The CIO Problem, Part 1

I get about one inbound request a week for help finding technology leadership for cities, states, and federal agencies. Leaders everywhere in government have realized how profoundly their success is tied to getting digital right (and often how frustrated they will continue to be if they keep getting digital wrong). I wish I could help them all, but I’d like to share one thing I tell pretty much everyone I speak with who’s looking for great digital talent.

That piece of advice is this: You may be recruiting for the wrong job. Most (but not all) of the folks who reach out are looking to fill either a position for a Chief Information Officer or a Chief Innovation Officer. But both of these jobs are framed in such a way that it can be almost impossible to succeed in them, and the candidate you want is also the candidate who can see that, and won’t take the job.

Let’s start with Chief Information Officers. Today, CInfOs (for lack of a better disambiguating acronym) are generally supposed to do four things (borrowed from a paper I co-wrote with my colleague Dan Hon):

1. Digital services: The services residents use to engage and do business with the City. GetCalFresh, RecordTrac , and Healthcare.gov are all examples of digital services. This can also include APIs and open data programs, though this is often the domain of the other CIO (the innovation officer).

2. Back office software: Day-to-day core services like email, human resources management, budgeting, fiscal and accounting that all departments rely on.

3. Mission IT: The business applications that run the internal processes of departments and agencies. These are often custom, but can now make use of underlying commodity technology.

4. Infrastructure: Network and connectivity, hosting and device management.

The problem is that outside of government, these are distinct jobs. Each of these things calls for a different set of experience and skills and few candidates are likely to have all four of them, though it’s common to get some of them together.

In broad strokes, making great digital services means shipping digital products, which is the job of product managers, entrepreneurs, and others for which there is often no government equivalent. Back office and infrastructure is something an “IT” department does. But “IT” in government has unhelpfully come to mean “all things technology.”

I can’t keep count of the number of times a non-tech government leader has said something along the lines of “I really want one of your great digital folks, Jen, but for now I just need someone who can keep the lights on around here.” I couldn’t agree more; no way are you going to succeed when email is down weekly. Back office and infrastructure are obviously critical jobs. Having email and payroll that are usually up, for instance, are base requirements for creating an environment in which you can build and maintain digital services.

But give the top technology job in your city/state/agency to a back office or infrastructure person, and no matter how great that person is (and we should all value the ones who are good at this a lot more), you are very unlikely to make great digital services. Which is a problem, because the cost-savings that great digital services will bring to the operations of your city/state/agency are going to be needed to support your IT infrastructure.

There is no easy answer to this problem, but let’s look for a second at what others have been doing to recognize and address this issue. Back in 2010, in the UK, Martha Lane Fox’s letter to Francis Maude , the Cabinet Minister, called for him to:

Appoint a new CEO for Digital in the Cabinet Office with absolute authority over the user experience across all government online services (websites and APls) and the power to direct all government online spending.

Maude did that, appointing Mike Bracken, who had previously been head of digital for The Guardian (very much a citizen-facing digital service). Before he moved on from the job, Mike used to make a point of the fact that the Chief Technology Officer reported into him, because he saw technology as fourth in order of importance behind user needs, policy needs, and operational needs. (IOW, you only ask what technology is needed once you understand the first three.) This model clearly gives one person full authority for digital services, but with control over all government online spending, he had some control over mission IT and infrastructure as well.

Mike’s title was Executive Director of Digital for the UK Government, but most similar efforts to give one person the necessary authority in the US have used the CInfO title. In 2014, a White House delegation led by Brian Forde recommended that the CInfO role in Detroit carry a “cabinet-level” rank to ensure it has a seat at the policy table and the authority and respect to successfully implement changes. Within federal agencies, CInfOs make the decisions about technology spending, and thanks to the Federal Information Technology Acquisition Reform Act ( FITARA ), they do so in an increasingly centralized manner. But the trend there has been to add a “head of Digital Services” position alongside the agency CIO position, reporting often to the deputy secretary, which is basically the equivalent of the COO in a company. These agency Digital Services leads are often dual employees of the White House and the agency in which they are embedded, and typically focus on particular public-facing transactional services, usually where a Presidential priority is at stake. While they tend to leave the overall technology strategy and spend to the CInfOs, the addition of Digital Service teams erode the notion that one person has “ authority over the user experience across all government online services and the power to direct all government online spending,” as Martha Lane Fox advised.

