Anil Cheriyan, a former bank chief information officer and chair of the Technology Business Management Council, has been director of the General Services Administration’s Technology Transformation Service, or TTS, for a little more than 100 days. In that time, he’s been listening and observing and says he believes the relatively small teams at GSA’s innovation hub can punch above their weight class if given the opportunity.
In a wide-ranging, exclusive interview with Nextgov, Cheriyan talks about his experience moving from the private to public sector, what it will take to ensure TTS has a broad impact across government, whether it should exist as an independent business line or be tied to acquisition, the significant evolution for the future of the Centers of Excellence and how to improve longstanding programs like 18F and FedRAMP.
The transcript has been edited slightly for length and clarity.
Nextgov: Did you ever think you would be a federal employee?
Cheriyan: No, I didn’t. I had retired from Sun Trust back in early 2018 and I was just spending my time on boards and doing some advisory work. Sometime toward the latter part of last year I got a phone call from someone in the White House saying, “Hey, do you want to do this thing called TTS?” At which point I said, “I have no idea what TTS is and I’ve never worked in the federal government and it probably doesn’t make a lot of sense for me to do it.” And the individual said, “No, no. You should think about it and consider it.” I said, “Send me some material,” which they did. I started reading about it and talked to a few people and slowly got more and more interested in the mission and what it was all about and I felt the mission of TTS is really something that’s as big as you want to make it. It’s a huge opportunity. What I call “big canvass” that we can create.
So, I got hooked. And if you think about it, I spent 35 to 40 years in the private sector and a lot of that in this country and this was an opportunity to give back and do something.
I’m not naïve. I don’t think I can come in here and change the world in a few years. It’s more helping the journey move forward. Helping the things move ahead. And if I move things forward a little bit, the good thing about doing it in the federal space is it’s a large canvass, so you can move things for 360 million people and move it for 3.5 million federal employees. So, a little change is worth a lot.
That’s what I’m hoping for: Get some momentum going in that direction.
What’s your impression 100 days into being a federal employee? How does the experience differ from in the private sector?
There are a lot of similarities, in the sense that you’re doing things with a fair degree of change. What I wasn’t aware of is the level of intensity of focus on change from the administration and, not just the administration, but everyone in the government in terms of IT modernization. The case for change seems to be reasonably well-made, and I didn’t believe that would be the case, having not done anything in government. That was an exciting, good thing to hear and see.
The organization TTS, as well, is full of high-energy, capable individuals. Some are term employees, coming here and spending time with a sense of mission, doing something for government. I didn’t think I would find that. The traditional view when you’re outside of government is everyone moves slowly and no one’s interested in changing. So, when I found that I was pretty excited because there’s a team that’s really driving toward making things happen.
There are perceptions out there about the people who make up the government, but there are also perceptions about the system being resistant to change. You manage a shop focused on innovation and say you have people who want to make change happen. What kind of barriers are you running into from the system itself?
Even in private industry, when I ran technology and operations for the bank—at Sun Trust—there were a lot of barriers to change. There’s cultural barriers: individuals who really don’t want to make the change happen. There are individuals who, even at the bank, who say, “This too shall pass. We will watch another person come in and want to make the change.” That’s human nature and it’s there. So, there are those cultural barriers that I’ve found working at Sun Trust over the years.
There’s a lot of opportunity to say, “This is what the law says we have to do,” and, “This is legislated this way and we can only operate in this manner.” So, there are those types of barriers that exists that, frankly, have not been challenged. I’d say the same thing if you’d interviewed me at Sun Trust. There are lot of people who’d say, “Well, audit wants us to do this,” and, “Legal wants us to do that,” and “You know we can’t make these changes.”
So, there are those barriers that people put up. Not that there’re all fictional. There are some of them that are true and need to be dealt with.
I think the ability to drive the transformation is not to come in and say, “I have a smarter way to do things and I will therefore teach you how to make the change.” I think that’s a no-win situation. The approach that I’m taking is more a: Let me listen, let me learn, let me bring the right ideas, let me bring the right team and put the right people together. And I have a firm belief that if you do that, you’re going to drive the change.
And it’s driven by more of a fact-based approach; it’s driven by more of a buy-in from individuals, who then really buy in to you, as well. Coming in with the right approach, I think you can get over most barriers.
Now, there’s always things that are bigger than you and things that are bigger than, frankly, GSA, and things that are bigger that need to be addressed. And hopefully those are all things people are working toward. This is not a sprint; this is a multiyear journey that will take time.
