We're funded by these television companies and more, including Comcast.
Oh, you think this is just a community center?
No, it's way more than that.
Comcast is partnering with a thousand community centers to create Wi -Fi -enabled lift zones so students from low -income families can get the tools they need to be ready for anything.
Comcast supports C -SPAN as a public service, along with these other television providers, giving you a front -row seat to democracy.
Up next, a look at efforts to modernize technology within the Veterans Affairs Department.
During the hearing, lawmakers and officials from the VA also discuss the agency's 2025 budget and benefits for veterans.
This is an hour and a half.
Good morning.
The subcommittee will come to order.
I want to welcome our witnesses to discuss the lessons we can learn from the VA's efforts to modernize its technology during the last two years.
The VA is an organization of nearly a half a million people that serves 9 million veterans and their family members using thousands of IT systems and has a total budget of roughly $370 billion a year,
making its second largest agency only surpassed by the DoD.
The scale of the IT systems are massive and presents a lot of challenges.
The department has repeatedly attempted to address these challenges with megaprojects that span years or even decades and cost billions of dollars.
This strategy simply has not worked well.
All six of the VA's multi -billion dollar IT projects have either stumbled badly or collapsed altogether.
And this should come as no surprise.
I challenge our witnesses to point to any IT project anywhere with a budget this large.
That has been successful.
We've spent many hours in this subcommittee discussing VA's three mega projects that are currently racking up cost overruns and scheduled delays.
Electronic health record modernization has spent nearly 10 billion dollars already and would cost more than 37 billion dollars to implement according to the Institute for Defense Analysis.
And the true cost across VA is even higher than that, as it doesn't include the enormous additional staffing requirements or the loss in productivity.
VA and Oracle are on the cusp of resuming the rollout, and there is no end in sight.
After six years, financial management business transformation has only installed the momentum system in a few small corners of the VA, and its cost estimate has more than doubled to $5. $8 billion, according to the IDA.
Finally, the Digital GI Bill has missed its original 2024 completion date and its final price tag will likely double as well.
Not only have these systems blown past their schedules and budgets, they're also struggling to live up to the users' expectations.
However, there are signs that the VA's leaders are learning to take different approaches.
Secretary McDonough had the good sense to pull the plug on the dimble system after its pilot at James A. Lovell went so poorly.
The second supply chain effort has been on indefinite hold since the subcommittee exposed its poorly defined objectives and more than $9 billion lifecycle cost estimate.
Finally, Secretary McDonough and Mr. DelBene and their teams came to their senses before moving forward with a nearly $1 billion human capital management We're good to go.
I think?
The department literally cannot afford to operate this way anymore.
Mega projects are also incompatible with the best practices that took hold in the software industry more than a decade ago.
I'm talking about agile, incremental development, small -scale deployments, testing by real users, and component -based architecture.
We're going to hear about successful use of these practices from our panel of independent experts and hopefully from our VA witnesses as well.
We've also seen some encouraging examples when the VA succeeded in implementing or solving problems with smaller systems.
VA .gov was riddled with bugs causing veteran submissions to disappear and I still question how long it took for the department to tell us everything.
But Mr. DelBene in his office were able to get to the root of the cause.
The problems fairly quickly.
There's also the 18 -month ordeal to get the program integrity tool to process claims data correctly.
But the system is small enough that OIT was eventually able to figure it out.
Mr. Riffesey and his team recognize the limitations in the large, outdated Veterans Benefit Management System, and they have been proactively segmenting it and modernizing it in individual parts.
As with most things in government, the larger a project gets, the harder it is to manage and the less likely it is to ever deliver useful capability.
On the other hand, smaller efforts deliver early and often and they can survive the technical challenges along the way.
As we close out this Congress and prepare to transition to a new administration with new VA leadership, I want to emphasize these lessons.
Learned.
We need to build a nimbler VA that's more responsive to the veterans' needs.
I look forward to our witnesses' testimony today to help us do that.
With that, I yield to Ranking Member Surfless McCormick for her opening statement.
Thank you, Mr. Chairman, and thank you so much for our witnesses for being here today.
Over the course of the 118th Congress, this subcommittee held 20 hearings examining the IT modernization efforts of the Department of Veteran Affairs.
It is clear to me that most of the programs that were reviewed suffer from the same issues.
Poor requirements development, poor contract management, And an over -reliance on contractors to determine what is possible versus what the department identifies as its needs.
I'm also concerned that the department is taking on too much.
Instead of attempting to modernize everything all at once, they should be thinking strategically and delivering one or two successful projects before moving on to the other.
I worry that business -led IT projects are part of the problem.
It seems to me that OIT's role in many of these projects is not sufficient to keep the projects on track from the very beginning, causing monumental downstream effects.
This lack of involvement, whether intentional or otherwise, contributes to the poor requirements development, which in turn leads to the significant contract Overages we see in the projects like the digital GI bill and the supply chain modernization.
All for VA to realize that they aren't getting what they need out of the product.
I also have major issues with the way IT is budgeted at VA.
IT is frequently the sacrificial lamb when it comes to needing funding for the mission.
While this is not unique to VA, we cannot allow this to continue.
As more and more veterans become eligible for care and benefits at VA, budget constraints require cuts in other places.
It cannot continue to be one of those places.
This is the reason that VA is operating on an IT infrastructure that is well past its service life in many places, and much of the software that VA uses is obsolete.
We hear a lot of talk about government inefficiency.
Fixing those inefficiencies starts by ensuring that VA employees have the best tools to do their job, which starts with fully funding the Office of Information and Technology.
Point blank, VA needs to do a better job of ensuring that its IT budget supports VA employees and veterans.
I look forward today to hearing the witnesses' testimony.
Before I yield, there's a few people that we, unfortunately, would be saying farewell to.
This would be the last subcommittee hearing for Assistant Secretary Dale Benning, one of my favorite chairmen, Rosendale, and the Republican subcommittee staff, Bill Mallison, who has been excellent.
Mr. Mallison has been with the subcommittee since its inception and has been a valuable partner in our efforts to ensure that VA and its employees have reliable and modern tools to provide world -class care and benefits to our nation's veterans.
His insight and his thoroughness will be missed, and I wish him the best of luck in his next chapter.
Also, as in this way, with a changing administration, we'll be losing Assistant Secretary Dale Benning.
I know you and I haven't always seen eye to eye, but I have appreciated your hard work over the last past three years.
I would also like to save the best for last, which is wishing Chairman Rosendale well wishes.
As this will be his last hearing with the subcommittee, there are few places in this body as bipartisan as a subcommittee we have been on, and I thank the chairman for being one of those people and working with me so openly.
I wish you the best on your next endeavors.
Thank you so much and I yield back.
Thank you very much for those kind words and appreciation from everybody, really.
I will now introduce the witnesses on our first panel.
Ms. Carol Harris is the Director of Information Technology Acquisition Management at the Government Accountability Office.
Ms. Lynn Overman is the Executive Director of the Beck Center at Georgetown University.
Ms. Overman has previously served in the U .S. Digital Service.
And finally, Mr. Reynold Schweikert.
Is the non -resident senior fellow for Congressional Modernization with the Foundation for American Innovation.
Mr. Schweikart previously served as senior technology advisor at GSA and in top technology roles in the House of Representatives.
If you all could please rise and raise your right hands, please.
Do you solemnly swear under penalty of perjury that the testimony you are about to provide is the truth, the whole truth, and nothing but the truth?