This is not necessarily a bad thing. At the risk of over-generalizing, Digital Service team leads and team members often come from careers at successful consumer Internet companies and bring a user-first, iterative approach to these projects, much like the approach Code for America is known for. (This is not surprising given that I was the original author of the Digital Services playbook and instigator of the USDS during my year at the White House, along with many other stellar contributors, who have taken this work much further and made it much better since my return to California!) Digital Service teams also typically have the backing from executives to challenge conventional wisdom around what Mike Bracken and his team call “government needs” — the seemingly bottomless pit of compliance with government policies and regulations that make it so hard to make digital services that are, in Code for America parlance, “simple, beautiful, and easy to use.” To be clear, they operate within that context and comply with same regulations that other teams do, but they (and others without that title) question the interpretation of everything from the Paperwork Reduction Act (PRA) to the Federal Acquisition Regulation (FAR) with the intention of the law or regulation in mind; they’ve also built infrastructure like cloud.gov that makes compliance much easier, so they can focus on what users need from them instead of what government needs.

Cities are now getting on the the Digital Services team bandwagon too. I’m proud to say that Code for America alums are behind a few of them. With the help of Ashley Meyers, a Code for America alumna serving as a Product Manager for the city, San Francisco just adopted a Digital Services Roadmap which calls for the appointment of a Digital Services Officer. Alicia Rouault, a CfA staffer and a 2012 fellow before that, recently stepped up to lead Metro Boston’s new Digital Services Group, housed inMAPC’s Data Services Department, serving 101 cities and towns in the Boston area. This is the first position of its kind in regional government in the country, and presents anopportunity to scale regionally without losing a localized approach to digital services.

In cities where the basic needs are met (email is up, payroll goes out on time, etc), some CInfOs are building great digital services right along side back office and mission IT and infrastructure. The CIO of Boston, Jascha Franklin-Hodge, has a decidedly civic tech background, as one of the founders of Blue State Digital, and has handled both the traditional CIO duties and made great digital services a priority, bringing in Lauren Lockwood as the Chief Digital Officer, reporting to him. Jonathan Feldman in Asheville, NC, has similarly handled the nuts and bolts of Asheville’s infrastructure while leading his team to create great customer experiences like Simplicity . But the need to create a separate unit for digital services is a reflection of how hard this can be to do within the priorities of a CIO office. In cities where the leadership feels particularly behind the eight ball because the basics don’t work, those conflicting priorities can make for a difficult search for a technology leader. Leadership knows they are looking for a change agent; they know the game most government CIOs have been playing is a losing one, and they’re looking for someone eager and able to play by a different set of rules with vendors, procurement, hiring, and more. But when that role is framed as a CInfO, it’s likely that the first priority of the job will be to negotiate large contracts for infrastructure and back-office with incumbent vendors. You might as well stamp the position description with “change agents need not apply.”

So if you’re a government leader hoping to recruit a senior technology leader who will bring your city/state/agency into the 21st century, what position SHOULD you be hiring? Obviously, that is going to depend on your circumstances. In the end, it may not matter all that much what you call the top technology job in government if you’re clear on what you want that person to do and what authority he/she is going to need to get it done. In most cases, you should consider recruiting two top people who have some combination of responsibility across the four domains we started with. If they are like-minded and clear on who’s got the ball on which reforms, it may not matter who reports to whom. If you can’t be sure of that, reporting structure may matter a lot. In some cases, you will find a great Chief Digital Services Officer who is happy to report to a CIO because that person will give them air cover they could not otherwise get. In some cases, you can get a find a Director of IT, likely focused on back office and infrastructure, who is happy to report to a Chief Digital Officer because IT will get more visibility as it’s more clearly connected to public wins. The one deadly combination is one leader protective of the status quo, and another hell-bent on modernization.

Getting the responsibilities, skills, leadership characteristics, and reporting structures right is a necessary step to recruiting the right person or people. Ideally, the process of figuring that all out would start before the search has begun, but in most cases that I see it doesn’t, and that’s okay. This is pretty new territory for government leadership, so figuring it out while doing it is to be expected. The only thing that IS deadly is NOT reacting to what you learn doing the process.

In my next post, I’ll tackle the role of the Chief Innovation Officer.