Harvard Students Hack Government with Empathy

What would happen if you took 25 grad and college students — from different schools, backgrounds, and ages — and expected them to tackle real government challenges in 13 short weeks?

This was my third time teaching DPI-663, Tech and Innovation in Government, at the Harvard Kennedy School, a field class that I run in partnership with government agencies like the City of Boston, the Census Bureau, and the U.S. Department of Veterans Affairs.

While the class has worked in the past, I’m always on eggshells as we approach Demo Day. This day is the culmination of the class, where five-person student teams present their field research, digital prototypes, and recommendations to fellow classmates, clients, Harvard professors and classmates, and the public.

Our most recent Demo Day was Friday, April 27, 2018, and the presentations and demos were just awesome!

Here is what we saw:

• A student team (Ayna Agarwal, Chris Kuang, Ben McGuire, John La Rue, and Molly Thomas) working with the Boston Public Schools presented six ideas to improve the parent experience, including improvements to the website and an internal customer relationship system. See their mid-semester presentation, blog, and story.
• A student team (Jackie Chea, Thad Kerosky, James Moffet, Erica Pincus, Jonathan Truong) working with the City of Boston on public records built a prototype to help requesters make more successful requests, advocated for a public online reading room, and helped the city with a RFP for a public records system — and had additional suggestions to make the overall city procurement process for software to be more user-centric. See their earlier mid-semester presentation, final presentation, blog, and story.
• A student team (Tony Thumpasery, Arjun Bisen, Ayush Chakravarty, Daniel Drabik, and Carissa Chen) working with the Census Bureau, part of the U.S. Commerce Department, developed a prototype for the new Census data website, so that different types of users can more easily find the data they need. They also had great ideas about how Census could engage with Google and other search engines and other third parties where people encounter Census data. See their blog and story .
• A student team (Chris Dylewski, Ifedayo Famojuro, Simon Jones, Rebecca Schwartz, and Irene Solaiman) working with the U.S. Department of Veterans Affairs on service members transitioning to Veteran status had many ideas on how to improve the process and make it more human, including a simple personalized web application that would help servicemembers find the most relevant resources. See their mid-semester presentation, final presentation, blog, and story.
• A student team (Dani Cinali, Brandon Lee, Amelia Sampat, Matthew Spector, and Katherine Spies) working with the U.S. Department of Veterans Affairs on supporting caregivers had 42 prototype ideas — including a pill dispensing robot! They built a caregiver portal prototype to help Veteran caregivers get the resources they need. See their mid-semester presentation, final presentation, blog, and story.

After Demo Day, in the following weeks, students then visited with the Boston Public Schools and City of Boston senior executives at Boston City Hall, and White House, VA, and Census senior executives in Washington DC.

A bit more about the students:

• A majority are enrolled at Harvard Kennedy School (HKS), with others in Harvard’s College, Engineering School, and Design School — with a couple dual degrees at Tuck and Stanford GSB. In past years, I’ve had HKS/HBS students and HKS/MIT students.
• They come from across the U.S. and the globe. This class included students from China and the UK, and past years include students from Israel, Egypt, Australia, and Japan.
• They have a variety of different backgrounds, ages, and professional experiences, by design. Some students come from politics, government and the military, others from businesses, academia, and non-profits. A few are still in college. Some had design, product, or software development experience — but the most important things they all had in spades: team-orientation, grit, and empathy.

What do students of DPI-663 learn?

• User-research and design thinking. Too much of government is removed from the lived experience of beneficiaries, citizens, and front-line government workers. In this class, students go out into the field, ask open ended questions, observe behavior, and use the best practices of user research.
• Synthesis of user research. Dozens of in-person interviews and observations are real data. But to make that data useful, teams needed to turn that data into representative groups of people (personas), stories of how people experience a government service end-to-end (journey maps), or how different groups relate to each other (stakeholder maps). Students synthesized their research into presentations of user insights just before spring break.
• Brainstorming, prototyping and user-testing. Building on their user insights, student team brainstormed ideas about solutions and interventions that might meet the user needs they identified. Starting with paper prototypes and moving to digital ones, students get a chance to take their ideas and test them with real people — quickly learning what works and what doesn’t.
• Teamwork. Inevitably students rate the team dynamic as one of the most challenging but also one the most rewarding experiences they’ve had at school. I’m a big believer in small interdisciplinary teams.
• Story-telling and communications. The best designers, engineers, and product managers can figure what to build and go do it. But if they can’t tell a simple and compelling story about it, it’s harder for them to get people excited about their recommendations, especially beyond an early prototype. I require the students to blog several times during the class, about their project. As someone who has written many blog posts inside and outside government, it’s harder than it looks. It’s hard to make it immediately interesting and hook a reader, make it widely accessible to the average reader, and completely devoid of jargon. I want students to have a personal public record of their work, as well as understand viscerally why working openly can be an important change-agent tactic.
• Hacking the bureaucracy with empathy. Ultimately, I want the students to learn more than just digital product management in government. I want them to unleash their inner change-agent, bureaucracy hacker, and public-sector entrepreneur (hat tip to my HBS colleague Mitch Weiss for advancing this idea). I want them to understand how bureaucracies really work (hint: understand and build trust with the people that work in them!) and learn to be effective in catalyzing change — even as a small external team of students.

