- [John] Alright, so hi. So we're gonna only do an
hour today 'cause our guest needs to jet, and so do I. So what I thought we'd do is
for a few minutes we would talk about a couple of observations, and then we'd have Jenifer
come up and talk about Code for America, which
I will probably call CFA for the rest of the time. And then we'll do questions and stuff. So some of the comments,
it was interesting, some of the feedback we got
after, kind of along the way, was that people wanted
more and more specifics and stories and specific situations. And so we tried hard on Tuesday to do a lot of that with Mozilla, I think. What you'll find is that
as we moved from investors to operators, operators
pretty much talk about what's on their mind all the
time, and operators are in the guts and in the details all the time. And so what I think you'll
find now in Alan's comment as he was headed to the airport is that everybody came up to
him to talk to him about community being the key to growth. And so he called me, and we
talked, and his big concern was, oh my god, are they over-generalizing? You just heard an hour about community. What do we do, like the community is not the only way to grow. And so his point of view
was, we need to help talk... We're gonna kinda keep coming
back to lightweight framers, which is here's kinda
how to think about it. Here's the specfics. Here's kinda how to think about it. Here's the specifics. It's a funny class. We've been trying to sort of
thread the needle a little bit between being like here's how
we think about the structures and the abstractions, and
here's how we operate. But then here's the details,
and we try to curate a very, very thoughtful set of panelists. So hopefully that makes sense. Does anybody have questions
before we get started talking about Code for America? I guess the other thing I
would say is like having said be careful about over-generalizing,
this is a funny week because this is the only week
we've got any non-profits, and both days we're doing
non-profits and really very, very community
oriented organizations. So Mozilla on Tuesday,
Code for America today. And so then that'll be the
end of the non-profits, so the rest will be for
profits, and Eric (mumbling) was a good example of that. All they do is make money at this point. So anyway, we'll just get started. I'm very happy to have
Jennifer Pahlka here. So one disclosure, I'm on
Jen's Board, and have been on the Board for a couple to three years. Jennifer's one of my
heroes because she looked, well I'm gonna let her talk
about it, but she looked at what needed doing a few years back. And she said, you know,
government's not very of the people, for the people anymore. And so she took her
entrepreneurial background and said I'm gonna change that. It came in the context of
starting this non-profit called Code for America, which
I'll let her talk about. And then I think she's changed the world. And for last year, she took
a year off, the year off from her start up to go take
a job in the White House to be Deputy CTO for the President. So we'll talk about what it's
like to leave your start up, what it's like to come back. And now as she's thinking
more and more about how does she scale impact,
how does she make this a durable, lasting organization,
the different types of considerations she has
from when she was starting the organization and getting
it ramped now to what it is to make it durable. So welcome, Jen, to the class. - [Jennifer] Thank you.
(applause) - [John] So here, have a seat.
- [Jennifer] OK, hi. - [John] So what I'm gonna
ask Jen to do first is can you just talk about
what Code for America is? - [Jennifer] Sure. Well, we started, and there
was an answer earlier about Teach for America, we
started as sort of the Teach for America for coders,
but working in government. So it was a service year
program for sort of mid-career technology and design
professionals to come work in government. - [John] What's a service year program? What does that mean? - Oh, this is a Stanford class. People don't know what a
service year program is? - [John] I don't know. - OK, so it's a year basically
that folks take off from their regular careers and
do something that doesn't pay very well but has very
meaningful, and they give back. So we started out by
doing this with 20 fellows in our first year in
three city governments. We worked for Seattle and
Boston and Philadelphia. And it's kind of grown from
there into essentially a program where we help governments
take an iterative, data driven and user centered
approach to problems. So we co-develop applications
with them, help them get an understanding of how you
would do that in a way that's very different from how
government does it normally, and try to build on all of
the outcomes that happen when you're able to do something
like that successfully. - [John] And you mean mostly
not the federal government, mostly cities. - We work with local governments because, as you said earlier, we're
sort of very much about for the people, by the
people, and if you think about what government does, local
government is actually doing a lot of the service delivery. We do increasingly work with
states, and we work with the federal government a number of ways, but it's usually to help
make change that has to cascade down from the federal
government to the states to counties and cities, and
we're working at that service delivery point that happens
with cities and counties. - [John] Alright, so we'll
have to bridge some of the language 'cause it's not
very start up language like service delivery, and that
just means delivering things to governments or services. - Yeah, exactly, products. - Products, exactly. That's a good word. So tell me about the first year. So you had this idea, we're
gonna start this thing, and you started what month? Like what part of the year did you start? - Well, I kinda quit my
job at the end of 2009. - [John] Your job doing? - I was running the Web 2.0 events. That was like a thing. - [John] That was a thing, Web 2.0, yeah. - Web 2.0, it was on the
cover of Time Magazine. - [John] That was also
a thing, Time Magazine. (laughter) - Touché. - [John] Magazines used to exist. - I was always into the
sort of business-to-business tech media stuff, and had this idea. So I was sort of doing it
part time until I could wrap up my last project,
and started at sort of the beginning of 2010 by opening
a call for cities to apply. So it was sort of smoke and
mirrors, and we didn't really have anything, but we
were building something, and we had an offering that
cities could apply to be part of our program. And we spent that year
getting our city partners, getting the funding that
we'd need to actually do it, which was a good thing we got
it 'cause I don't know what we would have done if we hadn't. - [John] So the first thing