Thank you, and let the record reflect that all the witnesses have answered in the affirmative.
Ms. Harris, you are now recognized for five minutes to deliver your opening statement on behalf of GAO.
Thank you.
Mr. Chairman, Ranking Member Sheriff Phyllis McCormick, and members of the Subcommittee, thank you for inviting us to testify today on VA's IT Modernization Challenges and Lessons Learned.
And I also want to take a moment to thank your outstanding committee staff and also best wishes to Mr. Mallison as well.
As requested, I'll briefly summarize our prior work on the department's efforts to acquire and manage its IT, as well as our work on critical factors leading to successful IT outcomes.
As you know, the use of IT is crucial to helping VA effectively serve our nation's veterans, and the department's investment in IT is substantial.
Over the past two years, VA has obligated over $21 billion for a range of IT products, systems, and services.
Unfortunately, the department's management of its IT modernization efforts continues to be high risk.
This morning, I'll highlight two key points.
First, we've made a total of 20 recommendations in recent reports to improve VA's IT acquisitions and management.
And while VA has concurred, none have been implemented thus far.
These recommendations address healthcare, Financial management systems, IT governance, and IT procurement.
For example, VA has faced long -standing challenges in its efforts to deploy modern IT solutions in two critical areas.
Its health information system, as well as its financial and acquisition management systems.
After three unsuccessful attempts between 2001 and 2017 to modernize its health information system, VA is now on its fourth attempt, known as EHRM.
However, in 2023, VA announced it was halting further deployments and instead prioritizing making improvements at the five sites using the system.
At that time, about 79 % of users strongly disagreed that the system enabled quality care.
In our 2023 report, we made 10 recommendations to VA in areas such as user satisfaction, system trouble reports, and change management.
Similarly, in 2016, VA established its FMBT initiative, its third attempt to replace aging financial and acquisition systems with one integrated system.
In 2021, we reported that full implementation of the new system was not expected until 2027 at a 10 -year lifecycle cost of nearly $3 billion.
As of July 2024, full implementation was moved to 2030 and lifecycle costs escalated to $7 .7 billion.
Our two reports made a total of three recommendations to VA on cost and schedule estimating and its efforts to manage risks.
If VA implements these 13 recommendations, along with the other seven we've made in the areas of IT governance, software licenses, and cloud computing, the department will be in a stronger position to course correct the major IT acquisitions it has underway.
And now to my final point.
VA should consider all available opportunities to ensure that its IT investments are acquired in the most effective manner possible.
We have previously reported on nine critical factors that have led to successful IT acquisitions across the federal government.
These factors include ensuring end users and stakeholders are involved in the development of requirements and in the testing of system functionality, that government and contractor staff are stable and consistent, among other things.
These factors are common sense.
But they are also easier said than done.
All too often we have found the absence of these factors or the partial implementation of them to be the root causes of cost, schedule, and performance issues found on troubled or failed IT programs.
Another set of leading practices applicable here is the use of iterative development for complex IT acquisitions.
Leading companies use this approach to rapidly develop complex IT products that are relevant and responsive to their users' most critical needs.
Iterative development essentially breaks large, monolithic IT projects into smaller, more manageable pieces so that you're delivering system functionality within weeks or months versus years.
The critical factors and leading practices I mentioned serve as the basis for many of our current recommendations to help improve federal IT management.
As such, VA can rely on them to help achieve successful IT acquisitions.
Moving forward in the two areas I noted, it will be critical for VA to fully implement our recommendations as soon as possible.
Doing so will position the department to more effectively deliver the systems and IT operations that meet mission needs.
That concludes my statement, and I look forward to addressing your questions.
Thank you very much, Ms. Harris.
The written statement of Ms. Harris will be entered into the hearing record.
Ms. Oberman, you are now recognized for five minutes to deliver your opening statement on behalf of the Beck Center.
Thank you, Chair Rosendale, Ranking Member Scherfelis -McCormick, and distinguished members of the committee for the opportunity to testify today.
It's my honor to share insights from my perspective as the Executive Director of the Beck Center for Social Impact at Georgetown University, as well as from my past experience in senior policy and delivery roles in the White House Office of Science and Technology Policy and the Domestic Policy Council.
For more than a decade, the Beck Center has led projects that have positively impacted more than 262 million people across the United States, working alongside governments to build a future with more opportunity and economic mobility for all.
Rooted in the values of our institutional home at Georgetown University, we are a network catalyst, a research hub, an advocate for policymaking in the modern age, and a training ground for tomorrow's innovators.
Technology modernization is central to our work.
The systems that veterans rely on to connect to benefits parents use to access childcare and seniors need to utilize healthcare must be robust, adaptive, and designed with user needs in mind.
Yet, federal technology projects often fall short, plagued by rigid approaches that prioritize process over outcomes.
These failures diminish public trust, waste taxpayer money, frustrate agency frontline workers, and prevent people from accessing services when they need them the most.
When agencies attempt to modernize, they often purchase static software, treating it like any other commodity, like computers or cars.
One -time purchases that simply work upon acquisition.
But software must continuously evolve to keep up with changing policies, security demands and customer needs.
Technology modernization is a continual process of addressing unmet needs, not a one -time effort with a defined start and end.
Too often, federal modernization projects are built in silos, scoped too broadly, and outsourced to vendors who require high -cost, multi -year contracts.
Those contracts are then often overseen by agency project managers who may not be technologists themselves, making it even more challenging to test vendors' deliverables and identify technical fixes when projects go off track.
The good news is that some agencies can and have shifted from this static project management model to what we call a product model that empowers internal agency digital services teams with full ownership of product development and avoids the one -size -fits -all solutions in favor of modular development that meets the needs of real people.
The product model, championed by my friend and colleague Jennifer Polka, starts with a small, nimble team conducting discovery sprints with the people who use the services to deeply understand their needs, And the service delivery challenges the software aims to solve.
By identifying high -risk elements early, such as whether a data integration will actually function effectively, the product team can test and address critical components of the product before investing significant resources in a full -scale solution.
A recent example of the product model in action is the development of the direct file tool for the Internal Revenue Service.
The 2024 pilot ultimately served more than 140 ,000 people with 90 % of users rating the tool as excellent or above average.
To achieve this, the IRS empowered their internal team with the right expertise, started with a limited scope and eligibility, developed the tool with deep user research, and tested with a small number of users.
In fact, the first test was with just one person.
Then they fixed what didn't work, tested again, User tested again, and then tested further.
This process saved users time, money, and resulted in extremely high rates of satisfaction.
What's more, the operational costs to the IRS to develop, test, and launch direct file, including customer service, cloud computing, and user authentication, were just $2 .4 million, partly because the U .S. Digital Service team was deployed at no cost to the IRS.
Simply put, agency product development teams are uniquely positioned to identify and address operational barriers.
This streamlines processes, reduces administrative burdens, and paves the way for more successful product deployments.
To ensure that federal agencies, including the Department of Veterans Affairs, can fully realize the benefits of the product model for technology modernization, I propose four recommendations outlined in my written testimony that Congress can work on with agency and White House leaders.
But perhaps the most important recommendation is to ensure that agencies streamline hiring processes to efficiently recruit and onboard digital service talent, Who can implement the product model, backed by flexible multi -year investments that support a people -centered,
iterative approach to delivering effective digital services.
