Hacking for Diplomacy: What we learned with the State Department

Photo: U.S. Department of State
Photo: U.S. Department of State

We just held our final week of the Hacking for Diplomacy class, teaching students entrepreneurship and “Lean Startup” principles while they engaged in national public service applying advanced technologies to solve global challenges. Seven student teams delivered their final Lessons Learned presentations documenting their intellectual journey over just 10 short weeks in front of several hundred people in person and online. And what a journey it’s been.

In this class, we partnered with sponsors in the State Department including:

  • Office of Space and Advanced Technology
  • Bureau of Political-Military Affairs
  • Bureau of Conflict and Stabilization Operations
  • Bureau of Counterterrorism and Countering Violent Extremism
  • Bureau of Population, Refugees, and Migration
    • Office of Assistance to Europe, Central Asia, & the Americas
    • Office of Assistance to the Near East
  • Office to Monitor and Combat Trafficking in Persons

Our sponsors treated our students like serious problem solvers who could contribute unique technical skills and unfettered customer access. In exchange the sponsors got access to fresh ideas, new technology and a new perspective on serious problems.

By the end of the class our sponsors inside State had experienced a practical example of a new and powerful methodology which could help them better understand and deal with complicated international problems and apply technology where appropriate.

And finally, our students learned that they could serve their country without having to put on a uniform. Today, if college students want to give back to their country, most think of Teach for America, the Peace Corps, or Americorps or perhaps if you wanted to offer your technical skills, the U.S. Digital Service or the GSA’s 18F. Few consider opportunities to make the world safer with the State Department, Department of Defense, Intelligence Community or other government agencies.

Lessons learned – Not a demo day

Silicon Valley folks are familiar with demo days – presentations where the message is: “Here’s how smart we are right now.” That’s nice, but it doesn’t let the audience know, “Is that how smart you were three months ago, did you get smarter or dumber? What did you learn?”

Hacking for Diplomacy Lessons Learned presentations are different. Each team presents a two-minute video to provide context about their problem and then presents for eight minutes about the Lessons Learned over their ten weeks in the class.

As an example, Team Trace worked with the State Department Office to Monitor and Combat Trafficking in Persons. The team was challenged to help companies push policies of responsible business lower down the supply chain. The key thing to note in this presentation is not only that the team came up with a solution, but also how in talking to 85 people, their understanding of the problem evolved, and as it did, so did their solution. (see Slides 12 and 25).

If you can’t see the video click here.

If you can’t see the presentation click here.

Team Hacking CT was sponsored by Bureau of Counterterrorism and Countering Violent Extremism with the goal of deterring individuals from joining violent extremist groups. After 100 interviews, the team realized that a bottom-up approach, focusing on support for friends and family of those at risk for radicalization, might be effective.

If you can’t see the video click here.

If you can’t see the presentation click here.

Getting Lean

Each of the teams used the Lean Startup methodology. For those new to Lean, the process has three key components.

First, students took the problems they got from their State Department sponsors and transformed those into what we call hypotheses.

For instance, one problem was: “We need to improve coordination among all the organizations trying to help Syrian refugees.”

That’s a big, unwieldy problem. Students had to break it down into a series of hypotheses. They had to identify who were the beneficiaries and stakeholders, and think about what specific service they were going to provide them, how they were going to get it to them and who was going to pay for it.

To help them do that, we have them map their nine critical hypotheses onto a single sheet of paper called the Mission Model Canvas.

aggregatedb-mission-model-canvas

Then, in step two, the teams got out of the classroom to test these hypotheses through interviews with people in the real world. Every team spoke to close to 100 potential “beneficiaries,” partners and stakeholders including NGOs, tech company executives, supply chain managers, foreign service officers in embassies around the world, and even refugees.

While the students were interviewing, they also employed the third piece of the Lean methodology: building the solution incrementally and iteratively. These solutions, called Minimal Viable Products, are what allow the teams to become extremely agile and responsive.