If you try to tilt yourself against the biggest and hardest thing, you’ll never get anything done. But if you start making the right decisions, start making the right changes, start building momentum, before you know it you have that flywheel effect of really driving change and some of the bigger [barriers] will start coming down at that point.
Does TTS have all the resources it needs? Do you need more people, more funding, more top-cover?
Resources are not a real obstacle. There’s always a battle for talent. That would be the case across the board: Whether you’re at TTS or anywhere, there’s always a need for talent. I think that we have done a reasonably good job at TTS so far in using innovative approaches to bring the right people. Whether that’s the right mix and so on is something that we can address.
I think the funding issue: We’re very much cost-recoverable in the main; very much a cost-recoverable group. Agencies will pay for value and if you deliver value and you demonstrate that value, I don’t think there’s a real funding issue from that standpoint.
What issues have you found? What have you learned and are there any immediate changes that need to take place?
Let’s talk about the good. The good is a really innovative team, a really high-energy team, a real culture for wanting to do the right thing from a mission standpoint. So, a really good base to work from. And a good momentum: A lot of the CoE work being done, there’s a fair degree of momentum there; the 18F team has really driven a lot of change and now are more cost-recoverable and driving real value.
But there are areas that need to be addressed. I believe that while we have good, strong subbrands like 18F and the PIFs [Presidential Innovation Fellows] and the CoE brand, we are not punching above our weight, really delivering impact at a level at which I believe we can. There’s a lot of good work. But is it really being done with the impact that is meaningful when you think about the overall mission of what TTS is and what they’re trying to do?
What level of program or impact do you think would be reaching for the stars but still within reach?
Take the CoE approach. We have an approach which is more top-down, transformational, agencywide.
Take the USDA transformation that’s going on. It’s a close partnership with [Agriculture Secretary] Sonny Purdue; with Steve Censky, the deputy secretary; with Gary Washington, the CIO. Really top-down driving changes that are connected to what the secretary of USDA wants to have happen: Consolidating to one call center, really driving the client experience, building the dashboards.
So, that’s the kind of a top-down approach that is working, where you can really say, “Here’s the impact. This is the real value and here are the millions of dollars we’re going to be saving on data centers,” for example.
Whereas, when you’re doing a lot of smaller projects that are at the individual program level, showing that impact is harder. Especially if it’s deep down in the bowels of an agency with an individual who might be three levels down from the CIO.
Not that they’re not important. They’re good projects. But translating them into real impact—what is the real outcome, what does it mean for our citizens, what does it mean for our federal employees—those are things that we need to punch above our weight, as it were, to really show that impact.
It doesn’t have to be big programs. It can be smaller, targeted projects. Small doesn’t have to mean small impact. You can do small projects that really have a significant impact.
From an impact standpoint, you just have to bring the right things together. If you bring in the CoE team and 18F team and the PIF team and maybe even leveraged some of OPP, you’d probably have a significant impact with one or two members of each of those teams actively participating together with a joint outcome in mind.
Which leads me to the second point. The groups have been put together over the years and it’s evolved over time—from OPP and 18F and the PIFS and more recently the CoEs—and they still operate in their siloes. There’s an opportunity to look across. And we’re actively doing things like that.
Take cloud, for example. OPP has a lot of cloud impact with FedRAMP and all of the—some of which actually come out of 18F—Cloud.gov. 18F does a lot of cloud work and cloud acquisition work. The CoEs have a lot of cloud playbooks. And, frankly, if you take that one step further, the rest of FAS has a lot of cloud schedules: ITC [the Information Technology Category] within FAS has got a pretty large cloud portfolio. There’s an opportunity to bring all those together and bring all of them to bear for an agency, rather than the individual siloes.
And the agency would benefit. They’d say, “Well, this is pretty good. You’re bringing the whole thing,” from solutioning to definition to migration through acquisition. There’s a whole game here that we can be playing that’s a lot bigger than the individual parts.
What is the future of TTS? Briefly, before this administration took office, TTS was raised up as a third line of business at GSA, beside FAS and the Public Buildings Service. Then, it was relegated back under FAS. Can or should TTS be its own delivery line or is technological transformation intrinsically linked to acquisition?