By adopting these recommendations, Congress can ensure that agencies like the VA are equipped with the tools, resources, and flexibility needed to deliver modern services that are focused on the end user.
Parents, veterans, seniors, and more that meet the evolving needs of the American people, ending the cycle of costly technology failures.
Thank you very much, Ms. Overman.
The written statement of Ms. Overman will be entered into the hearing record.
Mr. Schweikart, you are now recognized for five minutes to deliver your opening statement on behalf of the Foundation for American Innovation.
Thank you, Chairman Rosendale, Ranking Member Shriflis, and members of the subcommittee.
Thank you for the opportunity to testify this morning.
As mentioned, I am with the Foundation for American Innovation, but previously spent.
A considerable time here at the House, beginning in 1995, to modernize a wide range of legacy and dysfunctional software.
I have extensive experience in technology modernization, but I'm not a VA or IT health expert, so let me delineate that.
When we think about technology at the VA, It should empower the staff to provide better care without limiting the ability of doctors and medical professionals to problem solve.
I want to briefly focus on three issues that I think would improve the outcomes at VA.
The VA leadership challenges, improving the electronic health record modernization implementation, And identifying options to better address future challenges in financial management and supply chain.
I think that the next secretary should define several organizational objectives for technology modernization and place those objectives in each executive's performance plan to help create a more unified focus from VA leadership agency -wide.
The second issue that they should address is the tenure of the CIO.
I think Mr. DelBene had the best possible tenure given transitioning from an important private sector career with conflicts of interest to manage, etc.
So his three -year term is about as long as that is going to be.
The next administration should take steps to ensure that the CIO has a tenure of at least five years, either with a term appointment or some other process to accomplish that.
When you look at the record of the electronic health record modernization project, it would appear to me as an outside observer that the process is very...
Stepwise.
The sixth implementation in Chicago was apparently the first one with a community care center and that process of moving patients from the care center to clinical care and back again had been worked out in the legacy software but was an issue in the implementation.
To avoid sort of repeated discovery, as this project is implemented across VA, I'd recommend that the deployment plan look at how to operationalize all of the functionality to discover what it is across all the centers.
I think the fact there's 130 instances of the legacy system reflects a broad divergence across the facilities, and you can't discover that one at a time.
The methods previously discussed are appropriate to do that, but a key milestone, I think, is reaching a point where all of that functionality is running successfully in at least one facility,
so it's not a hypothetical and it's not pretty PowerPoint slides.
Looking at the supply chain issue briefly, Sounds like in these repeated failures, these are these monolithic projects.
One thing I would note is that the problem of failing to deliver creates a negative feedback loop and it causes a last train leaving the station phenomena where everyone loads every capability and every complexity because this is their one chance in 10 or 20 years to get there.
Thank you.
Conclude my remarks and look forward to your questions.
Very good.
Thank you so much, Mr. Schweikart.
And the written statement, Mr. Schweikart, will be entered into the hearing record so that we have the full statement.
We're now going to proceed to questioning, and I will recognize myself for five minutes.
Okay, Ms. Overman, good morning again.
Thank you again for being here.
Can you think of a multi -billion dollar IT project?
Heads up that this was going to be a question and I will not say that I did comprehensive research across every single IT implementation in the federal agencies.
A successful project or two.
And it's interesting because I actually think it reinforces the points that you made in your opening remarks and that I've heard from my fellow testimony given here today.
It was a cloud migration that the Department of Defense managed.
It's called the Defense Dias.
It's an 8 .9 billion 10 -year time frame contract to migrate securely to the cloud.
I think the reason that this actually reinforces what we're talking about is twofold.
One is cloud infrastructure is really functionally infrastructure.
It is something that you can acquire and deploy.
It is not something that is aimed at meeting specific end user needs.
So that has been a successful implementation.
And it was interesting because the way that they rolled that out initially, starting in May of 2023, was to roll it out to 11 ,000 users to start.
Now, for some agencies, that might feel like a large amount.
I think?
Okay, thank you.
Mr. Schweikart, can you think of one?
I cannot, Mr. Chairman.
Thank you very much.
Ms. Harris, what is it about the IT megaprojects that makes them so risky and prone to failure?
Because when you take that waterfall approach, which is that monolithic approach, You're developing all your requirements in the beginning, and then you're designing and developing the system, and then once you deploy five to seven years out, that's when you realize it's not going to meet your mission need.
And so you've already sunk in at that point, you know, hundreds of millions, possibly billions of dollars, and you're just set up to fail.
It's just too much risk for these agencies to take on to go with that kind of approach.
Very good.
Ms. Overman, you have been an advocate for a product model approach that makes government agencies accountable for outcomes rather than a project model.
You touched on it on your opening statement that lets agencies pass the buck to contractors.
What are some examples of the product model and the project model so that the general public can understand the difference what we're talking about?
Yeah, absolutely.
I do think the examples that you provided of some of these six VA systems are very good examples of the project model and what the challenges are with that approach, and I believe Ms. Harris also just highlighted that.
I'm going to give a couple of positive examples of the product model.
You mentioned the IRS direct file, which was...
First time we've heard them mentioned in a favorable manner.
Purely from a software development and deployment standpoint, it was actually quite effective.
I do want to point to another project that I think actually is another component of why this type of an approach can be quite useful, which was the covidtests .org.
Thank you.
Thank you.
First of all, they designed with users in mind and made the application online as lightweight as possible.
So they did not try to pile on data collection.
They did not try to do identity proofing.
They basically made a very simple form that took less than a minute to fill out.
That did not actually On the back end, what they did was they worked with vendors who had demonstrated success in handling high load input.
So while the front end was very simple, the back end was intentionally designed to handle high volumes of applications.
And indeed, on the first day, a million people applied and the system was able to bear that load.
So I think that combination of thoughtful design, working successfully with vendors, and having a product vision that was owned by the implementers was what made that project so useful.
Very good.
Very good.
I'm going to yield back now and recognize Representative Kennedy for five minutes of questioning.
Thank you, Chairman.
Thanks for joining us.
Thank you.
Thank you, Chairman, and thank you all for your testimony.
Thanks for your service to our country.
Ms. Harris, in your testimony, you talked a bit about the VA's IT budget.
Do you feel the budget is sufficient to support the VA's IT needs and ultimately the modernization efforts?
Well, I haven't scrubbed the full IT budget to give you that full response, but what I can say is that when we take a look at these critical success factors for these large -scale IT acquisitions,
one of these factors that they have in common are having programs receiving sufficient funding.
It's very important for these programs to have that.
And so I do think that in combination with other tools like having a working capital fund with money that the CIO can use over multiple years at his or her discretion is very important to give them that flexibility.
But at the same time, accountability is also critically important because the last thing we want is for agencies to have these slush funds.
And we have open recommendations to VA to improve their cost and schedule estimating capabilities.
So when they come out with these programs, like EHRM or FMBT, for example, we want those cost estimates to be reliable and comprehensive and credible.
That's critically important.
So we have those open recommendations for VA to improve that so that when they do come out with a budget request, the underlying programs have reliable estimates behind them to back it up.
Thank you.
And as far as the VA IT acquisition efforts go, what are the biggest steps that the VA could take to improve?
Again, breaking down these large programs into these smaller, more manageable pieces is a huge thing.
That will definitely set them up for success.
Also, I also agree with my panelists that having end users involved... in that process early and often, especially for testing is vital.
