Busting IT modernization myths

FCW recently gathered a group of IT modernization leaders from agencies large and small to discuss the myths and misunderstandings they most often confront. The discussion was on the record but not for individual attribution (see below for a list of participants), and the quotes have been edited for length and clarity. Here are seven myths the group focused on busting.

Myth 1: My current system is good enough

Although there is certainly pressure from the top to modernize in general, the owners of specific systems and business processes are often loath to change what works. And frequently, participants said, those stakeholders are oblivious to the extreme measures the IT team is taking to keep those systems running.

"Often they're working through the heroics of the IT organization and kept running," one official said, "but in some ways, we do ourselves a disservice by hiding how difficult and how fragile [those systems] are. Communicating that, I think, is important to trying to get some buy-in to modernize in the first place."

Calling attention to shaky systems is a delicate task, another official said. Program owners might question why such risks have been papered over, while the IT teams keeping the systems alive can interpret the call for modernization as a critique of their efforts.

"There's a real sense of professional pride," the official said. "I'm not sure we're doing a very good job, but we've got to figure out how can we lift some of that burden from them and not feel like we're stomping on their graves. People have been struggling and through heroic efforts have probably hidden the risk and the lack of performance from their customers. It's one of the things we're struggling with."

That's not to say that every system must be reinvented, however. One CIO warned against modernizing for the sake of modernization: "I've got a lot of old code running on brand-new hardware that's secure and safe. I'm not going to fight that battle."

Myth 2: IT modernization is about technology

One of the biggest myths of IT modernization, one CIO said, is that it will "fix all of the decades-old problems the system has" without any need to reinvent the business practices that system supports. "If you just modernize, you're just taking 10-year-old garbage and shining it up. There's this assumption that if we modernize, or if we put everything in the cloud, these 14 different silos will now magically be integrated."

In government, another official added, a lot of inertia must be overcome before the conversation can turn to IT solutions. "It takes so long to buy things and the contracts are so complex and prone to protests that we tend to reuse contract language from previous contracts to try to reduce risks, but that results in the same wording getting carried on generation after generation." Reusing contract language "actually hampers our ability to do some of the modernization."

"I think modernization really needs to be about the people," a third official said. "It's presented that it's about technology and it's this one-shot investment…. We'll move legacy stuff to modern infrastructure, and we'll be able to pay that back, and we're done. That's completely divorced from the reality, which is that it's a continuous, in-perpetuity challenge."

"We always have new requirements," the official continued. "We always have new technologies that we need to deal with. If we're not bringing the people along, then we fail."

Myth 3: CIOs can own the modernization process

As long as IT is seen as the cornerstone of modernization, CIOs will be seen as the owners of those efforts. And that is a recipe for disaster, participants said.

"A lot of us are saying this is a business transformation," one official observed, "but I would poll the CIOs around the room and ask them who's really on the hook for the modernization of your agency. Is it the CIO? The answer is probably yes, and that's not the right answer."

It is a business transformation, the official added, but nobody is held accountable except the CIO. "He or she cannot transform or modernize without significant influence or buy-in from the business, which we don't really have."

When CIOs do seize that role, a different risk emerges, another participant said. "When you do that in the absence of a mission owner and they look at it six months later and say, 'That's not what I wanted. I wasn't involved. How did you decide to do that?' — you just wasted a bunch of time and money."

"There is fundamentally an issue where the business administration people think that 'Yeah, modernization or technology — that's not my job,'" a third participant said. "The reality is there isn't a single part of our space today where even the most senior business folks have to understand the impacts of technology to what they're doing every day."

Myth 4: Modernization is too expensive

Costs matter, the group agreed, but too little attention is paid to the costs of not modernizing.

Too much of the conversation focuses on the money that will be saved by modernizing, one participant said, and "we have to say, that's the wrong discussion. It's going to cost more. We've been cutting costs for so long. We cut our software teams. The technologies that we're running on have been out of support for four years. If they don't invest and actually spend more money on it, we will go out of business from an IT perspective. Systems will shut down."

"It's a matter, at this point, of buying off that technical debt," another agreed. "We have cut costs to the point that we've accumulated so much technical debt that it's putting the mission areas at risk."

One participant said framing the discussion that way worked at his agency. "We wound up getting more modernization money than we were asking for. But it wasn't because we went in and said, 'We're moving to the cloud, and we're going to reduce costs.' We went in saying, 'We're going to move to the cloud, and we're going to cost you more because you'll actually have better capabilities that will be stable.' So it was entirely about the business of outcomes."

The Technology Business Management framework can be helpful in those efforts, several participants said. "We use TBM, so even though I own the IT budget, I'll tell everyone exactly what their costs of running their operation is," one CIO said. "So we've got the CFO wanting us to do some cloud work because we've shown their internal hosting cost. I allowed them to make the best decision."

Myth 5: The mandates always make sense

IT modernization has been a White House priority for years now, and the group agreed there were benefits to that top-down focus. One participant, who came to government after years in the business world, said many of the mandates reflect "what are generally accepted best practices if you're working within the private sector."

Others, however, complained that the myriad policy memos can send conflicting messages. "The strategy is not clearly connected," one said. "Also, most of these mandates come and they say, 'Do it now.' The challenge is we have a two-year planning process. Shaking that up and accommodating this mandate is a big challenge."

Myth 6: Central funding is the fix

Although it is not a mandate per se, the group cited the Modernizing Government Technology Act's Technology Modernization Fund as a prime example of top-down efforts that don't make sense for every agency.

"Our biggest challenge is the payback period," one participant said. "There's a lot of hesitation in terms of when you've got to return the money back to the TMF. I think it was a little aggressive for the reality that the CIOs and the business owners are aware of."

"If it was extended over a longer period, I think you would probably see a lot more adoption of it," the official added. "And again, it's great that it's there, and in some certain use cases, it works well, but in others, I think you're going to see less adoption."

Two other participants said their agencies were statutorily prohibited from using the TMF, and a third was even blunter. "Our budget director has pretty much said, 'Over my dead body,'" that official said. "If you get caught short, you're now basically having to pay that out of your program funds, and it's just completely untenable."

A fourth official added that TMF "has the risk of exacerbating the haves and have-nots of our federal government." Some agencies are well positioned to tap the revolving fund for projects that can show a return on investment within 24 months, while others are not. "But they still need modernization, and so they can't even go after those funds."

Others, however, said they were grateful for the potential funding. "We want to take advantage of the working capital funds and convince our CFO to really buy into that model," one said. "But we're having a tough time getting the buy-in."

Myth 7: You know what you're getting into

Technology cannot be the sole focus of modernization efforts, but it is a key component, and many participants lamented how frequently long-hidden IT problems surface to complicate their efforts.

"One of the misconceptions is that the legacy systems even work," one said. There are times when a full-blown business process re-engineering approach does not make sense, and "you just have to reach back and refactor apps." But in one such case, when the goal was to rewrite an ancient Java app "and just give us the junk we have today in a modern platform that we can modernize later," the team discovered that more than 60 percent of the legacy code didn't work.

"Over the past 15 years, they had done attempts to modernize it that failed, and they left the code in that didn't work," the participant said. Of the 400 screens in the app, only 17 actually ran. The others "weren't used and were never put in place, but specific functions from those were called and tied in and used. So we didn't know what pieces of that code did work and what didn't."

The official added: "What we thought was a grad school-level project — take this codebase and just rewrite exactly that same thing over here in this other language — how do you have them test something where three-quarters of the lines never worked in the first place?"