It’s funny. We were in a FAS leadership council meeting and—I don’t want to get into the historic reasons as to why they were not as integrated as they should have been. But the more and more I sit down in FAS conversations, the more and more I think of them as a machine that’s very large and very effective in doing acquisitions, especially in the space of technology. For the TTS organization to not leverage that, you’re not really driving the full value. You can make transformation happen at a small scale and you’ll never really be at the impact that you want to be unless you pull in the broad machine of FAS.
Now, is that the only channel for transformation: acquisitions? No. You can do transformation, and acquisition is a piece of it. You also could be partnering better with industry partners. You could be bringing in industry further ahead in the game—I think the CoE teams do that to some extent.
But I think there’s an opportunity to leverage industry a lot more—both the services as well as on the software side. There are a lot of players out there, especially new, innovative players that, frankly, government can benefit from spending a lot more effort on.
FAS is a major channel for TTS to be effective and drive change. But so is having industry partners on the services side and on the software side.
Organizationally should it belong or not? I don’t want to get into it. That really doesn’t matter. It’s really how you operate that’s more important.
Being separate and not being connected to FAS is wrong. Being part of FAS and still not being connected is not right either.
Technology and acquisition certainly seem to be linked. There have been a number of large IT contract vehicles being planned by agencies that are now moving through GSA. We’re seeing this a lot more with tech than buying other things, like air conditioners. What is it about buying technology that is pushing agencies in this direction and why are they coming to GSA?
I think it’s a combination of things. And I would say that GSA does a really good job of buying fleet and should be doing more, as well as in the buildings area, it does a really phenomenal job. Those are two areas where GSA is doing really well.
The evolution of technology, now, from being one where you buy widgets—you buy a server or you buy a network or you buy a piece of hardware—all of that is changing very quickly. Everything is now software-defined and becoming software-defined. If you’re familiar with the Wall Street Journal article “Why Software Is Eating the World” back in 2011. That’s pretty much what’s going on.
If you look at the telecom industry—why EIS [Enterprise Infrastructure Services] at GSA going to become such a big success, especially with 5G, is because it’s software defined. It makes it much more easily controlled and there’s a convergence of software and hardware that’s going on.
What is a cloud? It’s a software-defined infrastructure.
Everything is going toward that, where people are not going to buy individual components but they’re going to buy solutions. And there’re going to buy combined solutions. It’s going to be harder and harder for CIOs in each of the organizations to become familiar with all of those areas. And I think the solutioning around that, either they’re doing that on their own or they’re using GSA or they’re using third parties—integrators.
That’s one piece of it. Technology is changing so much that it makes the buy more complex and you need the assistance to do that.
That’s where the TTS role can become even more. We do some assisted acquisition but I think there’s a significant amount of work in really driving—what I call “moving left” in an organization—driving the change to be more business focused, outcome focused, solution focused, rather than, “I need 23 widgets and 44 servers”—that’s the extreme right of the acquisition.
Sounds a lot like the principles in the Technology Business Management framework, which is now a mandate for agencies to incorporate. Before coming to TTS, you chaired the TBM Council, which created a taxonomy for linking IT investments to specific business outcomes. But TBM is not the only framework out there. What makes it special? Why should this one framework be mandated across government?
I’m kind of biased. Having run the TBM Council—which I don’t anymore, having joined the government. But when I was looking at this as a problem to be solved in my prior career, I did a lot of research into all the different frameworks that are out there and, frankly, found nothing that really translated—from a taxonomy perspective—these component parts to a business solution. That’s fundamentally what TBM does, it pulls it all together in a taxonomy that’s clear, that’s manageable and needs a fair amount of work to do that translation.
But it really didn’t exist anywhere. And TBM was one—I was one of the founding members of the TBM Council back in 2011—to really drive that. There were six or seven of us as CIOs and now it’s grown to 7,500 members and it’s international and everybody’s following it, etc., etc.
There’s no one that’s really caught up with that real framework. There are lots of players that get involved with it. But they’re catching little pieces of it. That’s not to say that others won’t catch up. But I think at this stage, I would recommend it at any point in time.
But, frankly, I’m not involved in those discussions. I was on the other side of the equation when the government said, “Hey, we want to go, by 2020, have all budgets go to OMB using the TBM taxonomy.”
There might come a point in time that we build that competence as a CoE. But there are lots of other interesting things for us to do. But I think it’s a worthwhile thing to do that transformation.
Back, for a minute, to the consolidation of contracts at GSA. Based on your experience in the private sector, is there a tipping point where that becomes problematic?
