The line on LOB: Panel breaks down the fundamentals, and the payoff, of Lines of Business
The Association for Federal
Information Resources Management and GCN last month hosted a roundtable
discussion of the Office of Management and Budget’s Lines of Business
initiatives. The 10 participants in the roundtable generally agreed that Lines
of Business can work, while acknowledging that a lot of challenges lie ahead.
What follows is an
abbreviated version of the discussion. To read more, go to www.gcn.com and enter
428 into the Quickfind search box.
Tom Temin, editor in chief
for GCN and Washington Technology, moderated the discussion.
Scott Charbo, Agriculture
Linda Cureton, Alcohol,
Firearms, Tobacco and Explosives Bureau deputy CIO
Larry Gross, Treasury
Department acting CIO for E-Government
Charles Havekost, Health
and Human Services CIO
Randolph Hite, Government
Accountability Office director for IT Architecture and Systems Issues
Mark Krzysko, Defense
Department deputy director for Electronic Business, Defense Procurement and
Immigration and Customs Enforcement CIO
Price Roe, Justice
Department special assistant to the CIO
John Sindelar, General
Services Administration deputy associate administrator, Office of Governmentwide
Melissa Wojciak, House
Government Reform Committee staff director.
TEMIN: Well, Tarrazia
and I were discussing, when we came in, that there is not a lot known about how
Lines of Business will be accepted. It is a good thing in the eyes of OMB. So
what's your perspective on where we actually stand with picking Lines of
Business Centers of Excellence and moving toward them?
Price, I'm going to call on you.
ROE: Yes, I think we
know it's real because we are held accountable by OMB. One of the greatest
challenges for us—I say "us" meaning
the federal community—has been looking at the
volume and trying to figure out to what extent there is wheat and chaff as it
relates to an individual agency's mission.
It's the heartburn that sometimes accompanies these very good
So they are all real but there's going to be more—some
are going to have a greater alignment with the mission and, therefore, greater
focus and, therefore, receive more resources from within the department. And not
all are created equal in terms of what we are able to spend our time on.
TEMIN: You said that
there is a large number—you mean of mandates?
ROE: Yes, but also,
there is a large number of Quicksilver initiatives. Look at E-Payroll; that is a
great success story. We have a lot of movement on E-Authentication. E-Travel,
there are a lot of things.
At the same time, we have to look at the reconciliation with the
existing initiatives. Case management is something that is a natural for us at
the Department of Justice. However, we had a previous financial management
And there was a lot of back and forth on, you know, "Well,
are we going to shut this down and go with another program? Or are we going to
work closely with OMB to designate this as a
and therefore be a service provider, one of the service providers for the rest
of the federal community?" And we ended up going with the latter.
You have to balance your parochial departmental interests with
the larger federal picture.
And that's where OMB is very good at kind of grabbing by the
scruff of the neck and saying, "Look around," and they have the
ability to do that. And we feel the love sometimes.
GROSS: I think one of
the things that Price brings up is that this is no longer just an IT project. It
is forcing people to make business decisions on how their organization will run
today and years down the road.
MARTIN: I think there
are a lot of world drivers that are pushing us toward this goal. For example,
9/11 I think was a wake-up call for all of us, but when you look horizontally
across the government, there is a need to share data.
But at DHS, our big challenge is to share the same data about
the people that may have an effect on our nation's safety. And I think a large
part of what drives case management is how we will share that data. We call it
identity management. Who is in the country? What are they doing? What is their
So I think a large part of what OMB is trying to do with this
initiative is to force us to think horizontally, versus stovepipe. And I think
that's a huge driver for where we need to go as a country.
SINDELAR: You know,
the government tends to sacrifice the long term for the short term. And what we
are trying to do now is take the long-term outlook and build a shared-service
environment, starting with the 24 e-Gov initiatives and moving towards the Line
of Business effort.
I would say that the real driving force, the technical driver
behind this, is through the enterprise architecture.
And Larry is exactly right that this is not about IT.
This is about business transformation, how the government does business.
It’s been a long haul to get the chief financial officers’ community
involved, the acquisition officers involved in this as much as the CIOs because,
in fact, they all have to be at the table.
HAVEKOST: But the
realization that people need to have is that "systems" embody
processes and embody policies. And if we have hundreds of systems in the same
business area, we have embodied processes and embodied policies slightly
different hundreds of times.