We've seen success cases where agencies have literally built test labs where they bring in the users and they're working on prototypes and they're testing in real time, early and often.
That is also a great predictor of success.
Thank you.
To your knowledge, is there a single office or official responsible for tracking open recommendation from GAO or OIG related to VA's IT monetization efforts?
I'm not aware of that.
I'm not sure.
Do you think it would be beneficial to have a single entity?
I do think so.
I think a single point of accountability would greatly improve VA's track record in closing out our recommendations as well as the IG's.
And where should that entity fall within the department, would you say?
I'd have to take that back for the record.
I'm not quite sure where it would fall, but certainly someone that has direct reporting lines to the secretary or the deputy secretary so that they have that substantial backing behind them to ensure that these recommendations get implemented effectively and efficiently,
I think that that would be a good thing.
Thank you.
And in regard to the indicated risks associated with the lack of FITARA compliance, what do you think the VA's biggest weaknesses are with the lack of compliance?
Well, with regards to FITARA, I mean, one of the main things is for the CIO to Approve and review and approve all IT contracts going through the department.
And based on our most recent work in that area, it looked like roughly 39 % of those IT contracts were going under the radar.
So Mr. DelBene was not in a position to even be aware of those contracts coming through.
Having automated processes that ensure that those contracts hit his desk first so that he can review it and make a determination whether or not they are merited to move forward, I think that would be a good thing.
And we have an open recommendation relative to that.
And so as far as the VA's addressing these concerns and compliance, what is your suggestion?
Again, we have 20 open recommendations.
They should expeditiously implement those 20.
And in regard to...
I'll yield.
I'll come back.
Thank you very much.
Ms. Overman, why are the VA's megaprojects stuck in the project model?
And what problems does that create?
Certainly.
Based on my experience at the White House working on delivery and implementation and the work we do at the Beck Center, I do think the key problem is what has been indicated by my fellow panelists, which is this model of starting by trying to project over very extensive timelines what the ultimate end product is,
all its functionality, gathering those requirements.
Handing them off to a vendor and then hoping that, you know, five to ten years down the line, what is implemented or provided back, if it is in fact provided back, actually works.
I think there's a couple of fundamental challenges with that.
One, I mean, if you think ten years back, like cell phones, for example, how much have cell phones changed in just ten years?
How much have user needs changed over time?
The kind of fundamental I do want to emphasize,
I appreciate Ms. Harris's obviously extensive expertise in this space.
I do think a key component of this making the VA successful is obviously the more technical pieces of this, but having this internal capacity, these product development teams,
To oversee these projects so that there is a single owner or a team that is well -versed in agency mission and goals in these modern agile processes that own the product vision and then manage vendors on a much faster and more modular timeline is really a key to success to trying to solve these problems.
Very good.
Thank you, Ms. Everman.
Ms. Harris, are the megaprojects inherently...
Incompatible with some of your best practices.
What I'm trying to find out is where do we have these conflicts coming in?
It's like with your recommended best practices and the mega projects, the way that they are developing them.
Yeah, they're entirely in conflict with best practice and what we've seen both in the public space and the private sector.
We want these...
Thank you.
It's not a technical problem anymore.
What are the cultural...
So I think there are two things, Mr. Chairman.
One, the legacy projects that have been running, the approach hasn't gone back and taken lessons learned to approve legacy projects,
to update how they deal with legacy projects.
I think there's a...
So when you say the trust issue, break that down a bit for me.
From the people that are responsible for developing...
So from the end users, from the employees...
We're not going to get another delivery for an extended period of time, so let's go ahead and add everything, all our requests on right now.
That's right.
And is there a little bit of burnout?
Now we're on the fourth try and nothing happened, so why should I really participate because I'm busy?
Because this try is probably going to fail as well.
So why do I need to get involved?
And that's a management and a culture challenge.
I think the other challenge...
Is the use, build once, use many, to say, okay, we're going to build this, and there has to obviously be user input, but everyone needs to take it.
And secondly, the phrase from my CIO colleague at GSA, minimal viable product, is a shorthand way of saying, build just enough for it to be useful, get it into people's hands, and iterate.
But they need to do that enough, The people believe it's going to work.
They have to build that trust.
Very good.
Thank you very much, Mr. Schweikart.
I now yield back to Representative Kennedy.
Thank you, Chairman.
Ms. Harris, back to you.
Thank you.
As Ranking Member Sheriff Phyllis McCormick indicated in her opening, we are concerned about the business -led modernization efforts.
Well, I can tell you across the federal government, I've seen a mix of arrangements where it's business -led or it's CIO -led.
And I've seen...
Moderate success in both areas.
I think the critical key is for both to work in partnership together in close collaboration so that, again, those requirements are being fleshed out appropriately and that there's adequate engagement with stakeholders and end users.
That's really the key rather than who is leading the effort.
Do you think that OIT should have a bigger I think that OIT should be involved as early as possible in an effort that's led under VHA or VBA,
for example.
That they're involved in the acquisition phases so that they understand and that they get end users even involved in that.
Before we're even talking about contracts, so that requirements are adequately defined.
And I think OIT understands or has a better understanding of what best practices in IT acquisition are, so they should be involved in that process to be able to lead those discussions effectively.
Thank you.
Ms. Overman, in your testimony, you identified failing to conduct deep user research.
As a risk to successful technology modernization, I associate this with VA's struggle to do comprehensive requirements development.
How do you think an agency like VA could or should improve upon this?
I think you're hearing a lot of agreement here on this need for user research, and I think it has to be in combination with small start.
So start small, build a minimum viable product, test it with those users.
There's a hidden benefit to this as well, or one that hasn't been emphasized yet, is this will make for more successful longer -term development.
It also very early surfaces if something is not viable.
So you can actually stop projects that are not going to work before they get several years in and multi -billion dollars in.
I think it's also important to emphasize there's really two users that we need to test for.
The end users of the product, the people who are receiving the services, but also the agency staff.
Because the agency staff, I believe as Mr. Schweikart mentioned, how insanely frustrating it must be for people who are trying to serve veterans to have to deal with this clunky software over and over and over again, repeated trainings,
unsuccessful implementations that are getting in the way of the services that they're trying to provide.
So there's kind of, I can't imagine that everyone isn't frustrated by the current status.
The other value of starting small is that the implications of being unsuccessful are much less impactful on these end users as well.
Thank you.
You also indicated that you had thoughts on the budgeting and appropriations process
Sure.
I think, honestly, the thing that you're going to hear from me repeatedly is how important it is to have digital service expertise in the government.
And I think for two key reasons.
Ms. Harris has mentioned how valuable it would be to have OIT involved.
If they don't have sufficient staff capacity with this digital service delivery expertise, it's going to be very difficult for them to be actively involved in the number of contracts, excuse me, that the VA...
Thank you.
I think those are really the two key things.
And I do think the other benefit of the modular development, again, is that you're going to have a much more consistent spend over time as opposed to getting these kind of huge spikes around these large development projects where you think you're making a one -time investment, but as we've seen,
there is a big dump of money to start.
It doesn't work.
There's another dump of money, etc.
So over time, we genuinely believe that this product model approach is more cost -effective for the agencies as well.
Thank you.
I yield back.
Very good.
Very good.
Thank you.
Ms. Overman, you've worked with a lot of federal agencies.
Which ones do this well?
And what kind of cultures and management structures do they have?
You know, it's interesting.
