Note to readers: You can also join us for a discussion of what the next governor of California should do to promote delivery-driven government on November 15th at 6:00 pm at the Code for America offices, or by livestream. Sign up here.
A letter from the near future.
November 7, 2018
Dear Governor-Elect of California,
Congratulations on your win, and the opportunity to lead the great state of California. And now for what’s next. As Washington says to Hamilton, at least in the musical version of history, “Winning was easy, young man. Governing’s harder.”
One underappreciated reason governing is so hard is that as our society has hurtled into a digital age, we’ve left government back in the seventies and eighties. But as Brian Lefler says, “neglecting the machinery of government is a choice,” and we the people of this state can make a different choice, in partnership with our elected leadership.
Ten years ago, I was inspired by another newly elected leader, Barack Obama, to help to show that government can work as it should for all Americans. During his campaign and transition, President Obama signaled a willingness to explore how we could thoughtfully bring our government into the digital age, and thousands of people in technology, design and government rose to the occasion. I was proud to help, first by starting Code for America, and later by doing a year of service as Deputy Chief Technology Officer for Government Innovation at the White House, getting the United States Digital Service (USDS) off the ground. And because of hundreds of people who came together around these efforts, government now has a new playbook. That playbook is why veterans can now manage their benefits online, more families in need are getting help feeding their kids, and 11 million Americans got health insurance, despite the initial failure of healthcare.gov. (That site was built the old way, but fixed the new way.) This is a success story, but one that’s just begun.
With your election as Governor, we have the opportunity to write another chapter of this story. Choices you make now will have a huge impact on where the story goes next. California not only has the opportunity to lead the way now; California must lead, for the sake of the nation.
With enormous respect for you and the office you are about to assume, I offer the following advice as you prepare to lead California to lead the nation.
Focus on delivery. President Obama knew he wanted to modernize government, but in his first term, his administration focused on transparency and open data as the mechanisms. Meanwhile, across the Atlantic, the UK’s Government Digital Service (GDS) was taking a different approach, still championing transparency and open data, but centering the work around the delivery of services to the public and measuring its value by the improvement to real people’s lives.
When healthcare.gov sputtered so dramatically on launch in President Obama’s second term, we finally had the political will to make service delivery the priority and start the USDS, modeled after the GDS, and publish the digital services playbook. Lesson learned: delivery is the organizing principle that drives the kind of change elected leaders want and we need.
What do I mean by delivery? Ultimately, government is judged by the services it delivers to its constituents. If a service is poorly designed, inflexible, slow, and unfriendly, the conclusion is that government too is all of these things. Delivery means starting with the needs of the users of a service — the people waiting in line at the DMV or trying to sign up for CalFresh benefits or housing vouchers or MediCal, or trying to clear their record now that a former felony has been reduced to a misdemeanor. If it’s harder to sign up for these things than it is to sign up for online banking, that’s a delivery problem. What we have today across government is a delivery crisis, and thankfully many public servants in our state recognize that and are taking action.
Increasingly, service delivery in today’s world is digital, on websites and, more importantly, on mobile phones. Getting digital delivery right is something Californians are really good at. Silicon Valley leads the world in creating delightful, effective digital experiences. California’s government has to be just as good.
Make this a top priority. If you don’t make it a priority from the outset to improve government’s capacity to deliver services in a digital age, it will make itself a priority, through a crisis like healthcare.gov. Crisis is one way to motivate change, and we should, in Rahm Emanuel’s words, “never let a crisis go to waste.” But a crisis can mean something that dominates the headlines for three months straight, or it can happen unnoticed by the press to disenfranchised people in our country every day, like the failures of SNAP and Medicaid delivery systems in Rhode Island, Indiana, and other states, when food and medical benefits that families rely on simply stopped, with devastating consequences. Or defendants being wrongly arrested or jailed in three counties in three different states because the courts’ case management system vendor can’t fix an error. Or a child placed in a foster home with a known rapist because the only way to convey that information to the social worker was by fax machine. If it’s crises we need, we got ’em. We have a moral imperative to make this a priority now.
