State Government
Reboot
The state's IT systems are so old and outmoded
it needs a new culture as well as new computers
By Lawrence Bivins
During
the 1990s most corporations made the Information Technology leap from lumbering
mainframes to decentralized networks of desktop PCs. It wasn’t easy or cheap
but the astounding gains in productivity helped ignite the economic boom of the
last decade. Now it’s government’s turn.
Three separate agencies soon will tell the General Assembly that state
government badly needs to improve the way it spends nearly $1 billion a year on
IT expenses. The reports from the State Controller’s Office, State Information
Office and State Budget Office will detail their assessments of the state’s
inventory of computer systems, and all indications are the reports will not
paint a pretty picture. There are security gaps, out-moded systems, excess
costs, underperforming systems, minimal operator training and horribly
fragmented management controls.
State government has tried to drag itself into the Information Age, but its
attempts so far have been notable mainly for shedding light on the depth of the
state’s technology management problems. NC WISE, a $50 million effort begun
five years ago to link computers at public school systems statewide, may
ultimately cost taxpayers more than $150 million. A project aimed at moving
state government’s labyrinthine procurement system online has met with similar
disappointment. Recently, several state PCs were sold as surplus with sensitive
information still stored on their hard drives.
While buying new software and hardware certainly will be part of the solution,
the needed improvements are as much a challenge of state government culture and
organizational behavior as of servers and laptops.
“Everything we’re trying to improve on relies on culture,” says George
Bakolia, North Carolina’s chief information officer. His organization, known
as the Office of Information Technology Services (OITS), is involved in several
initiatives aimed at getting the state’s massive IT program under control.
State government centralized its computing two decades ago, Bakolia explains, by
consolidating its mainframes under a single authority. But the environment
changed rapidly during the 1990s as “distributed” computing — PCs for
everyone and software applications tailored for every purpose — led individual
agencies to develop more or less freestanding technology operations. “At that
point, we began creating silos,” Bakolia says.
Last year, the General Assembly asked OITS to launch an assessment of the
state’s IT assets, some of which are alarmingly old, and compile
recommendations and cost estimates on modernizing those systems in the worst
shape. The existence of these “legacy” systems are the most urgent challenge
to administering crucial government service in a secure, reliable and
cost-effective manner.
So disorganized are the state’s technology resources that it had no mechanism
for inventorying all its assets, says Bakolia, who has been CIO since May 2002.
That alone made it impossible for OITS to ensure network security, a basic
responsibility of the organization. “In order to guarantee security, I have to
know what we have,” Bakolia says.
The recently completed first phase of Bakolia’s assessment, which covers all
state operations except universities and the Administrative Offices of the
Courts, points out the magnitude of the state’s decaying technology
infrastructure. Nearly half its 52,315 computers, for example, are no longer
supported by their manufacturer due to their age. Another 25 percent have only
limited vendor support. Users of antiquated hardware must often pay extra to
obtain help when problems occur, or muddle through them on their own. Aging
systems are also more vulnerable to virus attacks and other breaches because
security patches are not readily available.
State government offices use more than 2,400 applications, the study found,
ranging from web portals to programs that process retiree pension payments.
About a third of those were deemed by OITS as “mission critical,” meaning
their failure would significantly impair an agency’s ability to deliver key
services. Among the state’s 824 mission critical applications, the assessment
uncovered 99 that were more than 10 years old and based on obsolete
technologies. State systems are also ripe for consolidation: there are, for
example, some 40 different applications in use for tracking employee time and
leave, the study found.
“Every agency senses that they have unique requirements, and they likely
do,” says Bakolia. “But from an efficiency perspective, we’ve got a lot of
duplication.” The incompatibility across diverse systems often means data must
be re-entered manually in some instances. And by purchasing such a patchwork of
hardware and software, the state fails to maximize its huge purchasing power.
It also never reaches a critical mass of skills necessary to operating many of
these systems. Making matters worse, many agencies facing budget cuts in recent
years chose to reduce IT training, Bakolia says. That’s false economy, he
adds: “They wind up having to hire outside contractors when they find they
can’t run their systems, and that turns out to be more expensive.” Indeed,
the OITS survey found vendor personnel operated 25 percent of all state
government applications.
Manpower skills are among the most troubling aspect of the state’s reliance on
legacy systems. Older applications are likely to be run by employees nearing
retirement. More recent hires are sure to be trained on the latest programs. In
many cases, it is impossible to find training on antiquated applications. “My
concern is that we’ll soon be left holding the bag unable to use these
systems,” says State Controller Robert Powell, the government’s fiscal chief
and a member of the Information Resources Management Commission (IRMC). IRMC,
created in 1992, is a 23-member panel of top state officials that sets and
oversees technology policy.