As you mentioned, I worked with the U .S. Digital Service.
I think USDS is probably the best example I can think of.
And I think the reason for that is...
Thank you.
I think DOD is further along.
I think DHS is further along in having enough digital service capacity to do these things on their own without having to rely on that external support.
And I think the interesting thing about DHS is not only do they have a lot of very good and deep internal technology expertise, they also have a very large customer experience office.
More like the private sector does.
Exactly.
They focus on customer service.
Yes.
Very good.
Very good.
Thank you.
Ms. Overman, how do agencies get in trouble when they buy commercial off -the -shelf software, as the VA has done with the Cerner EHR and the CGI Momentum financial systems?
You know, it's interesting.
I think, again, I feel like a little bit of a broken record here.
If you start with end users from the beginning, it is entirely possible that a commercial off -the -shelf system could work.
But you have to be very careful in understanding exactly what those COTS systems are offering and does it fit with what you're trying to solve for.
There are some downsides to COTS regardless of the fit.
One, it really only works if the agency is capable of adjusting its process and workflow to meet the COTS system.
And two, in a lot of these systems, you end up paying for functionality that you don't actually need because you're buying the full package as opposed to having a customized response.
I do think, you know, generally speaking, what I have seen is you don't want to be absolute about anything.
There are certainly use cases where having a COTS system might actually be the fastest and easiest way to solve for problems.
However, if you are dealing with what a lot of agencies deal with which is a user base that is not common in the private sector or is not fully met by private sector technology and you need customization,
that's where the product model becomes very helpful because you have the internal experts who can guide We're good to go.
So you can't modernize it yourself.
So even if it's fit for purpose at the time of acquisition, three to five years down the road, you're now stuck with a system that you can't change.
Very good.
Thank you.
Mr. Schweikart, do you think that the electronic health record modernization program reflects an agile, incremental, or modular approach?
I'm not sure it reflects any of those.
I think the challenge is how to create an incremental or modular approach.
And if I could touch back on One of your comments in your opening statement, Mr. Chairman, about underinvestment in infrastructure, we talk about modular approach, but you need the plumbing or the architecture so that these different pieces can be installed easily.
You know, that the door frames are a standard size so you can buy another door or a different door.
And so, you know, with where they are with this system, With the electronic health records modernization system, can they fully validate the requirements with small groups of users?
You know, a test lab would be helpful, but in my mind, the key milestone is they've actually got all of the functionality implemented once, which means that the other centers can come and look at it and say,
so this is what we're going to go do, and we're not going to...
Mess around every time we come to another center and say, oh, we need to change this because I like vanilla instead of chocolate ice cream.
But they've actually done enough exploration and enough validation and enough demonstration, really.
In the real world, this really works.
And we can roll it out rapidly to the remaining centers.
Makes sense.
As a matter of fact, I am sure that you all can go back to the statements that I made going back as long as three years ago.
We want to have a fully functional system at one facility before we start rolling it out in others.
I've used my time, and I think that we...
You're all finished with this panel?
Okay.
This panel is excused from the witness table.
I want to thank you all so much for coming in today.
And taking the time to answer questions for the committee.
And we're going to go ahead and switch out the next panel.
I'm going to go ahead and introduce the next panel as everybody is switching up so that we can utilize this time as efficiently as possible.
I'd like to now welcome the witnesses on our second panel.
To the witness table as we're doing the changes.
First, from the Department of Veterans Affairs, we have the Assistant Secretary for Information and Technology and Chief Information Officer, Mr. Kurt DelBene.
Good to see you again with us, Mr. DelBene.
We also have Mr. Robert, Robert, excuse me, Arifasi, the Benefits and Memorials Portfolio Director, and Mrs. Sherry Waters, the Health Portfolio Director at the Office of Information and Technology.
Once we are all situated, just like every meeting, we've had you seated.
Now we're going to ask you to stand back up again and please raise your right hand.
Do you solemnly swear under penalty of perjury that the testimony you're about to provide is the whole truth, the truth, the whole truth, and nothing but the truth?
Thank you.
And let the record reflect that all witnesses have answered in the affirmative.
Mr. DelBene, you are now recognized for five minutes to deliver your opening statement.
And I understand that we have a little bit of extra time.
Time?
It's a little longer, but not much longer.
Yep.
Chairman Rosendale, Ranking Member Scherfelitz, McCormick, and distinguished members of the subcommittee, thank you for the opportunity to testify on our critical mission to build a world -class IT organization and support the critical work of the Department of Veterans Affairs.
I'm accompanied by two of my executive directors in OIT, Ms. Sherry Waters, Health Portfolio, and Mr. Rob Arifasi, Benefits and Memorial Services Portfolio.
Over the past three years, OIT has transitioned from a traditional IT approach to a modern organization patterned after high -tech product development practices.
This model aligns with our responsibility in delivering a broad range of products and services to stakeholders and veterans.
While we've made much progress, much work remains to be done, and the ongoing support of Congress will be essential in advancing this transformation.
We're good to go.
I think?
Additionally, OIT has enhanced focus and transparency in spending, developing a clear one -to -end list of prioritization of IT investments.
This aligns with FITARA to improve management and oversight of IT resources.
These efforts ensure that every dollar spent benefits veterans through improved services and support systems.
In the past three years, we've made significant technical strides in our major IT programs and projects, greatly enhancing the services we provide to veterans through focused efforts in cybersecurity, claims processing, digital experience, electronic health records, financial management, and education benefits.
The most critical technical priority is cybersecurity.
We have refocused the team from merely executing cyber as a policy and compliance exercise to a risk -based approach.
We've embraced a zero -trust strategy, achieving over 95 % enforcement of multi -factor authentication, encryption at rest, and encryption in transit.
VBMS has been a cornerstone of our efforts in supporting streamlined claims processing for veterans.
In 2024 alone, VBMS supported processing 90 % of the claims within established timelines, showcasing our commitment to creating a user -centered platform that empowers veterans.
To our efforts to improve the veterans' digital experience, OIT has successfully launched the VA Health and Benefits mobile application, and it's been downloaded 2 .8 million times and is actively used by 1 .2 million veterans each month.
We've also supported millions of monthly visits to VA .gov, ensuring veterans and customers receive the same information regardless of how and where they interact with VA.
The EHRM initiative has undergone a thorough reset since early 2023, focusing on integrating user feedback into systems design.
We're planning a restart of deployment efforts in 2025, ensuring that our electronic health record system meets the needs of our veterans and VA staff.
The rollout of IFAMs within FMBT has modernized financial management processes across VA. Enhancing our ability to manage our resources effectively.
As of November 2024, we have successfully completed multiple phases of the initiative, resulting in a more agile financial management system.
Our work on Digital GI Bill has resulted in technical advancements that ensure timely and accurate delivery of payments and benefits information for enrolled veterans.
This initiative exemplifies our commitment to innovation and reliability in delivering essential services.
While we celebrate these successes, challenges also lie ahead as we continue our modernization efforts.
Enhancing the technical skills of our workforce is vital in adapting to evolving cybersecurity threats.
OIT must also navigate the complexities of modernizing supply chain and HR systems.
Our efforts in modernizing VA supply chain management systems have made slow progress, and we're committed to leveraging a more deliberate, incremental approach.
Likewise, HR modernization needs sustained investment beyond the current implementation.
The enterprise human capital management modernization efforts aim to enhance the employee and user experience while ensuring information transparency.