As teams talk to stakeholders they gather evidence to either validate, invalidate or modify their hypotheses. If they find out that their assumptions are wrong (and almost all do,) they Pivot, that is, they make fundamental changes to their hypotheses, instead of blindly proceeding forward simply executing a plan. This ability to gather data, build and test MVPs, and then change course is what gives Lean it’s tremendous speed and agility to deliver rapid solutions that are needed and wanted.

As an example, Team Aggregate DB was working with the State Department Bureau of Conflict and Stabilization Operations (CSO). CSO helps embassies and diplomats to visualize, understand, and stabilize conflict. The team’s challenge was to get helps embassies and diplomats get more information about informal leader networks. Getting out of the building and talking to 87 people gave the team got a firsthand view of the downside when an embassy does not have access to the right local contacts. (Slides 3-9)

If you can’t see the video click here.

If you can’t see the presentation click here.

As they developed MVPs, our students took these solutions out into the real world for feedback. At first the solutions were nothing more than drawings, wireframes or PowerPoint slides. As they came to understand their problems more deeply, they refined their solutions into the final products we saw.

h4dip-mvp

For example, Team 621 – Fatal Journeys worked with the State Department Bureau of Population, Refugees, and Migration. The team’s challenge: how to get more data on missing or perished refugees. In this presentation, note how the team’s understanding of the problem evolved over the course of talking to 88 people. They realized there was a missing link between key stakeholders that limited identification of perished refugees and prevented emotional and legal closure for their families. The team pivoted three times as they gained deeper and deeper insight into their problem. With each pivot, their solution radically changed. (Their first pass of problem/solution understanding is on Slides 1-29, but then they get additional insight in slides 36-50. Finally, slides 51-64 is their third and final iteration).

If you can’t see the video click here.

If you can’t see the presentation click here.

Hacking for Diplomacy was profiled this week in the L.A. Times. We’ve also had L.A. Times> Beijing bureau chief Julie Makinen, who is on a JSK journalism fellowship at Stanford, helping coach students this quarter on interviewing and research techniques. Julie has been sharing her impressions of the class on this blog. Here’s her last installment:


In the Netflix age, suspense is an increasingly rare commodity. If we’re intrigued by an hour of “House of Cards,” we need not delay gratification – we can just queue up the next episode and push play. But following Stanford’s Hacking for Diplomacy class over the last 10 weeks has been like watching a TV drama the old-fashioned way. There were cliffhangers every time, and you had to wait seven days to find out what would happen next.

The class, which meets just once a week but requires massive outside work, is run not as a traditional lecture where professors drone on in front of passive students — just the opposite. It’s the students standing up in front, discussing what they’ve found out in the past seven days, what progress they’ve made, what obstacles they’ve run smack into. The teachers sit in the back row and lob questions and critiques forth — sometimes very direct critiques. That format keeps students and teachers alike on the edge of their seats.

Conflicts and misunderstandings within student teams — and between students and sponsors — cropped up as the students tried to learn about the State Department, their sponsors problem, and Lean Startup methodology all at once. Students, teachers, and interviewees said surprising, intriguing, even stunning things. Some days, you could see teams going off the rails, but instead of just shouting at your screen, “No, don’t go down that alley!” a professor would actually speak up from the back with something blunt like, “You’re way off track, and we’re firing your idea.”

And just when you thought a team had struck upon a brilliant notion for a product, they’d report back during the next session that everyone they put it in front of hated it. I started looking forward to each Thursday at 4:30 p.m. like my parents looked forward to watching “Dragnet” as kids, because the suspense was killing me.

Thursday’s season finale did not disappoint. Teams that just two or three weeks ago seemed to be foundering pulled off some amazing comebacks.

Take Team Exodus, which had spent a substantial part of the quarter focused on how to match private companies seeking to assist Syrian refugees with NGOs working in the field. Late in the term, the students scrapped that idea after finding competitors who were already deeply engaged in that space. They did a major pivot and decided to concentrate directly on refugees as customers — building on all they had learned during their first eight weeks of interviewing and research.

In week 9, they decided to build an AI chatbot on Facebook’s Messenger platform to allow refugees to ask questions like, “Where can I get clothing?” The bot will tap into a network of NGOs to source answers. A very basic prototype, built primarily by team member Kian Katanforoosh, a master’s student in computer science and management science & engineering, is already up and running.