If this is a top priority, your cabinet will reflect it. You need a cabinet-level position focused on technology policy and digital delivery. Government has for too long made digital just a thing you buy from vendors, the work of low and mid-level bureaucrats with little understanding of what they’re buying and even less strategic input. That doesn’t cut it in 2018. But even more important than a digital leader is the rest of your cabinet. Whatever domain they are experts in, they must be willing to lead not just policy, but also delivery, because the two are inextricably intertwined in this day and age. This does not mean they need to have technical or design backgrounds; but they must be curious, humble, and coachable as it relates to the digital world, even as they are at the top of their profession in other disciplines. If they’ve spent their careers in government or legacy industry, they must be willing and even eager to unlearn the waterfall processes of the past and relearn iterative, user-centered, data driven approaches. They will need to build multi-disciplinary teams and break down silos. They will need to learn to ask different questions. They will need to talk to the geeks as much as they do the wonks. Plenty of career government leaders have made this transition brilliantly, and can share how much more rewarding it is to lead this way. They will not be alone. But they must be willing.
Understand, and help others understand, that it’s not about technology. Digital is a means to an end. The problem is not that we don’t have enough technology in government, or enough investment in technology. When government routinely spends in the billions of dollars for technology projects that don’t work, when government services take hours or days to navigate when they could take minutes, we have too much technology, not too little. What we have too little of is service design, agility, understanding of user needs, and feedback loops that allow us to iterate to get to better outcomes. This is what we call delivery-driven government, and it’s the basis for doing smarter, iterative policy, too. Tech and design are the means to providing services to the public in the most effective, efficient, and respectful ways possible, and rebuilding faith and trust in our public institutions.
Build on the team California has today. I said earlier that California must lead, for the sake of the nation. And here’s the thing: California is ready. Over the past four years, the government operations leadership, in partnership with the Health and Human Services Agency at first and now several other departments, has remarkably and courageously begun transitioning to the new playbook, starting with a decision to break with the status quo on a Child Welfare System procurement in 2014. It hasn’t been easy, and it hasn’t gone perfectly. But today you see an agile, user-centered team regularly shipping code that works for its users, and you can’t overstate what a difference that represents from what would have been under the old model. All over the country there is a recognition that government’s standard operating procedures for building and buying technology work incredibly poorly, and California is now among the leaders in showing it doesn’t have to be that way.
Kudos in particularly to CIO Amy Tong, Secretary of Health and Human Services Michael Wilkening, and Secretary of Government Operations Marybel Batjer, who’ve shown enormous courage and skill in leading and navigating this transition. You won’t do better than these three, and there are many others like them already serving this great state. You may want to add some new faces, but fight like hell to retain the talent that’s already there if you want to keep up the momentum, because there’s still a long way to go on this path.
Really, actually, truly value learning and continuous improvement. And show it. This is a hard one, because every incentive in government structures and culture works against you. It’s cheap and easy to say that government is rigid and risk-averse, and should be otherwise; it’s a very different thing to actually support real learning on a path to transformation. As Clay Shirky said of healthcare.gov’s initial failure, the waterfall method that produced it (and produces most deliverables in government) “amounts to a pledge by all parties not to learn anything while doing the actual work.” The trouble is that to learn, you must fail. Waterfall processes attempt to avoid failure at all costs, tending to result in spectacularly large failures from which nothing is ever learned. The trick is not to avoid failure, but to fail small, fail fast and learn. And apply that learning immediately, and then do it all over again. But that takes leadership and courage, especially in a government context. If you don’t want to lead a risk-averse workforce that moves at a snail’s pace, you’ll have to show your people you have their backs right out of the gate, and make sure your top team does the same.
Change procurement, but start with 90% practice, 10% policy. You already know that the way government procures hardware, software, and services makes little sense. Most of the rules were written in a pre-digital era, so not only do they apply poorly to things like software, but they’ve also had years to accumulate and accrete, making them both overly rigid and overly complex. Understanding what can be changed in the law, policy, and regulation that governs procurement will be necessary, but don’t underestimate what you can get done within existing law, policy, and regulation. Much of what needs to change is are complex and rigid practices held over from the past. They can be changed more than you might expect by demonstrating that there are other ways of doing procurement, and by retraining contracting officers and others in government to have them follow a different set of practices. Understand that fixing procurement is, like everything else, not a simple matter of finding the one right law or policy to amend. There are thousands of human beings in the procurement apparatus and the departments they serve who need to be brought along.
While you don’t want to wait for changes in law and policy to change procurement practices, you should still initiate those changes. This will not be easy. One approach you should consider mimicking is what the Department of Defense has done with the Defense Innovation Board (on which I serve). The Department appointed a dozen outside experts to look at ways to speed innovation, and then Congress charged the board with creating a Software Acquisition Practices subcommittee responsible for making recommendations to Congress about changes in law that would reduce complexity and improve outcomes in DoD software. This is long hard work and specific to the context; what works in the DoD may not be quite what works in state government. But getting specific and practical about the statutes that hold government back is important for long-term change.