Continued investment in OIT's major initiatives is essential to building progress we've made over the past three years.
These initiatives are crucial to fulfilling our mission to provide secure and effective IT services to support veteran care.
Above all, we in IT must continue to hone our craft, focusing on a clear vision, developing comprehensive roadmaps, and being relentlessly focused on operational excellence.
Investing in our people, our greatest asset, is key.
We've made great strides, but the work must continue into and beyond the next administration.
Chairman Rosendale, Ranking Member Shurflitz, and members of the subcommittee, thank you for the opportunity to present this testimony.
Our journey to become the best IT organization in the federal government continues, and I extend my deepest appreciation to the subcommittee for your oversight, your attention, and your unwavering support of our nation's veterans.
I look forward to your questions.
Thank you very much, Mr. DelBene.
The written statement of Mr. DelBene will be entered into the hearing record.
We will now proceed with questioning, and I will recognize Representative Kennedy for five minutes.
Thank you, Chairman.
Mr. DelBene and other panelists, thank you for being here.
Thank you for your testimony in service to the country.
Mr. DelBene, I have serious concerns about the size of the VA's fiscal year 2025 IT budget.
And I understand you do as well.
It's clear to me that the budget isn't just going to keep the lights on.
It's going to severely curtail many of the department's modernization efforts.
Do you believe that the funding for next year's levels are sufficient?
Thank you for the question.
I do not.
As I've talked about in the past, the budget is down in a significant way from the previous budget of 24.
But even a flat budget in this environment of growing cybersecurity threat and the need for modernization you talk about, that would not be sufficient.
So we have some of the key highlights I would probably flag is development is down 99 .2 % in funding there.
Staff and administration services are up 5%, but much of that is covered by a pay increase, so the actual net increase in headcount is zero, essentially.
Modernization efforts down 66 .5%, over 24.
One critical area is infrastructure readiness program, which is off 65 % in the current budget, and that is the rollout on a sustained basis of new PCs, new network infrastructure, etc.
That will take a significant hit as well.
FMBT, that program is down 63%.
So it's a maintenance budget?
That's probably the right character.
So what are the short -term and long -term impacts of this so -called maintenance budget?
Well, on a positive front, we are, as I talked about with our one -to -end prioritization and really focusing on execution is...
We are figuring out how to make it work.
But there's a few things that I would call out.
One, while we have a 20 % increase in cybersecurity investment, the transformation to a zero -trust architecture, which I firmly believe in, is an expensive operation.
And OMB put out a pretty comprehensive roadmap that they asked us to follow in terms of our implementation.
We will not be able to do that entire deployment.
We also have to make very difficult decisions about which programs to invest in modernization of.
I would also say that because of the good work that's been going on in the PAC tax and serving more veterans, The headcount levels have increased in the administrations.
That means we have to pay for more PCs.
We have to do more customer support.
And as a result, we are short in the field on the number of people that we need to actually support folks.
We're actually coming from a legacy of very good customer support.
Our users' satisfaction in the field is very, very good.
I worry that those numbers will come down as we are spread more thin.
Thank you.
And if you had the ability, what would you do to restructure the OIT budget?
Well, I think it just needs to be at a higher level.
I think there is an opportunity to have money that goes over multiple years.
However, I should point out at this point that the common theme in this hearing of incremental instead of what I call big bang development, and a lot of folks do, I fully agree with that.
And so...
What it means is not if the budget were larger, we would have super large expenditures in the programs.
It would mean we would have an adequate investment across modernizations, a broad set of modernizations that we would do in an incremental approach.
And so that kind of sustained budget at that kind of like the whole notion of an MVP followed by an investment on a yearly basis, getting us to a point where all the systems are what we call evergreen.
Which means you're always doing an incremental amount of improvement every year.
That's how I'd like to see the budget follow that structure as well.
I'll come back for more questioning.
I'll get back.
Thank you.
Thank you very much.
Mr. DelBene, good to see you again.
Thanks for coming before us.
The VA started developing the Veterans Benefits Management System in 2013 to digitize the paper -based claims process.
The system was a disappointment for many years, but in 2022, you changed your strategy and the VBMS has notably improved.
Could you explain how and why the VA changed its strategy with the VBMS?
Thank you for the question.
I'm very proud of the work that's been done there, and I will pass it to Rob Riffesey, who leads that program, but there's a couple of things I think in my perspective that I want to highlight first.
The first thing is when I came on board, but not as a result of my efforts, certainly, but kind of a focus around operational excellence, we looked at the uptime of the system and made changes in terms of scalability to make sure it would just not fall over as the demand was continuing to increase.
And so we have gotten VBMS to the point where it has a highly reliable, high uptime system, which is, I think, the first key.
And then the second key is just going to this notion of Creating a target platform for the modernization where you want to go, and then incrementally getting us there.
So we have this new platform called BIP, and we want to move everything to BIP, but you don't do it all at one time.
You do it incrementally.
But Rob, do you want to take it from there?
Sure, thank you for that question.
And really, back in 2022 and earlier, we could only release a version of VBMS about once a quarter because VBMS was so large.
There were so many pieces of it that needed to be tested.
And gone through and then fixed as bugs were found.
And it did not make sense to continue in this model in which we had to wait so long to get a release of VBMS out.
And we looked at the various strategies that we could take for being able to release VBMS faster and get the capability into the field, be able to test individual parts without having to rework the whole entire system whenever we found an issue.
And that led to the modular approach we're now taking with VBMS today, where we're able to work on it in very smaller modules that make up the parts make up the whole.
And we're able to deploy right now.
We're deploying every two weeks.
We're deploying functionality to VBMS.
And we have a very ready, agile release with VBMS working hand -in -hand with VBA to make sure they have the functionality they need.
Very good.
Thank you so much.
Mr. DelBene, in 2023, the VA was soliciting proposals for a $900 million project to replace its human capital management systems.
You recently abandoned this idea, and now you're considering a modernized in -place strategy.
Why did you decide against the megaproject, and what is your strategy now?
Well, I think there's two aspects to that.
The first is that the austerity environment that we find ourselves in, we're not going to be able to go out for a very large bid to move the entire HR system to a new platform.
I do think moving to a new platform is important.
The current system is an on -premises system and all of the major vendors are moving to a cloud -based platform.
I should note that the contract, if you look at it, actually had the notion of a minimum viable product in it.
So it basically said, you will create a pilot, it will do just the atomic functions necessary, and we would only pay for that first milestone before we went more broadly.
But with fiscal austerity as it is, we are moving to a model that says we want to find, kind of similar to how we're doing with VBMS, find that new platform we want to move to, figure out, and it will be a cloud -based platform,
figure out how we would take a small modicum of functionality from the HR, I think of it as a process flow from you're recruited to you're off -boarded at the end of your career, we'll take a slice of that, we'll figure out how we would move that to a modern platform,
and then we'll figure out how to move the other pieces over time.
That will change the notion of funding as well to be a very incremental thing over time.
And we will come to you in the future once we have figured out the strategy of like this is how much we will need in each year to do that continuous modernization.
Very good.
Thank you, Mr. DelBene.
Earlier this year, the VA was planning another mega project to consolidate 63 supply chain systems at a cost of over $9 billion.
Why did you decide not to move forward with this project, and how is the current strategy better?
Tell us the improvements on that.
Yes, so early on, Demos, as you mentioned, I believe, was the target environment there because it was being used by the DoD, and also felt that sharing catalog of inventory elements would be a good kind of synergy across those two organizations.
