Weighing the options

Information technology used to be an issue left to "propeller heads" in

the basements of government buildings, but those times are gone forever.

With IT now a core component of many government programs — and a pricey

one, at that — agency managers and legislatures need a way of judging the

worth of technology investments.

There have been a number of attempts to develop a process that would

enable that, but it's fair to say there has been, at best, only partial

success. The state of Iowa, however, thinks it may have the best method

devised so far to judge the merits of IT programs.

A result of Gov. Thomas Vilsack's "leadership agenda," Iowa's Return

On Investment Program targets benefits to both government and citizens as

the measure of the state's investment in technology.

It's that second group that Richard Varn, Iowa's chief information officer,

believes sets his state's approach apart. While other states have evaluated

technology from a business perspective, he said, no one else has included

a formal examination of the ROI for the citizen.

"This is the first time that IT evaluations have been done this way,"

he said. "I think program administrators will be welcoming of this because

they are faced with budget cuts, and now they can argue that such cuts could

impose a hidden tax on citizens without taking into account the full ROI

impact."

Say individual citizens or businesses spend an average of 40 hours a year

doing business with the government — for example, filling out forms — and

administrators can show that a new IT program would eliminate 30 hours of

that. That benefit should be part of the argument in favor of a government

investment because of the benefits that would accrue to the public at large.

Once the benefits are known, Varn said, the state can make choices about

how to reallocate IT funding to lessen the cost to government or to decide

how to do more with the dollars. The ROI process in that way becomes a guide

for the whole of the IT investment program, he said, from the front end

to the back end.

Glen Dickinson, division director for the Iowa legislature's fiscal

bureau, thinks the ROI process will make an impression on the state's lawmakers.

Most people in the legislature, particularly those on the appropriations

committees, feel they don't have the expertise to mark up IT projects and

are "too overwhelmed" with the subject matter to make rational choices,

he said.

"There are certainly other considerations involved [because] the budget-making

process is an inherently political one," Dickinson said. "But having a standard

ROI program in place should help them make a decision between various projects.

It will be useful for the baseline discussions that need to happen."

The ROI program will be used for the first time in the full appropriations

process during consideration of the fiscal 2002 budget. Agencies began submitting

their proposals in October, and every one that contains a request for IT

funding must have an ROI evaluation document appended. On July 1, agencies

will start spending the funds appropriated to them.

But the program has already produced dividends. Using the ROI methodology,

agencies reviewed their current IT programs and identified as much as $1.6

million in savings, and by combining some parts of separate projects, an

additional $840,000 is expected to be shaved from IT expenditures.

The IT Project Evaluation document first calls for an analysis by agency

executives of the project for which the funding is being requested. It asks

for a clear statement of the project's objectives and for a plan of action

that shows how the project will be implemented.

It also asks for a description of the purpose of the project, its expected

results, what criteria will be used to judge its success or failure, project

management and risk mitigation plans, and a detailed itemization of the

hardware and networking technology that will be used.

The ROI part of the evaluation runs the project's applicant through

a set of financial requirements, such as estimated costs of hardware, software,

people and other resources and supplies that will be needed, as well as

sources for the funds the agency expects to get its funding from, including

state, federal and local government money as well as private and other funds.

It also requires a certain amount of judgment on the part of the applicant.

In addition to "hard" money figures associated with reduction of costs to

the government, and those such as hidden taxes that the citizen would save

because of the project, the evaluation also calls for an estimation of citizen

or government benefits for such things as avoiding risks to health, security

or safety, providing enhanced services, avoiding the consequences of not

complying with enterprise technology standards, and so on, some of which

may or may not be quantifiable.

From all of that, the ROI of the project — defined as total annual benefit

minus total annual project cost, divided by the amount of requested state

IT project funds — is calculated.

However, that's only the beginning of the process. Each of the proposals

are further evaluated by the state's Information Technology Department (ITD)

and ranked according to a standard 10-step set of criteria. The Information

Technology Council (ITC) — a 17-member body formed to oversee the ITD and

executive branch IT activities — goes through a similar exercise, and the

project rankings of both the ITD and ITC are compared, with the final result

a single ranking of projects that is sent to Gov. Vilsack.

Following a discussion between the ITD and the governor's financial

affairs office, a master list of prioritized projects is put together, and

the governor uses this to provide budget recommendations to the legislature.

Lawmakers will then, theoretically at least, use that ROI-bolstered list

to make an educated decision on IT-related appropriations.

The list will not necessarily be the final say in what decisions are

made, Dickinson stressed, because the legislature will make its own analysis

of the various IT projects and the information that goes into the ROI process.

But lawmakers found that this kind of activity was useful when they

had to make decisions on Year 2000 issues, not least because it held agencies

accountable for the information they presented on their various projects,

he said.

"They expect this [ROI] process will provide the same kind of guidance

on IT generally," Dickinson said. "It's a good starting point for improvement

because they've never had anything before that they could use to make comparisons."

The same can be said of the executive branch, according to Cynthia Eisenhauer,

director of the Iowa Department of Management. Individual agencies may have

had their own methods for calculating the worth of IT projects, but this

is the first investment tool that will have an impact across the entire

IT enterprise, she said. Previously, any estimates of the worth of enterprisewide

projects had been, in Eisenhauer's words, "haphazard and unfounded."

"This raises the whole level of the discussion to that of the formal

ROI process, rather than leaving it to conversations around perceived priorities,"

she said. "That inevitably becomes a political dialogue that ends up at

the pork barrel."

But the biggest impact will be felt first in the agencies, said Tom

Shepherd, a technical adviser to the ITD's project office.

In the traditional appropriations process, agencies essentially guess

what they need to spend on projects and, typically, spend their allotted

amount conservatively for the first nine months of the fiscal year and speed

up their outlays tremendously in the last three in order to meet their appropriated

budget. This "brings all the hogs to the trough," he said, something that

doesn't "engender efficiency."

In the new way of funding projects based on the ROI process, agencies

will have to submit invoices for what they spend and will only be reimbursed

those actual expenditures. Any savings will accrue to the entire enterprise

and not to the individual agency. That means agencies will have more dollars

available for projects they convince the ITD are important to the enterprise.

At least, that's the goal, said Paul Carlson, director of the ITD project

office. One of the ways to get there, something that the ITD will encourage,

is for agencies to partner with each other on projects that have mutual

advantages. A more cooperative atmosphere will replace the stovepiped approach

of the past, where agencies kept their projects to themselves. Budgets will

follow the technology lead.

However, he said, Iowa is nowhere near that yet. As the new process

comes into play, there will be the inevitable turf battles, but Carlson

is confident people will come around. In the contemporary world of government,

IT has to battle with other projects for the available funds, "and at the

end of the day, everybody agrees that priorities are priorities."

It was never expected that this program would change things overnight,

Varn said. New approaches typically take anywhere from five to seven years

before they are completely accepted within government, he said — "three

if you are really lucky."

What will be needed is a period of analysis to see how the projects

started under the ROI program progress and, most importantly, if government

resources are allocated more effectively as a result.

"That," Varn said, "will be the real test."Robinson is a freelance journalist based in Portland, Ore.