And that's not a good way to do business, and that's not a
particularly transparent way to do business. But the Lines of Business give us a
way to have systems that embody process and policy a fewer number of times
across the government.
It makes the process more transparent. It makes us more nimble
about being able to update policies and processes, because they only need to be
updated in a very few number of systems as opposed to having some negotiation
about how to embody those processes and policies across hundreds of systems that
then would need to be updated.
SINDELAR: In the Line
of Business effort, the goal is to alleviate the requirements for every agency
to stand up their own financial management system, their own HR system, and the
business processes related to that over time, and allow the agencies to focus on
CURETON: It think of
the Jack Walsh model at General Electric Co., what he did with GE. What is the
business of GE? What is their main business? What should they get out of? And I
think this makes a very sensible way for departments to focus on what their
missions are and to divest themselves from some of the ancillary but very
important activities that support the mission.
TEMIN: I just want to
follow up on what Charlie said about duplication and with it the possibility of
the policies and processes being slightly different on all of those systems.
Maybe Randy can answer this: Should every agency have the same policies and
And if not, then do the LOB service providers simply become
time-share operations running everybody's applications? Does that make sense?
HAVEKOST: I'll jump
in there. That's a very reasonable discussion in this space. There are a lot of
processes that people perceive as policies just because they've been the process
for so long. It compels us to take a look at these things and say, "Are
these real? Or are these just vestiges of how we've done business for so
Are we in such a rut that we can't see that there may be a more
common, a more standardized way of doing business that gets us to the same end
and gets us there in a way that is more transparent, that is more understandable
to external entities, and it also can use systems that are used across?
GROSS: We were
looking at an approach where we were going to allow you to analyze your business
and make a determination of: At what point in time does it make sense to adopt
this business process here, from the
? And what part of that
do you want to utilize within your organization?
So it's not a one-size-fits-all. It does allow you, as a
business owner, to decide what fits your organizational needs.
CURETON: I think that there are probably some mission-specific
policies at play, and there are probably some specific laws. But there's a large
amount of that which is just past practice or just how we always do things. And
so think the real challenge is to be
able to balance between the two and not sacrificing one for the other in looking
at the long term versus the short term.
KRZYSKO: The policy
at the top level, what we have institutionalized in organizations are generally
tribal customs, the policy interpretation at lower levels implemented through IT
infrastructures that ultimately become the bane of our existence. The policy has
Now, you had the premise, Tom, that what happens is the policy
goes to automation. Policy is not a requirements document for an instance of IT.
And in many cases, the policy gives you broad-level latitude for implementation.
In the different organizations and their local implementations, there is
certainly uniqueness, but it's generally not from a policy perspective. When you
get to the requirements document or the requirements specs as we would do it in
a common business process, it becomes more difficult then because many people
implement that based upon their view of the world, not from the policy guides.
We can agree that we have a policy framework to worry about,
whether it be the financial framework, whether it be the Joint Financial
Management Improvement Program, whether it be the Federal Acquisition Regulation
and, in our case, the Defense FAR, because we do have some specific laws passed
for the Defense Department.
But fundamentally, you can find a core, an 80 percent solution,
that you can implement, but you have to have very strong leadership to bring
HAVEKOST: Can I offer
another example? This is when I was running Grants.gov. Some agencies require or
recommend a letter of intent that you intend to apply for a grant. And in
general, that is not part of regulation; it's part of how they do business.
And when it comes right down to it, most of the places that
require that, they don't actually require it; they recommend a letter of intent.
And the reason they recommend it is so that they can estimate how many
applications are going to come in, so they can estimate their workload at the
application due date. But there are other ways in the electronic world to figure
out how many. How many times has the application package been downloaded is a
way to estimate what the interest is. It takes quite an epiphany for people to
say, "Well, maybe we don't need a letter of intent."
TEMIN: We've talked
about what to preserve of investments that you've already made as an agency and
what to let go of. Sometimes, as in the case of the FBI, they let go of $100
million-and-some just because it didn't work.
In the other case, the financial system, the investment was
already made, and it turned out to be a service provider for others.
So at what point can agencies feel safe in just simply letting
go of sunk costs, because there is a Line of Business provider that might be
able to do that function better, and not get zinged for it?
