error rate of 5 percent to 7 percent. But TIERS was under the microscope because of the hype associated with the call center switch. When the pilot indicated there were problems with the application, public outcry was swift because social services are of such vital importance and under such close watch by the media.

Furthermore, it's difficult to predict how employees will adapt to new programs and which mistakes they will make, McCurley said, explaining that those factors are why an agency starts with a pilot and makes adjustments to employees' actions.

"No matter how good a job you do of drawing up the process in a lab, [employees are] going to do what they're going to do," he said. "You really have very little control over their behavior."

The HHSC cut off its call-center contract with Accenture in March, and the controversy around TIERS and the push for private call centers came to a boil in April, with the release of Flood's report and a series of Legislative hearings. House committees questioned Albert Hawkins, executive commissioner of the HHSC, on management decisions and ultimately put the program under closer Legislative oversight.

"When you're doing a project this massive, it's a best practice to be sure you have a single point of oversight," Flood said. "That oversight makes sure that the technology side of the house communicates with the business side of the house. So it's not a tech marvel that produces

Patrick Michels  |  Contributing Writer
Patrick Michels is based in San Francisco and Austin, Texas. He writes for Government Technology, Texas Technology and Emergency Management magazines.