Make targeted civil service reforms to support your agile strategy. It’s hard to hire anyone in government; it’s even harder to hire and retain digital professionals. Job classifications and hiring rules written decades ago mean you can’t evaluate digital talent using relevant criteria, and often you can’t write job descriptions that match what you really need. (The great team you’ll inherit has already created a job classification for data scientist, for instance, but there’s still a long way to go.) Talent is crucial, and don’t believe those who say you can’t get great digital talent in government. In fact, some of the best digital professionals today are working in government or trying to, because that’s where they know they can most help their fellow Americans. You’ll just need to direct your team to spend some time deep in the weeds to make it easier to get them on board.
Open an Oakland office for digital. Sacramento is where California government happens, but it’s not where the majority of tech and design talent lives. Enough of the people you need will take a pay cut to do some good, but too few of them will upend their lives and the lives of their families and move to Sacramento. But build a digital service unit close to the Oakland or Emeryville Amtrak stations, and you’ll have access to most Bay Area talent, while making it easy for them to get to Sacramento regularly to work with their partners in the various departments and agencies.
Get ahead of the pushback. Not everyone wants government to run the new playbook. For some civil servants, the old way is what they know how to do, and change is uncomfortable. But more importantly, there’s a lot of money at stake. $2B for a court documents system that never saw the light of day, $900M for a deeply troubled financial system, and hundreds more projects with that many zeros. Doing digital right in government will mean fewer of those monolithic non-starters, and some of the vendors who would have gotten those billions will fight the change. And they fight dirty. Aggressive lobbying is just the start. They will make friends with you and your team, and give you a thousand reasons why the new way won’t work. They will exploit the rules designed to make contracting fair, particularly protesting any contract that follows the new playbook, slowing every project they can down and increasing their costs. And the ones who fight this change the most also also have the most power and resources.
But there’s a lot for the vendors to like in the new way too, and some of them see that, especially the newer entrants who already run the new playbook, but also some of the legacy vendors who are just sick of how dysfunctional the whole system is. In the new model, they’d get to be heroes instead of villains. There would be smaller contracts, but more of them. And they’d get to do more technology and less litigating. (Many people describe the process today as contract, build, fail, sue, and one reason governments don’t like to do business with smaller vendors is that they’re worried they can’t recoup their losses suing a small company into bankruptcy.)
Two strategies will be critical here. One I’ve already mentioned: back your people. The second is to proactively paint a vision of what the government digital contracting ecosystem could look like that clearly articulates the upside for everyone, including the vendors who will play along, and make it clear there’s a huge downside for those who won’t. Don’t demonize anyone from the start; tell your story and let them make a choice, and then celebrate those who contribute to your vision, not just with contracts but with public gratitude, and hold a hard line with those who won’t. They’ll all be watching; if they see that the old tactics still work, they’ll use them. If they see the game has truly changed, and that your administration is serious about a new model, and that there’s a win for them in it, most of them will come along. But be prepared for a fight, because it will come, and everyone will be watching.
Think capabilities and capacity, not just solutions. Most elected leaders I talk to want solutions. What software solutions should my team be buying? What policy changes will solve the procurement quagmire? Which vendors are the good ones? These are all fine questions, but they won’t get you where you really want to go. For decades, government has failed to developed basic competencies in digital, and addressing that issue is more like deciding to diet and exercise than taking a magic weight loss/fitness pill. One is hard work and takes time, the other sounds great but doesn’t work long-term. As I’ve said before, digital is something you do, not something you buy. I recognize that “I built the digital competency of our great state” doesn’t sound like reelection gold, but I promise you that strategy, when done right, will also result in enough concrete wins that you’ll have great outcomes to celebrate on the campaign trail. But if you only look to buy a system to solve a problem without building a workforce that continually learns and improves its ability to deliver, your gains will be shallow and short-lived.
This is a lot. I know. I lived it with dozens — maybe hundreds — of civil servants, government executives, technologists coming into government, vendors — you name it. It’s hard work, but the alternative is even harder in the long run. And it’s worth it. The world has gone digital and the people of California expect — even demand — better of their government. And you have the team and the ability to provide that. What does it look like from where you sit? Deploy these strategies with a strong message and mandate from the very start of your administration. Show the country — and the world — what governance looks like in a digital age.
Governing’s harder than winning, yes; but good, modern, digital governance is winning.