On the eve of Thursday’s class, team members Katie Joseff and Berk Coker had a call with the United Nations’ refugee agency, UNHCR, and learned that the organization was very interested in working with the students to bring an Arabic chatbot to the field, most likely starting in Jordan.

“At the end, our team kind of came out of the weeds,” said Joseff, an undergrad majoring in human biology. “We finally got to the thing that Steve Blank talks about – where you can see the whites of a customer’s eyes and they just really want the product you’re talking about.”

Team Exodus: Coordinating information to better serve refugees

If you can’t see the video click here.

If you can’t see the presentation click here.

Watching the students’ process and progress was an eye-opener even for many State Department sponsors and private-sector mentors. Team Space Evaders, who in week 7 seemed to be vying for the title of Team Whose Proposed Products Generated the Most Yawns from Potential Customers, had an “ah-ha” moment and decided instead of focusing on tracking objects already in space, they’d pivot and concentrate on objects that will be launched in the future.

They’re proposing a “debris footprint” that would rate satellites before they’re sent into orbit on how much space junk they could generate. The team hopes that this could lead to international design standards to reduce space debris.

“They had a fundamental insight – don’t track ’em, solve it before they even get into space,” said Jonathan Margolis, deputy assistant secretary of State for science, space and health who came all the way from Washington to meet with the team and sit in on the class in Week 9. “It’s a reconceptualization of a problem we’ve really been struggling with.”

Team members Dave Gabler, a master’s student in business and public policy with an Air Force background, and Matthew Kaseman, an Army vet and freshman in aerospace engineering, said the next step is to produce a white paper that fleshes out the mathematical formulas that could underpin a ratings system, then take that to academic and industry conferences. “That would help start a public discussion and push the debate,” said Gabler.

“The math is probably the easy part,” said Pablo Quintanilla, a former Foreign Service Officer and current head of public policy for Asia for Salesforce, who served as mentor for the Space Evaders team. “There’s so much more to the behavioral side – who in the international space community will adopt this?”

Quintanilla said that working with Space Evaders drove home for him the merits of forming diverse teams to tackle problems. Besides Gabler and Kaseman, the student team included Kate Boudreau, a junior majoring in biomedical computation, and Tyler Dammann, a junior in computer science.

“This cross-functionality and working across disciplines is really effective,” Quintanilla said. “I feel like this is living proof that you should work everywhere like this.”

Team Space Evaders: Reducing space junk

If you can’t see the video click here.

If you can’t see the presentation click here.

Professor Jeremy Weinstein, a co-instructor for the class who recently served as deputy U.S. ambassador to the United Nations, acknowledged that 10 weeks is a really short time frame for the students to make any meaningful impact. But unlike an internship, where a lone student “plugs into an existing bureaucratic hierarchy and rules,” the Hacking for Diplomacy students had the advantage of being able to work in teams — and approach the problem more as outsiders.

“The students don’t have to play by the same rules [as insiders]. They can ask the non-PC questions,” said Weinstein. “To be ignorant of the rules is a blessing at times — if you can do it respectfully.”

Getting students to have a healthy appreciation for how government policy is made — sometimes painfully slowly — is part of the educational process. And so perhaps is getting bureaucrats to be more open to fresh ideas. “There’s not going to be a flip of the switch” in State as a result of this class, Weinstein said. “There is some skepticism. But I think more broadly, we’ve won some people over.”

Thursday’s wrap-up session attracted a diverse audience, including representatives from leading Silicon Valley tech companies as well as diplomats from France, Britain and Denmark. Susan Alzner, head of the U.N. Non-Governmental Liaison Service’sNew York office, said after watching the student presentations, she wants to take the customer discovery and interview methodology back to her agency.

“The U.N. has lots of small teams of people who often believe they already know the solution to a problem. … And the U.N. does way too much consultation digitally. Interviews are critical. It’s so elaborate to see these students doing 100 interviews to understand a problem, but it’s so important to orient yourself before making a plan to do something.”

Team Hacking 4 Peacekeeping: Better data on, and decision-making about, peacekeeping forces

If you can’t see the video click here.