SINDELAR: I've been
on detail to OMB since March of last year, so I'm your surrogate. But the Line
of Business concept—again, through the
enterprise architecture and the business case process—is
to focus on new systems, development modernization or enhancement to your
current system, meaning your current investment. So it's not to go after
mission-critical systems or even systems that are in place today that are
operating fine and shutting them down.
And, in fact, there's an alignment report that is being
generated as we speak that will identify where we see redundancies as related to
the programs and systems that are up in place for e-Gov, such as IE and others,
and why those systems still need to be funded.
But the focus has been on new investment money, on the Lines of
HITE: With regard to
your last question, that is practical reality, the fact that investments are
going to be ongoing, whether you are talking about Lines of Business across the
federal government, or within a department, or within an agency.
And the question about, are we, through Lines of Business,
talking about establishing a common policy and a common set of procedures to be
implemented across all these agencies? No, we are not, and architecture doesn't
talk about doing that either.
Architecture talks about, for your enterprise, identifying what
are those aspects of the enterprise that we need to do consistently across that
corporate body, and which things don't we need to do necessarily consistently.
Because there are unique mission that are performed by component organizations.
So that is the million-dollar question when it comes to
architecting an enterprise, that minimalistic approach to creating the blueprint—just
enough, not too much.
Actually turning lines of business into a reality is just a
huge, huge undertaking from a governance standpoint. Who is responsible and
accountable, and has the authority to do what in this arena?
KRZYSKO: You need
Those physical data elements, those business rules, those
how-you-report, are definitive. We've been working on that for a year in the
deployment of a system, a federal procurement data system.
That certainly touches and drives back through any architecture,
because procurements and grants are fundamentally all integrated in the
financial management process. That architecture needs to be documented.
MARTIN: I was
actually going to ask you a question about the governance. Has OMB begun
thinking about how that would work or how we can play into that?
There are still a number of skeptics out there, and rightfully so. But OMB can't
always be seen as the big daddy with a big stick. What we need to do as agencies
is own this and develop an effective governance structure as Randy talked about.
I think one of the models right now is probably Integrated Acquisition
Environment, IAE, in terms of a good governance structure in place.
They are engaged and they have enforced a budget discipline.
departments and agencies worry: "By having another agency or another
agency's contract vehicle provide LOB services, will I get the attention and
service? I can't send, you know, just a cure letter. Do you trust the other
agency to handle your family jewels?
CHARBO: I don't think
Line of Business is all that new. You
know, we've had the
for a long time. We've used the Bureau of Public Debt for HR functions for
awhile. So there have been pockets of that for quite a long time; that's out
There are a lot of things that we have to do, just trying to
make it a bit simpler, I think it's as simple as it gets. It's a customer and a
service provider. And why can't you leave them if they aren't providing the
product effectively? I mean, if that's not part of the equation, then it will
You tend to go
where you get good products and good services, and they develop a brand. And
even within the government, there are a few of those brands out there where you
traditionally have gotten good products and good services.
And now I think
what the Lines of Business are even doing is bringing those groups together and
saying, "Notch it up to the next level. Let's compete those. You three,
four, five even, let's compete. Let's go to the next level and maybe improve it
SINDELAR: You need
service-level agreements with metrics that essentially stand up as contracts
between agencies. But at the end of
the day, you should not care where you are getting your service from, as long as
it is effective and meets the metrics.
There has to be
another mechanism to address where there are issues with performance and the
ability to move from a failing service provider to the new service provider. And
that's one of the things that we are struggling with and addressing in the
governance discussions that are ongoing.
KRZYSKO: You also
have to have strong program management governance to be sure that the execution
of those programs meet those critical needs, all things considered. It's not
just as simple as letting a contract; you have to entertain the transitional
elements of moving in Department of Defense and moving that whole structure to
TEMIN: Well, you
can't even look at DOD as an entity in this context. The Marines don't even want
to move to the Navy's personnel system.
KRZYSKO: But in
IE we've tiered our governance structure down to the Army, Navy, Air
Force, all 27 components, where they all have a voice at the table and we take
their vote. So they have to come to
the requirements table; so when we go to a shared service environment, our
office represents a sole voice for the Department of Defense in terms of
fine except for the situation that happens when the need of the few outweigh the
need of the many. And in the mission
space, there are reasonable but rare, probably, occurrences where that happens.
And so, the governance process needs to be flexible enough to take those
situations into account.