I think we lacked a clear understanding of what the gaps in the systems are today from an end -user perspective and what an incremental approach might look like to improving our supply chain's approach.
And I think we're stepping back and looking at that now.
But let me pass it to Cherry Waters, who's been involved in that project.
Thank you.
Thank you very much for that question.
I'm actually really excited to tell you about what we're doing in the supply chain initiative.
Very much like what we heard about from our first panel, we are having a very end -user focused approach and we're doing an iterative strategy.
We recently brought a group together and we looked at what are the most challenging problems that we want to resolve with this system.
Then we had teams take them back and evaluate all of those top 24 that were identified from a people, process, and technology perspective.
And by looking at it in that way, it helps us begin to prioritize how we get quick wins so that we can deliver a minimally viable product, improve on products that we already have in place,
and work with our partners to improve processes at the same time.
Very good.
Thank you.
I yield back.
I will now recognize Representative Kennedy again.
Thank you, Chairman.
Mr. DelBene, I'm concerned that the VA I share the concern there,
and we've worked closely with the administrations to come into a better partnership there.
The authority I have and the leverage points that we've really tried to drive have been around the authority to operate process, which says if you're going to be on the network,
we have to review specifically from a cybersecurity perspective, but we can broaden that approach as well.
And then FITARA review as well.
And I think the points that were made around whether everything goes through the FITARA process is a really good one.
I think the GAO work there was very good to survey particular projects and see how many do and don't.
We have been working within our Office of Strategic Sourcing in OIT to put processes in place to make sure more and more goes through the FITARA process.
I am a big fan of the FITARA process because it gives us the ability to review everything as it comes about.
At the same time, the capabilities to do the technical review of these projects is not strong within the administrations, and I think they would recognize that as well.
We have reached out to those groups and said, look, we want to play a bigger role in doing the technical oversight of those projects.
In most cases, there's a receptivity there.
I think the work in particular in Digital GI Bill has been good, where we said, hey, let us get more involved.
And how would you restructure these efforts going forward and how do you think it would improve the VA's success?
My biggest concern overall, bar none, is cybersecurity.
So, you know, we, the sacredness of the data that we hold is held in trust and we have to make sure that we don't get a breach.
So the first thing we are pushing for is our policy 6500 requires that people purchase software that is cyber secure.
And we're holding people accountable to that.
That means that you have to have technical resources on your program, even if it's a business or an IT program, to actually assure the cybersecurity of it.
That is not necessarily...
It's our goal to review that.
It is not our responsibility to assure that the development that you went through is a secure one.
The second thing we're doing is we're...
And we're doing this really, I would say, at a pilot level right now.
Is we're engaging in drafting an operational agreement with the administrations that says, not just in cybersecurity, but in operations as well, we will do this, you will do that, and we will work together in these areas and in this way.
We did a lot of this in the commercial sector when we had, for instance, the Windows team working with the Office team delivering things together.
You basically have to come to a common operating model that says these are your responsibilities, these are ours, and this is how we'll work together.
My first approach there, which has been very iterative, was just to clarify by virtue of a memo and a policy what the responsibilities are, but we quickly found that it's much more complicated in a back -and -forth thing,
so this operating agreement principle.
Thank you.
I yield back.
The whole notion that you can determine exactly the cost of a program,
the precise requirements for a program of large size from the beginning is just very fraught.
Once we get involved in a project, things evolve in terms of the requirements, new requirements emerge, and the technical complexity of the project invariably increases.
And so this notion of building a minimum viable product that you can build upon and build success upon success, it basically allows you to hone what you think you're going to deliver more broadly.
But it also builds on success from people's minds as well.
They see it roll out.
They see the initial difficulties.
They see it getting better.
And they start getting excited about the project as well.
And I completely agree with the sentiment of the first panel that there are very few successes where a big bang approach has been successful.
And this is true in the commercial sector as well.
And you've had the unique experience of being able to operate on both sides.
Yeah, we've had our share of big bangs in the commercial side as well.
Thank you.
Mr. DelBene, you've clearly learned some lessons in VBMS, HR, and the supply chain project to move towards more agile, less risky approaches.
Where else are you doing that?
Where else are you implementing this?
Well, ironically, I think where we are ending up in the EHRM program is an incremental approach.
If you look retrospectively, We have established six locations where it is today.
With the pause, we are resetting to getting it right.
And in a sense, hindsight is 20 -20, but you probably should have done that from the beginning.
So the scope of the project from the beginning should have been this first pod, and then you would have broadened it beyond that.
I think it's the model that we're using everywhere, though.
I think even if FMBT, in a sense, I mean, it's been talked about as Big Bang.
It's not really.
It is get the blueprint of the financial transformation right.
Start with some of the core offices and add VBA.
And now the thing that we're going to have to do in the future, which we're just doing with the central office in this first VHA central office first, is the rollout to the visits.
And that will be complex.
But I don't think it's completely accurate to say it was a big bank project.
And the other thing I'd say is the nature of ERP projects in general, which is what that program is, it is a big commitment.
There is no way to absorb half an ERP transformation.
I did part of our transformation at Microsoft in my previous role, and there we were consolidating 34 different ERP systems into a single global system, but the target was one single large global system.
You just need to do it in an incremental fashion.
Very good.
Mr. Albany, six years into the FMBT and EHRM systems, is it too late to change them wholesale to make them more agile?
As I said, with FMBT, I actually think it is pretty agile.
I think the challenge there is it's going to be an expensive program.
It is an essential program.
A successful organization has a strong financial management system and has been noted.
Ours is 30 years old.
It is time to do FMBT, but it is going to be an arduous process to get it done.
I think it is important to go in the approach we've taken.
I think the next thing they're going to do is to take a particular vision.
And use a willing vision that is excited about the project and getting that one right.
They are focusing on process transformation that has to happen in concert with the actual rollout.
So that one I think is going to work.
But it's going to take time.
On EHRM, as I said, we just need to get the criteria right.
We need to get the situation right where we are, which is what we've been working on.
And then at that appropriate time, we'll do the planning to move to another set of And Lovell had the additional staff available to them to give that support,
which, again, when we start looking at the total cost, that really has been a big part of the problem.
We don't have just the overrun.
Well I...
I think we're already seeing some of those.
So, for instance, in FMBT, the success of the project so far has been good, and the uptime of that system has been strong as well.
So I think in many ways, FMBT, although a very large project, which is not an easy project to do, has been fairly successful.
We are seeing on the EHRM side greater stability and uptime of the Oracle system.
They are meeting the requirements around uptime, on incident free time, on end user responsiveness as well, so those are good signs.
I think the harder work there, and this is ironic to say when technology is so hard, are the process aspects of that.
Getting the system to meet the processes.
And there is, and I think this was spoken to before, there's a lot of diversity in the processes even across the business.
And so another thing that's going to lead to success is that process standardization, that role standardization with EHRM.
And that's been a big part of the work as well.
I think, as the Secretary said, you should expect to hear from us about a restart.
Within the fiscal year 2025.
Mr. DelBene, why is it so difficult for large government agencies and the VA in particular to implement these megaprojects successfully?
That's a very good question.
I think megaprojects are difficult to do any time.
I think it is probably the industry minimizes the complexities involved when they...
We go out and ask for a new system, which probably we should say we will only want the MVP.
