Most Active Stories
All Tech Considered
Wed October 23, 2013
U.K. Official Urges U.S. Government To Adopt A Digital Core
Originally published on Wed October 23, 2013 6:43 pm
When he read about the technical failures plaguing HealthCare.gov, Mike Bracken said it felt like a real-life version of the movie Groundhog Day. During the past decade, the government in the United Kingdom faced a string of public, embarrassing and costly IT failures. Finally, a monster technical fiasco — a failed upgrade for the National Health Service — led to an overhaul of the way the British government approached technology.
Instead of writing behemoth, long-term contracts with a long list of specifications for outside contractors, Parliament greenlighted the creation of the Government Digital Service, a "go-team" of 300 technologists who began streamlining 90 percent of the most common transactions the British people have with government. It appointed Bracken, a tech industry veteran, as the first ever executive director of digital — a Cabinet-level position.
Two years later, gov.uk is a single, simple platform connecting the hundreds of British agencies and allowing people to pay taxes, register for student loans, renew passports and more. Doing technology this way is saving British taxpayers at least $20 million a year, according to government estimates.
Not everyone is onboard with the reforms. For one, becoming "digital by default" means those who prefer a more analog relationship with government services are forced to adapt. And one of Bracken's biggest critics is a man named Tim Gregory. He argues that putting technologists at the heart of government stifles business investment in the U.K. Gregory is the U.K. president of CGI, the global contractor whose American arm was the biggest contractor on HealthCare.gov. (Bracken calls Gregory's complaint "beyond parody.")
The energetic digital chief was in Washington, D.C., this week to speak with the Presidential Innovation Fellows, some of whom are part of the "tech surge" aimed at helping fix the system. He sat down with me for an extended chat about the "not sexy" heart of the HealthCare.gov failure, his hopes for what comes from this crisis and the lessons he learned abroad that could help the U.S. (The interview has been edited for length and clarity.)
What does being "director of digital in the U.K." mean?
Great question. That means I am responsible for making public services digital by default in the U.K.
How did you come to be in this role?
I'm the first person in this role, and the position has existed since mid-2011. It came about because of a report written between Martha Lane Fox, who is [a] digital champion [in] the U.K. and an entrepreneur, and Francis Maude of the Minister of the Cabinet's office. The report's called Digital by Default and it recommended four things: that government has a digital center as its heart, [as] it never had one of those before; digital capacity and new digital skills right at the heart of government ... and then [that] it fix publishing, fix all its transactions, make them both digital by default; and then finally, link them all up with open data.
Why was there a need for government to have a digital heart?
We have grown accustomed to too many technology failures in government, and also our digital services aren't keeping up with digital services outside of government. In the U.K. it's widely accepted that our public services are pretty good. We're a reasonably populous country at 70 million people. We have a long tradition of public services. We do an awful lot of different types of public services to a high standard. Yet it's also accepted that when it comes to transposing those to a digital world, our services lack a little bit in quality. It's not for want of spending money on technology, but in comparison to things like booking airline tickets or buying books or all the stuff we do in our daily lives that government services and public services are not keeping pace.
How did the U.K. arrive at this conclusion?
Well, it arrived at it over time, to be fair. Throughout the 2000s. But one of the critical events was in late 2008, 2009. We had yet another big ... IT program ... this one called NHS IT — the National Health Service. It's not actually under [me]. It's outside of central government. But there'd been a massive multibillion-pound IT program, [and it] hadn't yielded good results. And I think that was the moment when both politicians and civil servants felt, it's time to try another tack here. Because pouring money into these big IT programs just isn't yielding great results.
And when you say "isn't yielding great results," what was happening where we might be able to sort of see some parallels?
Well, a bunch of things were happening. We were getting too bogged down in long-term multiyear procurements. We were trying to predict the future in a digital world that's changing rapidly. Because we were trying to buy things with five, eight, 10-year cycles, we just couldn't possibly keep pace with what was happening. Also we weren't getting good values as a result of that. We were trying to outsource the whole thing into a certain contract type and weren't really in control of that. And the final thing is, you weren't able to react to user need. Users' needs change. Because we were locked into these big timelines and because we were costing a lot of money, we couldn't react to users' needs and actually to changing policy needs.
So all those factors combined to really create an inertia at the center of government, and to unlock that inertia you need many different things. But the critical thing you need is delivery skills. You need a new younger generation. Not just a younger generation but younger skills. It's like the Internet generation. You can create stuff really quickly, that can get stuff out the door maybe in alphas and betas, try early projects, and iterate and change them depending on user need. ... And that's a fundamentally different model than trying to predict the future and buy your way out of it with big contracts. We've had to really unpick some learned behavior that government's had for the best part of 20 years.