SINDELAR: And OMB has
said, as recently as this week, that if you have unique requirements—and
small agencies do have unique requirements—where
maybe the common solution doesn't meet in a cost effective way, they need to
come in and make their case and they will hear them out.
TEMIN: Linda still
looks worried. Can you give us an example of what your experience has been,
something that really is unique that you would still want to retain control of?
CURETON: I'm worried
about the small component, the small agency that doesn't have the clout of a
large department; the Homeland Security Department versus the Federal Trade
Commission, and how the two would fare in making their business case to whoever
that they have a legitimate, unique requirement.
And I'm thinking of a situation like that where the two CIOs may
not have the clout to actually impose their will to the decision-makers. And
that causes me some concern.
SINDELAR: Well, I
think it comes down to the question that Tom put in the first place: trust. OMB
has said their door is open if you want to come to them through the business
case process. I mean, I understand your fear.
KRZYSKO: It's about
showing up and understanding the requirements, because we had to deal with that.
I mean, while somebody would say, "DOD would have the big
vote at the table," you know, we have components that are very small that
may have uniqueness that we have to represent and understand the requirements.
TEMIN: What kind of
legislative cover for Lines of Business that doesn't exist now do we need, if in
fact we do need it, to make Lines of Business happen in a meaningful,
widespread, deep way?
HITE: Lines of
Business is good government; it's good management. You don't need to legislate
CURETON: I disagree
just a little. And this will sound kind of schmaltzy, but the principles that
our government was founded on, the separation of powers, in and of itself
suggests some redundancy and some duplication, and that to maintain to balance
MARTIN: Part of what
has been challenging is to make one of our organizational elements a shared
service provider for others. And we've lived through a lot of pain in trying to
do that. But everyone comes to the table, 22 CIOs sit around the table and some
will say, "Well, this one legislative issue won't let me do that," or
"This one policy issue won't let me share or let this provider be my
service provider," because they find every trick in the book, every
business process, risk factor, et cetera, to say "It will not work."
It's because the culture is "I've got to stay within my own domain, manage
my own data, be in control of my own destiny."
TEMIN: And I don't
mean to trivialize what Randy said, but there are a lot of things that are good
that are nevertheless legislated.
HITE: Oh, absolutely.
And if things aren't moving fast enough, sometimes Congress will take matters
into its own hands. And part of that deliberative thinking about "How do I
best architect my enterprise," needs to recognize "Where does it make
sense that I need to seek legislative change so that I can do things
WOJCIAK: I tend to
agree with Randy. I think that many of the legislative tools that you need in
order to give you cover to go to Lines of Business already exist in statute.
So I think we need to see Lines of Business evolve before we try to
develop legislation to help you all.
TEMIN: Oh, you're
absolutely. I think it's a very good, important initiative that makes a
tremendous amount of sense. It really is going to crystallize, I think, a lot of
what was trying to be done with the Quicksilver initiatives.
It's a very important management tool, and Rep. [Tom]
is very supportive of it as an initiative.
But we have the Clinger-Cohen Act, the e-Gov Act, Federal
Information Security Management Act, and all of those are very nice pieces of
legislation that deal with management principles across agency and information
sharing initiatives. So I tend to think that the capability is already there.
And you know,
after such a rush period of legislation as well, I tend to think we also need to
make sure that we are reviewing the legislation, that we are reviewing how the
e-Gov office is operated, and doing our oversight before we do any more
legislation. Sometimes—and this might shock you—if
we rush to legislate, we don't do a very good job, and we can harm your
capabilities to move towards these management initiatives.
TEMIN: I want to get
a little bit into the weeds now, here, on how Lines of Business might work.
How does it, from a management sense, work if your agency—say,
your department, and then getting down to your agencies and bureaus—are
all using all different Lines of Business providers? We could really have a lot
of different lines of communication going out. How would that work?
And the follow-on to that question would be: How do you make
sure that your Line of Business provider is, in fact, technically up to date?
Even though systems are only the enablers of all this, technology does change,
and the changes in technology do, in fact, enable better business function.
GROSS: Let me speak
to that as an advocate for e-government because it does make good business, so I
would look at it just like as if you were home. You buy various services for
your house; you get cable television; you get gas through the utility company;
you get water through another utility company. And you pay for those services
with an expectation that those services will always be there. It allows you,
using your home as an analogy, to focus on those core things that y