People on the industry side said, we've got the solution for you.
It'll solve all of your problems.
And that probably sets the expectations of folks too high in terms of the simplicity of doing it.
I think cost estimates are probably part of the problem, that they tend to underestimate the cost.
To your earlier question about have you ever seen a program that didn't overrun cost, Part of that may be as you get into the complexity, things get harder and more expensive.
But part of it is the underestimate of the cost estimate in the beginning.
And so it is independent lifecycle cost estimates almost invariably increase over a period of time as more is known.
But I think it's tempting to say, oh, this seems easier than it is.
And you just have to be very wary.
These are very, very complex projects that have deep...
So, from my view, It seems just like, quite frankly, bad legislation.
You take a piece of legislation and you continue to pile everything on it because everybody thinks it's the last opportunity to get a bill passed, and what you end up with is a really big, ugly, unworkable piece of legislation.
That's the problem with this city, quite frankly.
And so, using that in relation to... to software if people think this is the opportunity that we have to fix then everybody starts trying to throw their piece into there as well and it becomes just an unmanageable great big mess instead of listening to the experts saying We'll take a component of this.
We'll get this.
We'll make sure we have that right.
And then we'll take the next piece and build upon that.
Mr. DelBene, you inherited EHRM, FMBT, and all the other megaprojects, along with the decisions that were made before you arrived, which stinks, but again, it's the nature of the beast.
If you could go back in time and start over with one of them, how many of those would you have designed or started differently?
Well, the critical question on EHRM, as we've discussed in the subcommittee, is modernization of VISTA versus building of a new system.
There are real values to building from a commercial system, but it is a very difficult debate that you could have debates on both sides of it.
And there have been failed attempts to modernize VISTA.
This whole notion of getting VISTA into an incremental modernization.
Where it's evergreen is kind of what I would have, had I gone back, that would have been kind of the leading candidate on the improving VISTA side.
I was not involved in the evaluation.
There's only a few commercial vendors that the VA could have gone with on the COTS side of things, so I don't have that much visibility into how that selection was made.
But there are real merits to going to a commercial product as we've done.
I think we are over the hump now.
And those merits outweigh the whole re -looking at or re -litigating this question of modernizing Vista.
We do need to continue to modernize Vista because it's going to be with us for quite some time still.
But I would have probably, given where we're at, certainly now I would continue with the rollout of Oracle.
FNBT, again, it is by nature an incremental effort.
I think they actually probably took the right approach.
The one thing that's interesting there is the choice of a system that is very federal government specific, which is what CGI system is.
But that even makes sense because they do have a very large market share in the very unique aspects of what it means to be an ERP system for federal government.
So those kind of make sense.
The supply chain one and the HR one, I think we should have been incremental from the beginning.
And I definitely would have taken that approach.
Very good.
Well, I'm glad to see that we are starting to get our arms wrapped around some of these and taking smaller portions of that and trying to get them functioning or taking facilities, individual facilities, and get it to work.
As I say, if you talk to a vendor, a contractor, an engineer, we can build a bridge from here to Europe.
Is it possible?
Absolutely.
But how much time and how much money is it going to cost?
So any of these things, sure, as far as a megaproject, the engineer, the developers, they will tell you, sure, we can do that.
And they can eventually, but how many challenges and what is the cost of it going to be along the way?
So I would urge your successor to be skeptical of the contractor's promises and hopefully stay...
On the same path that we have all worked together here to get us going down, because it has started making some incremental improvements.
I guess everybody is done.
I have no rancor to give me any closing remarks, so I want to thank the witnesses for joining us today and both panels.
The potential benefits of modernizing VA's IT systems are undeniable.
With the right IT strategies, the VA will deliver health care and benefits to veterans faster and better.
There's no doubt about it.
But if the VA's leaders keep repeating the failed ways of the past, they will burn through billions of dollars of taxpayers' money, only to be left with dysfunctional systems that do nothing to improve services for veterans.
Today, the ranking member and I laid out the subcommittee's position.
Based on all the evidence we've gathered during the past two years for how the VA should be approaching the IT modernization, the VA should not be doing IT megaprojects.
They just don't work.
And they always turn into black holes sucking billions of dollars of taxpayers' money into the contractors' pockets.
Put together, EHRM, FBMT, and the Digital GI Bill are conservatively Going to cost the VA at least $27 billion more than what they were initially planned.
No organization, and that includes the federal government, should be allowed to operate that way.
This is just not our opinion.
Our expert witnesses explained the pitfalls of buying or developing software without a clear understanding of the organization and its requirements.
They know that once a faulty project is set in motion, it's very difficult and expensive to restructure it.
And I appreciate them pointing out that sometimes the best course of action with a project is just simply to cancel it.
And we've seen that, and I appreciate it.
Our VA witnesses also understand this.
We've discussed some examples like supply chain and human capital where Secretary McDonough and Mr. DelBene have wisely stepped back and reassessed the requirements and restructured the efforts in smaller, less expensive,
less risky pieces.
We've also seen the VA begin to convert VBMS into a modern modular system.
This is encouraging.
But as long as the VA leaders continue to answer the siren song of megaprojects, I'm afraid that we will see more poor results, overspending, and budget problems.
No project should be too big to fail, especially when it's failing because it is too big.
Finally, I would like to thank Bill Mallison, his staff, Sam, I understand he's going to be stepping up.
Leanne, everyone who is behind us that has helped us do this work.
Representative Shurfless McCormick, the GAO for all the information that they have helped us with.
Secretary McDonough, I have truly enjoyed working with him and the VA and all of us working together to identify The problems that we are facing and to make some incremental steps towards resolving several of the big issues which have plagued the VA for years.
Have plagued the VA for years.
And I will tell you that it's been my honor to work with you and to serve on this committee.
With that, I ask unanimous consent that all members have five legislative days to revise and extend their remarks.
Include extraneous material.
Without objection so ordered this hearing is concluded.
Mr. Silverstein, author of The Rising, shares stories about the rebuilding of the World Trade Center complex following the attacks and discusses the business, political, and engineering challenges he faced during his 20 -year rebuilding effort.
I said, it's got to be replaced.
Because if you don't, Lowman Hat is going to become a ghost town.
People are going to leave it.
They'll never come back.
I said, secondly, if we don't rebuild it, we're going to give the terrorists exactly what they wanted.
I think this is an attack.
Not on the Twin Towers.
Nothing like that.
Much more serious.
It's an attack on America and everything we stand for.
So we have an obligation to rebuild it.
Larry Silberstein with his book The Rising.
Sunday night at 8 Eastern on C -SPAN's Q &A.
And you can listen to Q &A and all of our podcasts on our free C -SPAN Now app.
Attention middle and high school students across America.
It's time to make your voice heard.
C -SPAN's Student Cam Documentary Contest 2025 is here.
This is your chance to create a documentary that can inspire change, raise awareness, and make an impact.
Your documentary should answer this year's question: Your message to the president.
What issue is most important to you or your community?
Whether you're passionate about politics, the environment, or community stories, Student Cam is your platform to share your message with the world.
The deadline is January 20th, 2025.
For over 45 years, C -SPAN has been your window into the workings of our democracy, offering live coverage of Congress, open forum call -in programs, and unfiltered access to the decision makers who shape our nation.
And we've done it all without a cent of government funding.
C -SPAN exists for you, viewers who value transparent, no -spin political coverage, and your support helps keep our mission alive.