If you can’t see the presentation click here.

Weinstein noted that scaling up Hacking for Diplomacy may not be as easy as expanding Hacking for Defense, simply because “there are millions of people who work in the Department of Defense… while the size of the State Department foreign service corps is smaller than the total number of people who play in military bands.” That means there are fewer people who can serve as sponsors.

At the same time, the class could tap a wider array of sponsor organizations. “Scale maybe has to look different — we can look to the [State Department], but also UNHCR, the foreign ministry of the U.K., other international organizations,” Weinstein said. “You have to think of a different array of partners.”

Most of the State Department sponsors for this year’s class, Weinstein noted, were not political appointees but career foreign service officers or career civil servants.

“They are the glue that holds the agency together and they are key to getting anything implemented in government. And so the buy-in is there,” he said. “But they also need permission; they need a blessing to experiment with radically different ideas. And you need political cover in these bureaucracies to do this kind of work.”

“I hope,” he added, “we’ll have that cover in a subsequent administration. More than cover. Endorsement. Enthusiasm. Excitement.”


Our teaching team

Like the students’ efforts, the teaching of this class was also a team project. I was joined by Jeremy Weinstein, former deputy to the U.S. ambassador to the United Nations and a Stanford professor of political science; Zvika Krieger, the State Department’s representative to Silicon Valley and senior advisor for technology and innovation; retired U.S. Army Col. Joe Felter, who co-created Hacking for Defense and is a senior research scholar at the Center for International Security and Cooperation at Stanford; and Steve Weinstein, chief executive of MovieLabs who teaches entrepreneurship at Stanford and UC Berkeley.

h4dip-instructors

Our teaching assistants were Shazad Mohamed, Sam Gussman and Roland Gillah. We were fortunate to get a team of seven mentors currently or formerly served in the State Department and selflessly volunteered their time to help coach the teams. Each team also got a mentor from the tech industry who helped guide them through creating their final products. Of course, huge thanks to the Stanford students who gave their all through this class.

Going forward

While our previous Hacking for Defense class gave us a hint that doing the same for Diplomacy would work, we’re a little stunned about how well this class with the State Department went. A surprising number of students have decided to continue working on foreign policy projects after this class with the State Department or with NGO’s. Other colleges and universities have raised their hands, and said they want to offer Hacking for Diplomacy or potentially a USAID Hacking for Development class at their school.

Meanwhile our Hacking for Defense class continues to scale through H4Di.org the non profit we set up to curate the problems from our sponsors (JIDO, ARCYBER, AWG, USMC, NSA, AFNWC, SOCOM, 75th Ranger Regiment, USTRANSCOM, Cyber Force Protection Brigade, National Defense University, and the Center for Technology and National Security Policy). And H4Di.org supports the universities teaching the class this year: Stanford, UC San Diego, Georgetown, Air Force, University of Pittsburgh, James Madison University, Boise State, and RIT.

If you’re interested in offering Hacking for Diplomacy (or Defense) in your school, or if you’re a sponsor in a federal agency interested in solving problems with speed and urgency, join us at our next H4D educators class January 17-19th at Georgetown.

Lessons learned

  • Our sponsors inside State saw examples of a new and powerful methodology – Lean which could help them better understand and deal with complicated international problems
  • Lean offers State speed and agility to deliver rapid solutions that are needed and wanted
  • Our students learned that they could serve their country without having to put on a uniform
  • Other universities are willing to have their students work on diplomacy and development problems
  • The class was a success

(This post is a continuation of a series. See all the posts about Hacking for Diplomacy here.)

About Steve Blank

Steve Blank is a retired serial entrepreneur-turned-educator who is changing how startups are built and how entrepreneurship is being taught. He created the Customer Development methodology that launched the lean startup movement, and wrote about the process in his first book, The Four Steps to the Epiphany. His second book, The Startup Owner's Manual, is a step-by-step guide to building a successful company. Blank teaches the Customer Development methodology in his Lean LaunchPad classes at Stanford University, U.C. Berkeley, Columbia University and the National Science Foundation. He writes regularly about entrepreneurship at www.steveblank.com.

Comment