I don’t know enough, frankly. This is one of those where it would probably be wrong for me to comment about it. I’m not that steeped in the acquisition process as perhaps I should be. A little more than three months in and I’m learning it.
Should 100 percent of all contracts across government come through GSA? I don’t know. Maybe there’s a point where you get to value and then there’s a point you don’t want to go beyond. But I’m just guessing.
I ran acquisitions for the bank. But that’s a different story than running an acquisition arm of the government.
Getting into the programs, let’s start with the Centers of Excellence. There are plans to open two more centers—for seven total—and maybe in the future, you just mentioned, a TBM center. What will this program evolve into?
Let me back up a little bit and say that the term “CoE” is being used in two different ways right now. One is “center of competency,” of which there are five right now: client experience, call center, IT modernization, cloud and analytics. Those are centers of competency.
However, the term is also used as the approach being taken, coming top-down, agencywide transformation.
Agencywide transformation coming top-down, working with the secretary or deputy secretary on down is a good thing. We should continue to do that.
The competencies that we bring right now are five. We’ve got two agencywide transformations running: USDA and HUD [Housing and Urban Development]. And we’re pushing for another three this year. Will they all show up immediately? There’s a whole process I’m going through for determining where and how. And it’s not just me—it’s [GSA Administrator] Emily [Murphy], as well as OAI [the Office of American Innovation] and [Federal CIO] Suzette [Kent] and [White House Deputy] Chris Liddell—we’re all actively looking at evaluations of what the next agency should be.
Are you ready to break news on the third one?
No, not yet. You’ll hear about it when it happens. I’m hoping in April, but these things take time.
But going back to those five centers. We are actively looking at rationalizing those five and thinking about what additional ones we should be adding to it. Shortly, we should have a list of what those competency areas are. Client experience is clearly one; call centers kind of falls within client experience. Cloud and IT modernization are very tied together.
And we’re actively looking at new ones. RPA [robotic process automation] is one. We’re looking at identity security. Really, we’re looking at what do agencies in the government need. I think those are ones we’re familiar with—for security and identity we have Login.gov, a lot of security work gets done at FedRAMP.
When you say rationalizing, you mean looking at whether they should exist on their own or combining some?
Think of taking the word CoE and using it to define solution areas. The approach that’s being used for agencywide innovation is really the right approach. Do we need to bring all of these solution areas, some of these solution areas, one of these solution areas to an agency to drive transformation from the top down?
One major change from the first two rounds: Phase I contractors couldn’t compete in Phase II. That’s changing going forward. Is that safeguard no longer necessary?
It’s still there. What we’ve done is to say that if you’re in Phase I in one domain area, you still can’t do Phase II in that same domain area. But there’s no reason you can’t do Phase II in another domain.
If you’re doing Phase I call center, you won’t be doing it for Phase II. But you could still do Phase II cloud, or something else.
Fundamentally, what we’re saying is don’t give the full henhouse to the fox—or however you want to call it.
Let’s go to 18F, which was part of the genesis of TTS becoming its own thing. There have been a number of GAO and inspector general reports on that team in the past. What issues have you found in the first 100 days and how are they being addressed?
Frankly, the team has done a really good job before my joining to fix a lot of the IG and GAO issues. I think it’s commendable their taking that and really driving that. Their leader, Angela Colter, has done a really good job of making that happen.
18F is doing really well and thriving. It’s doing its thing—path analysis work and user-centered design—and doing it for probably about 18—it’s in the teens—agencies that they’re actively engaged with. One of the issues was if they’re cost-recoverable and are their consultants busy enough—fully utilized. That’s what’s really changed. We’ve really got that up and running in a real way. All of that is really now cleaned up in a significant way, in partnership with our general counsel and their team, there’s been a lot of good work in fixing that.
Where I want to take it moving forward is really that impact conversation that we had earlier. How do we drive bigger impact, pick and choose the right projects, connect it to the rest of the family—bring in PIF when we need them or the centers of competency when we need them—build platforms, if appropriate, which is really what 18F has done with Login.gov and Cloud.gov—there could be lots of other platforms we could be looking at.
There are many more programs, too: PIF, FedRAMP and some of the other groups under the umbrella of TTS. Can you talk about the rest of the family?
The PIFs: We just had another 19 or 20 PIFs join in January. They’re working at 15 or 16 agencies. We’re actively recruiting, looking for the new wave. April 22 is the timeframe for when we need all the applications in. We’re looking at late summer to have another wave of PIFs come in.