In the U.S. we have a chief technology officer, Todd Park. And there's been smaller initiatives like the Presidential Innovation Fellows. How is what Todd Park does or the "power" that he's given different from what you've been given?
Todd Park is an amazing person and his enthusiasm and his skill show through and he's great. However, I think that one of the issues that you have here and other countries have is the absence of a delivery capacity — the absence of being able to put your hand on teams of highly skilled, multidisciplinary technical and digital and policy people and deploy them at points of real urgency means that Todd can only be as good as his influencing skills to the existing supplies and the existing contracts. They need rooting around and doing differently. And that's a fundamental problem.
Government only has a finite amount of delivery capacity. In a world of policy, in a world of political imperatives, the next big thing comes along and your best people have moved on to that. So unless people like Todd Park have an organization and a structure and skills like we have in [Government Digital Services], then I don't see how they're ever going to be able to really structurally change some of the problems you have. And I do think that will lead to more and more experiences like the one you're potentially having right now with HealthCare.gov.
There's good signs here. Jen Pahlka has come on from Code for America for a year. And if she's given the backing and the team to build the sort of mini GDS or macro GDS, that's great. There's the Presidential Innovation Fellows. There's plenty of places where you could put these skills in the center, but until you actually have them in substantial number — we have 300 in the U.K. — ... then I fear that you'll always be working through it through influence, however talented some of your technology leaders are.
I imagine one of the reasons your government arrived at this is cost?
Yes. So it saves us huge amounts of money. So when we have that team, the way we deploy them — there's still a finite resource; it's still not enough. Because government is quite a distributed thing as I'm sure it is here ... the first thing we did was create a common platform so everyone in government can use it. That's called gov.uk. So if you go to www.gov.uk now, it's one country, it's a domain for the entire country. There's still some agencies to come onboard, but most of the big departments are done. ... In the next year all of government will be on it.
Right now in the U.S., we have a bunch of agencies, hundreds of agencies, but they contract out for their own sites and their own dashboards and their own content management systems.
The problem that everyone in government as a user has is, you don't just get information from government. You transact. And we all know the big stuff — driver's licenses, passports. Governments are pretty much the same the world over. They do roughly the same stuff. So using that finite number of those 300, what we've done is gone after 25 of the top 50 transactions in U.K. government. The top 50 account for 97 percent of volume. And you can guess what they are. I've mentioned some: passports, driver's license, tax, health care, all this stuff.
What is your reaction to HealthCare.gov and what you're reading and seeing regarding failures of what was meant to be an Expedia shopping for health coverage?
Yeah ... I'll say this with no sense of enjoyment whatsoever, but it feels a bit like Groundhog Day to where we were three or four years ago. Hundreds of millions of dollars, large-scale IT enterprise technology, no real user testing, no real focus on end users, all done behind a black box, and not in an agile way but in a big waterfall way, which is a software methodology. And basically not proven good value, and I'm afraid to say I've got example after example in the U.K. in the past where we've had that experience. So it looks just like one of those.
My hope is that the current shockwaves of what you're going through here are strong enough to implement a new approach and actually to get political will behind having digital skills in the center. Because delivery is the only thing that will solve this problem. I've not been thrilled so far by the response in terms of the view of technology that you can send one or two or a whole fleet of people and then crack this problem. The real problem is systemic. You actually can't build technology like this. Technologies aren't things that are binary. You don't procure them. They're living systems and you have to have people who look after them and develop them iteratively and change and grow with them and you need those skills in the heart of government.
What is your fear? What does the shockwave reaction now portend for us?
You're destined to repeat the problem. This is something that governments have done all the time. Very few have them have this public [attention]. So often projects like this can be sort of quietly shelved and buried in hundreds of millions of pounds or quietly put through on the side. And because end users aren't touched by them there isn't, if you like, an outcry or so much of a view. This is different now. So, my fear is that governments continue to see technology as big white elephants, that you create a service and then you leave it. You have to take a different approach. Reform of procurement is the elephant in the room. If there aren't steps to take substantial procurement reform then that's a problem. Procurement reform is the killer for any dinner party conversation. I recognize that.
Not a sexy topic.
Not a sexy topic, but it's so important because we can't keep buying technology like we're buying bridges or we're buying motorways. It's not the same stuff. It's much more involved in its faces and users. We have to be much smarter about it and we must have in governments all over the world, I'd say, the skills at the heart of government ... skills to understand how to use technology. By the way, I don't mean do it all yourselves. ... You should use suppliers, use vendors, use the best that's out there. But not in this sort of one-size-fits-all kind of way.
