We Must Run Government IT Like A Startup

A recent piece on InformationWeek.com by David Carr pointed me toward the federal government's HowTo.gov and the site's digital strategy guide. It's a well-intentioned body of work consisting of all of the consultant models, numbered steps, PMP-type milestone tables with action owners and, of course, acronyms. It includes a quote from President Obama: "I want us to ask ourselves every day, how are we using technology to make a real difference in people's lives?"

Indeed. But making a difference in people's lives through innovation isn't just about technology; it's also about throwing out flawed process models and adopting new ones.

I agree with Carr, who wrote: "If you work in government, be prepared to lose control. If you don't go out and meet social media head-on, it will happen anyway, all around you."

[ Find out how the federal government innovates with less. Read Busting Through The Federal IT Budget Ceiling. ]

Most government entities are simply publishing information; they're not engaging in discussions with constituents. It's not social if there's not a two-way exchange of information. Please don't tell me you're not resourced for it. That's like a private-sector business telling me it's not resourced to handle those annoyances called customers.

Carr's assertion that government entities could lose control is just the tip of the iceberg. Government needs to completely change. Yes, technology is involved, but the way governments use technology also needs to change. It's not just about using social media. It's about decentralizing. It's about instituting agile processes. It's about delivering benefits to citizens rather than adhering to bureaucracy.

Governments -- slow, lumbering and unresponsive -- must adopt a startup culture. Old way: command and control. New way: decentralized collaboration on shared goals. Political factors and life in a fish bowl discourage government staffers from engaging in experiments with any risk. After all, if their experiments don't work out, they'll be derided for spending taxpayer money on a boondoggle, a potential career limiter.

Global CIOs: A Site Just For YouVisit InformationWeek's Global CIO -- our online community and information resource for CIOs operating in the global economy.

Within our city's IT organization, where I'm CIO, I've been gratified that some of my staff have gone out, with very little prompting, and gotten educated about startup culture. Some of them participated in a Startup Weekend; others have integrated agile principles into project management.

Here's one recent, incredible example. Using the principles of "minimum viable product," "short, iterative product cycle" and "small experiment" from the lean startup camp, my staff preempted our city's plan for a truck routing system based on off-the-shelf software with a cheaper, less complicated idea of their own.

They set up a small experiment without telling me or any other executives in our organization. They talked with sanitation truck drivers, in effect saying: "Hey, wouldn't it be cool if we could save some fuel? Want to do a small pilot that won't take a lot of time?" The answer, of course, was yes. They marked down the odometer readings on a brush pickup truck for the start and end of a run. Then they plugged all of the address locations for the pickups into our existing geographic information system and ran a route optimization.

It turns out that, with basic route optimization, we could cut mileage by more than 20% on our own! The "minimum viable product" for drivers is merely a set of turn-by-turn directions on a piece of paper. Getting fancy, we could upload the data points into Google Maps for iOS and let it shout out the directions. Point is, we figured out that we probably don't need the capital project -- once we started acting more like a startup.

Government doesn't need to change all at once. If we take iterative steps and acknowledge that everything won't always work out, we can make small changes that will later scale to larger changes. We need to start getting out of the way of our staff and let them start to solve problems. And most important, we have to stop doing things the way they've always been done, and start copying the success of our private-sector brethren.

InformationWeek's 2013 Government IT Innovators program will feature the most innovative government IT organizations in the 2013 InformationWeek 500 issue and on InformationWeek.com. Does your organization have what it takes? The nomination period for 2013 Government IT Innovators closes April 12.