DPI-663 Students: the real deal

Students in my class continue to amaze me. Over the course of three years, alumni of DPI-663 have:

• been hired by the U.S. Digital Service, by the Commonwealth of MA, by the City of Boston;
• taken jobs with top tech companies in Silicon Valley;
• written a book explaining tech;
• founded a college club, the Harvard Open Data Project, to hack on Harvard University data;
• founded a non-profit, Upsolve, to help low-income Americans with Bankruptcy;
• founded a non-profit, Coding It Forward, to place tech student in the federal government; and
• won the party nomination and is currently running for Congress this year!

You might ask, why do I teach this class?

• The best school of government in the world — the Harvard Kennedy School — needs to teach these topics, and the future leaders of governments around the world need to realize that understanding people is key to ensuring government really serves its people. If we aren’t teaching design, product management, tech and data fluency, and public-sector entrepreneurship — then we are doing a disservice to the leaders of tomorrow.
• Because students learn by doing. We can teach them about how digital is changing policy and service delivery — but there is no better way to internalize it than rolling up your sleeves and learning the hard way.
• Selfishly, because I miss the mission, scale, scope, complexity, and responsibility of government — and this is a chance to help by staying involved. I love the curating, matchmaking, and trouble-shooting with teams.
• Because they let me! A huge thanks to David Ellwood, Doug Elmdorf, Archon Fung, Suzanne Cooper, Nicco Mele, Nancy Palmer, David Eaves, Vanessa Rhinesmith, Dan Schrag, Richard Parker, Steve Kelman, and many other colleagues at HKS. It’s been an honor to be a part of the Shorenstein Center (researching and writing about open government data) and Digital HKS (where I led a seminar on hacking the bureaucracy last semester), and I love collaborating with other colleagues at Belfer, Ash and other parts of HKS.
• And of course, because I get a kick out of every time someone says Professor Sinai with a straight face!

Despite it being the third time around, I still learned a ton from the students, course assistants, and visitors to the class — including:

• Students are relentless in their desire to make government better. They come up with creative ways of getting outside of the Harvard bubble, and are full of ideas about how to modernize government for the people it serves.
• Many students, especially in the Kennedy School, have a bias towards quantitative surveys and operational data, rather than the messier data from interviewing and observing people. By the end of the class, students see the value of the design process and how it can strongly complement the more traditional data sets they are used to analyzing.
• Students worry about their ideas being sustainable and being carried forward. They are cognizant of how hard it is for their rough prototypes and ideas to really stick after the semester — as bureaucracies sometimes seem designed to repel outsiders. But the City of Boston is moving forward with a public transparency tool based on the student input, and plans to modernize a real estate tool on Boston.gov based on a past project. The Boston Public Schools is already moving towards simpler and more functional digital services. And the Census Bureau has told me that this class is helping shape a planned redesign of Census.gov.
• Real-time feedback is invaluable. I make the students do a “one-minute test” at the end of each class where they respond to a few questions (specifically: what is the biggest takeaway from today? What surprised you most? What is your biggest remaining question?) and typically read and respond to it over the weekend. In retrospect, I should also do it too!
• Students will do amazing things when challenged. While the students have a lot of other things going on at Harvard, they keep stepping up and delivering for their clients.
• Teaching is humbling and awesome — and a great privilege! I re-learn this every year, as I work to make the class a great experience for the students, and realize that each year I can make it better.

This class would not be possible if it weren’t for the awesome course assistants, research assistants, guest lecturers, and other student, alumni, and faculty contributors.

• Dana Chisnell and Mina Hsiang guest lectured this year — on user research, and Healthcare.gov and bureaucracy hacking, respectively.
• Stephanie Nguyen and Brian Lefler, both former U.S. Digital Services members, were amazing course assistants this year, building on the legacies of past course assistants Angel Quicksey and Chante Lantos-Swett.
• Alisha Ukani and Hadley Debello worked tirelessly behind the scenes, as research assistants, and helped the course run smoothly.
• Past contributors like Erie Meyer, Ryan Panchadsaram, Marina Martin, Eric Waldo, Ben Willman, Mary Ann Brody, and Kate Krontiris have been critical in bringing their experiences, perspectives, and mentorship to the class.

A special thanks to the clients of DPI-663, who have invested time and energy in making the class a success. The challenge of scoping a problem that is relevant, immediate, and accessible to outside students is a formidable one. I’m forever appreciative of the government officials who spend the time coming to class and working closely with the students during the semester.

So what’s next?

• I’m going to teach in the spring of 2019. If you are a prospective client, drop me a line a nick_sinai at hks.harvard.edu. I tend to gravitate to government units that are ok with ambiguity, and those that have an interest in understanding the underlying problem from the perspective of end-users and line employees (i.e. they dig human-centered design). The people at the heart of the problem — whether they are external customers or internal employees — need be local to the Boston area, since it is hard for the students to regularly travel during the semester. We expect government agencies to devote real time and energy to the class, make their agency accessible to the students, and include senior executives directly in the class.
• If you are a Harvard student interested in the class for 2019, here is the application. No experience is necessary, but I’m interested in what you bring to the class, and if you are going to make this class a priority.
• The students are going onto amazing things — please hire them!

While it’s already summer, and the students are in the middle of internships, travel, and starting new jobs — I’m already getting nervous about Demo Day 2019. But I’m confident the students will continue to surprise and impress!