One year later, a look back on the growth of 18F

Share

Written by

Nearly a year ago to the day, Greg Godbout was visiting Disney World. But instead of traversing the park, meeting Mickey and his pals, or screaming as he plummeted through Space Mountain, he was calling cohorts from his recently finished Presidential Innovation Fellows class. Godbout didn’t know then the Pandora’s box he was opening, one that would become the hottest thing in federal IT and a basis for how government would begin to expect digital services — he was recruiting the first employees of 18F.

“Hey, your ‘PIF-dom’ is ending,” Godbout, the 18F executive director remembered telling his PIF cohorts. “You’ve got to come do this, we want to do something permanent.” What exactly “this” was, he wasn’t sure. “It was very rough and spitballing, like what do our processes look like and stuff like that?” he told FedScoop.

Looking back, Godbout said after his initial team conceived a vision for the General Services Administration’s 18F, which was then called GovX, they were spot on with some things and somewhat off with others.

Ask anyone what 18F’s biggest achievement in its first year was and they’ll probably say something involving agile development. That’s what Godbout and his team thought, too, but they were a bit off.

“I’d put that as like the No. 2 change. It’s still huge, it’s still enormous,” he said. “But without a doubt, no question, the No. 1 biggest problem that government should be addressing right now is user-centered design versus stakeholder-centered design. As that kept happening and I was going around and talking to people in government, I kept thinking ‘In almost none of these projects do they talk to an actual end user.'”

On every early project 18F worked on, Godbout said there was a major pivot when the end users were involved — not a bad thing, but something agency stakeholders resisted.

“People did not want to go back to the stakeholders and say a change was going to be coming,” the director said. “That is where we met a lot of contention early on in our projects, every single one of them.”

But predictably, he said with the cultural aspect of 18F, they got it right.

“I think we hired the right people,” Godbout said. “I think that’s what it comes down to, we got the right group of people.”

That’s arguably the biggest challenge for the federal government: hiring a talented workforce. Yet Godbout said that hiring is one of the team’s biggest achievements so far and that there was never really an active recruitment process. Fittingly, 18F’s 100th member joined the team in December.

By January, 18F stood at about 15 members. That number stayed stagnant until May, when, as Godbout put it, they “fixed hiring.”

“We enter May with 15 and then we exit October with like 90,” he said. Elsewhere in government, an agency is lucky to hire someone anywhere between six to nine months after they apply. But at 18F, they knocked it down to six weeks, and they’re shooting for four weeks. “We kept iterating the hiring process, and those iterations finally started to kick in in May, and there was a backlog of people who were waiting to get through that hiring process,” Godbout said.

 

I think we hired the right people,” Godbout said. “I think that’s what it comes down to, we got the right group of people.

But before they could hire talented technologists at freakish rates, 18F had to have them interested in working in the federal government, and that’s where the culture became most important.

“If you’re going hire, you better get not only good at the process and figure out what authorities you’re going to use, but most importantly build the culture,” Godbout said. “Looking back, I keep coming around to if there’s anything I’m thinking about the future, it’s: Yeah, the culture really is important. Like, incredibly important, and in fact it’s probably the answer to our success if I had to pick one.”

But doing so was complicated; it required clearance from above. And it wouldn’t work, Godbout said, unless the higher-ups gave 18F the independence and autonomy they craved: “to invest in our own work, to feel like we control our work and we’re driving it,” he said.

Daniel Pink’s “Drive,” a 2011 book revered by many as a manual for interpersonal relationships, lists three principles for business motivation: autonomy, mastery and purpose. Godbout made sure to fit each of the others in addition to autonomy.

On mastery: “Mastery is important. We work with open source, we’re using modern technologies,” Godbout said. “You could say ‘Let’s bring in a bunch of technologists into the federal government,’ but if they’re going to come here and work on COBOL — and that’s an extreme example — it’s going to be hard to convince them to take the job, even if you made a great environment for them.”

On mission: “Of course government has mission everywhere,” he said, ” and that’s our one big advantage.”

It couldn’t just be a pure team of technologists, though. Godbout said they needed a mix, “not just your developer, your UX person and your project lead, but your finance person, your HR person — all of these people are tied to the outcomes of the program office of the business unit. I don’t think you accomplish this unless you do that.”

A year later, with other digital services production floors like 18F popping up all around — the U.S. Digital Service in the Office of Management and Budget, and the Department of Veterans Affairs Digital Service Team — Godbout doesn’t think 18F was the tipping point. That happened long ago, he said.

“In a marketplace where we can’t possibly serve the demand of people coming to us, there’s something wrong with that marketplace,” Godbout said. “You realize it was as much a tipping point that you inspired as it was frustration at not being able to get to these services,” an “opening of Pandora’s box.”

And for that reason he expects “a massive proliferation of these methodologies in the next year, not only spreading out to other agencies, but just in general new standards and expectations. If we can hire in six-to-eight weeks, why can’t all the other agencies?” he said.

Godbout sat in on the creation of USDS and VA’s team, and he has his finger on the pulse elsewhere around government through agencies’ requests to help build similar in-house teams. He thinks more of these teams are going to be created quicker and quicker.

“It’s not because we’re good at selling it, though I’d like to give us credit for that,” he said. “I think it’s because they already wanted it, and it frustrated them, and the marketplace wasn’t there to provide it. And now it’s there.”

-In this Story-

18F, Agencies, Archives, data analytics, Defense Advanced Research Projects Agency (DARPA), Department of Defense (DOD), Department of Justice (DOJ), Departments, General Services Administration (GSA), Government IT News, Greg Godbout, Innovation, open data, open government, Tech
TwitterFacebookLinkedInRedditGoogle Gmail