You were here in Washington to speak with the Presidential Innovation Fellows, a group that I'm sure you're very supportive of. What was your message to those folks? What were you here to say to them?
I was giving them a bit of feedback from the three years we've had at this. The first was, go at the big stuff quickly. Do the stuff that matters. Solve the big problems around procurement. Make sure we get digital capacity into the heart of government. Start to look at end user services.
Get the data for your services and put them in the public domain ... how many people applied for benefits today? How many people were successful? Those people who weren't successful, why did they fail? Use that as a baseline and try to drive services. Too often, and this country's no different, I talk to CTOs and CIOs and they say, "I've got all that data." Great. Bully for you. Well, publish it. Let everyone have a look at it. Because actually it's not technology data. It's data about public services, and we all pay for them.
Finally, the president said Monday that the policy of health care reform is not the website. How would you characterize the relationship between what government services purport to do and how much they're dependent on the technological systems?
It is the prevalent distribution model of our time. I don't think you would hear politicians say, "Well, the government buildings, they're not the government," because you have to go to government buildings to transact with them. So websites, digital channels, mobile services, applications, APIs, they are the government. And that's a critical thing. Digital services are public services. The Web services are indivisible from public services. And that's a generational message that I think the Web generation understands. And potentially because we've thought of technology as procurement for so long, that message has been missed in political circles.
MELISSA BLOCK, HOST:
A federal IT project plagued with high-profile problems, integration breakdowns involving dozens of contractors, and taxpayers footing a multimillion-dollar price tag. Well, that scenario has also played out in the United Kingdom - so many times, in fact, that it led to big changes in how government tech work gets done there. NPR's Elise Hu has that story.
ELISE HU, BYLINE: By now, you can hardly turn a channel without hearing about the tech failures plaguing HealthCare.gov. Serious problems make for biting satire, like this from "The Daily Show's" John Oliver.
(SOUNDBITE OF TV SHOW, "THE DAILY SHOW")
JOHN OLIVER: There is a small risk that applicants might get sucked into a Tron-like, cyber-bureaucratic nightmare with no escape.
HU: The site to help Americans in 36 states sign up for health coverage cost at least $80 million. And weeks after its rocky rollout, it still doesn't completely work.
MIKE BRACKEN: It feels a bit like "Groundhog Day" to where we were three or four years ago.
HU: That is Mike Bracken, a technologist who watched national IT programs fail again and again in his home, the United Kingdom. The tipping point for the U.K. government came when a system for the National Health Service got contracted out, cost more than a billion pounds and - you guessed it - didn't work right for its end-users.
BRACKEN: I think that was the moment when both politicians and civil servants felt it's time to try another tack here because we have grown accustomed to too many technology failures in government and also, our digital services aren't keeping up with digital services outside of government.
HU: So the U.K. changed up its entire approach to government tech projects. Instead of multiple agencies procuring different contractors for their individual jobs, it put digital at the heart of government. Parliament created the government digital service, attracted 300 technologists to work for it and gave Mike Bracken a cabinet-level position as its boss.
BRACKEN: Technologies aren't things that are binary. You don't procure them. They're living systems and you have to have people who look after them and develop them iteratively and change and grow with them. And you need those skills in the heart of government.
HU: This in-house team simplified nearly all public services across agencies. Voter registration, tax payments, student loan applications and passport renewal are all available through one site and one streamlined system. Doing it this way instead of contracting out has saved British taxpayers nearly $20 million a year.
BRACKEN: We don't go in the government digital services to save money. We go to create great services and saving money is the byproduct of doing that.
HU: Not everyone was on board with these tech reforms. Bracken's biggest critics said putting technologists inside government would stifle business investment. That critic, Tim Gregory, the president of the U.K. branch of CGI. Yes, that CGI, the biggest contractor on healthcare.gov. For his part, President Obama is trying to make a distinction between his health care law and the technology powering it.
PRESIDENT BARACK OBAMA: The Affordable Care Act is not just a website.
HU: Bracken says that's the wrong perspective.
BRACKEN: I don't think you would hear politicians say, well, the government buildings, they're not the government because you have to go to government buildings to transact with them. So, Web services are indivisible from public services. And that's a generational message that I think the Web generation understands.
HU: He hopes America's huge, high-profile tech failure can lead to big, systemic changes in how government approaches technology. Elise Hu, NPR News, Washington. Transcript provided by NPR, Copyright NPR.