There’s a high demand for them. All the agencies love the PIFs and they work with them, particularly on the CTO [chief technology officer] side of the conversation, there’s a lot of good connectivity there. So, that’s going to continue to grow.
Where does the demand for PIFs come from? At 18F the demand comes from tackling a specific problem. For the CoEs, agencies come to you. When people come to you and say, “We need a PIF,” what are they asking for?
They’re typically looking for assistance in a technology area that’s very specific. For example, we’re doing a fair amount of work with the Marine Corps. Their CTO [Jennifer Edgin] really wanted some technology assistance in what they were trying to get done. VA [Veterans Affairs] is another one, where, actually, the CTO [Charles Worthington] used to be a PIF. So, he knows our capabilities and he’s already laid out some critical areas.
They could be in artificial intelligence, machine learning. They could be in use of visual recognition tools. There’s a whole list of specific needs they come to us with.
On the CoE side, I do want to say that they don’t all come to us. We work with several individuals to jointly define where the opportunity is.
On the OPP side, which is the Office of Platforms and Products, FedRAMP is probably the most well-known. I’m getting a lot of good input from lots of players on the good, bad and the ugly on FedRAMP. If I were to say, “Here are the opportunities,” on FedRAMP, people want it to be faster, cheaper and more reused across government. That’s at the heart of the complaints.
The team’s done a phenomenal job of getting it done faster. They’ve certainly improved on the cycle times. We have an opportunity to get the voice of the customer better into FedRAMP.
FedRAMP has been dealing with these same issues since its inception. When you say getting more of the “voice of the customer,” is that a novel idea? What can be done to get over those hurdles?
Some of the noise that I hear right now is very anecdotal. What I’d like it to be is much more compliant driven and much more curated, as it were, so that we have a forum and an approach.
The FedRAMP team does do a lot of clientsat [client satisfaction reviews] and actually their clientsat is very high. But I want to get the broader picture and we have to figure out how best to do that.
We can continue to speed it up, leverage technologies available out there. It’s fundamentally a risk management framework—there are lots of players who do those things. The scoring mechanism to make that happen, in terms of what level you have—there are lots of opportunities to automate those further. And then, work very closely with [Federal CIO] Suzette [Kent] and the CIOs to drive adoption.
There are three issues: the cost, the time and the reusability—the portability of security accreditations from one agency to another. Sounds like you’re working on the first two, but what can TTS and FedRAMP do about the third?
There’s a lot of outreach that we could be doing. There’s a lot of communication leveraging the CIO-wide network [the Federal CIO Council] that already exists that Suzette runs. We don’t need to recreate a lot of these. I think this is more of a drive by influence.
Making it a law is probably not the easiest thing to do. But making it a needed activity with the right buy-in from everyone. ATOs [authorities to operate] are done often by the CISO [chief information security officers] in the organization, getting them to buy in—getting them to have a say, getting to understand what their issues are—I think that’s part of that whole change that needs to occur.
What is the future of TTS? What new programs need to be launched? What areas of technology transformation does TTS not cover now but should?
Let me put it another way. We’re actively looking at how do we do three things: one is build momentum. Building momentum is really about how do you get more agencies to leverage TTS and build it in such a way that we are doing what they need us to do. If they want us to do blockchain in the next year, maybe that’s a competency we need to build. I’m not sure that’s real. But meeting the demand and understanding the demand and building the momentum to drive that is one area that we are really focused on.
That means having the different players at TTS play together in an integrated fashion and really playing above our weight and having more impact.
The second piece is about establishing a sustainability. It’s a very small organization. I don’t think it should be a humongous organization. But I do think it should be a sustainable organization, in the sense that, when you have competencies, is there reusability, is there a community of partnership there, are you leveraging the right external sources, do have a method by which you are doing something sustainable—even the demand process has to be sustainable.
Building momentum, establishing sustainability, and the third piece is really driving for excellence. I think there’s an opportunity here to not make money but to be excellent in what we do, and be financially independent—or cost-recoverable—which I believe if you do the right things in the right places we will get to.
Those are the three things we’re looking at. And if it generates new opportunities—RPA is one that is a good opportunity, maybe that’s one where we spend more time and effort building. All of that will come based on those three main strategic goals.
Part of that sustainability, by the way, is leveraging GSA more, leveraging the FAS organization, leveraging our partners more. There’s a lot to be done. But I think we can be the tip of the spear on a lot of transformation that’s going to take place.