The Controller’s office is responsible for the business functions of state
government, including its sizable accounting system. Thus, technology management
is something more than a passing interest for Powell. In January, his office
completed a two-part comprehensive review of state government’s business
systems infrastructure, working with Bakolia, State Budget Officer David McCoy,
Personnel Director Thom Wright and technologists from Deloitte Consulting.
The study, also conducted at the behest of the General Assembly, found dated and
duplicative technology in wide use across government. It was emphatic in its
assessment that the state is missing out on opportunities to improve efficiency
and customer service due to the lack of strategic technology planning and
management. The absence of strong “core systems” led most agencies to
implement their own solutions, leading to the Tower of Babel-like assemblage of
uncommunicative applications. “Unless the state steps forward and addresses
the need for a new enterprise business solution from a statewide perspective,”
the report says, “the efficiencies and costs savings that can be realized from
a system enterprise approach will never be realized.”
The Deloitte study offered the state three options: The first, and least
dramatic, calls for enhancing the capacity and reliability of existing systems,
starting with the areas of highest risk. It would create an enterprise “data
warehouse” to ease the gradual replacement of legacy systems and begin the
process of designing a long-term strategy for introducing new financial,
budgeting, and tax and revenue systems. The overall approach would require the
least up-front investment, the report says. While addressing near-term risks of
catastrophic failure, the option does little to ensure against system problems
recurring in the future.
The second option phases in new “best of breed” solutions over a five-year
period. Such an approach calls for a somewhat larger, though still manageable,
initial investment. It would require the state to maintain several technical
platforms, boosting the risk of communication breakdowns between new and legacy
systems. The option is the “best compromise of business and technical
approach,” the report suggests, “however, (it) may be too aggressive for the
current political and economic environment.”
A third option calls for the aggressive implementation of a fully integrated
solution for handling core business functions. Going that route would provide
the greatest technical and business benefits, according to the report. The major
drawback: its high initial cost. The option also would amount to a “major
change management challenge.” The strategy also likely means the state would
rely on a single vendor, elevating the stakes higher still.
Getting a Handle on Projects
For any option to succeed, the state would need to organize its project
management resources wisely, establishing a governance structure charged with
minimizing disruption and ensuring North Carolina gets the most out of its
investment. “We need to put more emphasis on project management,” according
to Powell. The state’s lack of strong in-house project management skills has
been the source of some of the past disappointments, he says, although with
every project important lessons are learned. The very nature of technology
projects requires leaders to accept a new array of expectations about costs,
timetables and objectives. “If you’re constructing a building, you can go to
the site daily and see the progress occurring,” explains Powell. “But with
an IT project you really don’t know what you’ve got until you test it.”
Yet a third General Assembly-authorized study of the problem is under way at the
State Budget Office. Set for release in April, it will offer insight into how
technology operations are financed and budgeted by the state. “It’s clear
that everything we do with IT relates to the way we do budgeting for IT,”
according to Bakolia. His organization, for example, offers services and support
to state and local government agencies on a fee-for-service basis, relying on no
recurring General Assembly appropriation. That means the more cash-strapped
agencies might be unable to access the technology resources they need. On the
other hand, better funded government operations are free to adopt whatever
systems they desire, bypassing OITS and operating with little concern for what
other agencies might be doing.
“It’s very disturbing to me that this is so fragmented,” says Sen. Linda
Garrou (D-Forsyth), an Appropriations Committee co-chair. “We’re putting a
huge amount of money in these programs and people are right to ask, are we
getting our money’s worth?” She would like to boost the authority and
capacity of the OITS, an entity that has, in various incarnations, drifted from
department to department and now rests under the Office of the Governor. Garrou
says it’s vital that the state utilize technology to improve service delivery.
Some states, for example, offer driver’s license renewals over the Internet,
while North Carolina motorists still wait in long lines at DMV offices.
“This reminds me very much of the kind of problems companies had to tackle
about 20 years ago,” says Bill Coley, the retired Duke Power president now
serving on the N.C. Business Council for Fiscal Reform. The council, working in
conjunction with NCCBI, is examining ideas for achieving greater efficiencies in
government operations. “ I think the State of North Carolina should attempt to
benchmark itself against the private sector,” Coley says. Along with Duke
Power, for which he led comparable technology management reorganizations, Bank
of America and Wachovia Bank offer ideal examples of how sound technology
planning and management strategies can enhance competitiveness, he says. “I
know that government operates differently than the private sector, but we need
to move closer to those models.”
While the state spends close to $1 billion on its various technologies, the
issue has to do with more than money, Coley says. “Just investing in
technology alone won’t get you there unless you also change the way you’re
doing things.” How North Carolina delivers services to citizens and companies
is a clear reflection of its attractiveness as a place to live, work and do
business. “We want to be a destination for new jobs and investment by
high-technology companies,” Coley explains. “The best way to achieve that is
for our government to be seen as a model.”
Return to magazine
index
|