Company representatives hold an on-site "kick-off session" with agency officials to walk them through the information gathering process. Each official is given a detailed questionnaire, and the representatives make sure officials understand exactly what is being asked in every question.
"[The Hackett Group] provides a workbook that says, 'Here are the definitions of all the processes we are talking about, and this is where to go to fill in the data,'" Wilson said.
It usually takes two days to gather the information and eight weeks to 10 weeks to conduct the studies. The cost of one benchmark study for one administrative area in an agency is $80,000. A progressive discount applies to each additional study purchased after that, as the state benchmarks multiple agencies.
The Hackett Group recommends that states study all agencies for comparison purposes. Sometimes one agency outperforms another in the same state doing the same task, according to Wilson, who added that states would likely attain updating services from the company every three to five years.
Before NASACT established the contract, The Hackett Group and Accenture performed pilots in Tennessee, Alaska, Nebraska, Oregon, Washington and Arizona. Participants were quick to warn of the pilot's deficiencies, but all agreed that it gauged the status of states' back-office operations.
Among the compiled pilot results in the High Performance Best Practices Study for the Benchmarking Committee of the National Association of State Comptrollers was a striking observation -- the public and private sectors spent similar amounts on financial processes, but the distribution of costs and resources was dramatically different.
The public sector paid a lot of entry-level employees to perform tasks mostly done by technology in the private sector. Additionally private-sector organizations hired fewer workers overall, but more professional-level employees, making higher salaries. And private companies vastly outspent governments on technology by more than 2-to-1.
The public sector's use of technology best practices lagged behind the private sector's, according to the pilot studies. Productivity in high transactional areas -- such as cash disbursements and compiling and reporting transaction data -- was significantly lower for the public sector. Government's cash disbursements productivity was, on average, 60 percent lower than the private sector, and transactional compilation and reporting cycles took longer in the public sector than in the private sector.
Jan Sylvis, chief of accounts for the Tennessee Comptroller's Office, said she was initially apprehensive about participating in the pilot.
"I thought, 'Oh man, there it is in all of its glory -- the good, [the] bad and the ugly that we have accomplished,'" Sylvis said.
Washington was one of the first states to perform a multi-agency benchmarking effort in 2002, before the NASACT contract was established. That state's back-office employees had similar concerns.
"The concern about benchmarking that [I] discovered [was] typically, 'If we go out and benchmark, and we find out that somebody else is doing this in 18 seconds, and it's taking us three days, are we going to get nailed or blamed? Is anybody going to take into consideration the fact that they can do it in 18 seconds because they have bigger, better computers?'" said Mary Campbell, special assistant to the governor of Washington for the Office of Management, Accountability and Performance.
Sylvis and Campbell both said a good benchmarking effort emphasizes that faulty systems caused the inefficiencies, not the employees themselves.
A key theme throughout the pilot was automation -- the private sector got more professional-level work done for its money than the public sector by automating most clerk-level tasks.
Clark Partridge, comptroller of Arizona, is negotiating his state's benchmarking services, but warned that states must be willing to spend the money required to change business practices and purchase the necessary