you did was go find customers. - First thing we did was
find customers, yeah. - What cities want a little
bit of Silicon Valley magic. - Right. And you know, it's kind
of amazing I think, we were able to put up
a decent looking website very quickly, and it's amazing
how that makes it look like you have something. - [John] That's a good
lesson, like back when the web mattered I guess. - Yeah. - [John] Now you have an app, I guess. - Yeah, so I mean, you know
I think people in local government could go in
and look at this thing, and go, oh, that's a thing. They have a nice website. And so we got, yeah, we
went and got customers, and then we went and got
our teams, our fellows, recruiting technology
design professionals. But kind of that whole
year felt a little bit like you're laying down the tracks
as the train is reaching the part where the tracks need to be, with the funding coming
in and staff coming in sort of just in time for us
to launch the beginning of the first fellowship year
starting in the beginning of 2011. - And you said it was 20
fellows the first year? - Um hmm. - [John] This is I think a
year before I got involved. And so 20 fellows in three
cities, and how did you get anyone to put their hand
up and wanna do that? What people did you try to
recruit, and what did they want? - We definitely were very
lucky, and I think just the way you do it, we
made it a competition. And we have actually used this since then. You're not trying to sell them something, you're giving them an award. So we said cities will apply,
and we will pick the ones that we wanna work with. That was also a bit of smoke and mirrors, but it worked out fine. So you know, the first folks
that we found were these guys in the city of Boston
who ran an organization I've never heard of before,
but it's now somewhat common in city governments. In fact, there's one in Austin,
called the Mayor's Office of New Urban Mechanics. So these guys had already
been creating a space for innovation, and
really a space for risk, within local government in Boston. And what they had said to folks is, we are the risk aggregator. We're gonna do crazy
projects that don't normally happen in government, and if they work, and we're partnering with departments, and if they work, you and the
department get all the credit. And if it fails, we'll take all the blame. And they had this sort of
ability, they had this sort of mandate from the mayor
to be able to do that, who was a very beloved mayor and had very high risk tolerance. So folks like that were kind
of natural first customers. Mayor Nutter in Philadelphia
also just, you know, sort of fire in his belly
and a lot of executive support for it. So they were just the folks
I think who could stand the risk of doing something new. But we made a lot of mistakes that year. And in fact, we initially had four cities. I don't know if you remember this. - [John] I did not. - We were gonna be
working with the District of Columbia as well, and
we had a great change agent partner there as well. And right after we signed the contract, Mayor Fenty lost the
election, and Gray came in, and he fired all of our
fellows at lunch time on their first day of work. - [John] (laughs) When
they were in D.C. already? - Um hmm.
- [John] Yeah, that's cool. - They made it from breakfast to lunch, and they were escorted out by security. - [John] And obviously it's
good to hear these stories 'cause there are so many
commonalities with what we heard in the first couple of weeks
about sort of faking it until you get your first
customer, faking it until you get recruiting, you
know, finding a kind of, we haven't talked about this,
but finding a very special class of customer, like this risk agent. - Yeah. - That's kind of important
because you almost don't want your mainstream customers
'cause they're not ready for all the mistakes. - Yep. - So that's interesting. So it's not the scaling
customers, but it's the early customers. Do you have a question, Chris? - [Chris] Yeah, so the government's paid to fire these people. - Um hmm, they did. Yeah, we had some, and it
remains today that we have a mix of funding from
philanthropy and from governments themselves. - Yeah, one of the things
that comes up is around funding models for non-profits,
and you have some mix of earned revenue, sort of
just actual business model government, and so corporation giving, and then individual giving usually. So ideally, like Mozilla you
get to this sort of nirvana when it's all earned revenue, and you don't have to go raise. But in the early stages
it's a little bit like raising from VCs, which
is you get partners who are thought partners,
and so think we've been lucky thought partners (mumbling) and Knight and other places as well. Do you have a question? - [Voiceover] Yeah, how did you decide how to price your packages? - So the question is how
did you decide how to price your packages? - Honestly, what we did
is we took the amount that we were gonna pay fellows and
added a little bit to that, and threw it over the wall. And we didn't get the full
price from all of our partners. I think Philly and Boston
paid us what we asked, and Seattle sort of paid
us what they felt like, and we took it. And then it all went, that
one didn't work out well. - We'll talk about
productization as we go, too, and then please feel
free to ask questions. I think that's a commonality,
too, which is sometimes you try to say, well
we're gonna base price on what it costs us, plus a little bit so
we have enough cushion. Sometimes you say, well,
why don't we just ask 'em for this much? It almost always seems
crazy to the start up, and depending on the
organization sometimes it doesn't sound like much at all. So sometimes you ask, and
if they say yes too quickly, you know you've underpriced it, which is usually what happens. OK, so good. So how did you know, when's
the first time you felt like you had a product market fit for CFA? When did it feel like you
really had a match between what you were offering
and what somebody wanted? - I think, and this is
gonna get into sort of what our product really is, but
the first time I kind of realized we were onto
something was after one of the fellows in Boston stood
up an application called Discover BPS. And the background for this
is that they were working on actually other apps in
Boston, and there was sort of a larger team there than we
needed because the D.C. thing had cratered, so we just
moved those fellows over to work on Boston. - [John] Well, let's just talk about the process a little bit. So the fellows, they're
groups of four or five people. - Now we just do pretty
much exclusively three, and occasionally four. - [John] And it's usually a
user interface type person and a developer or two. - Yep, yeah. What we say is amongst those
three people you've gotta have all the skills to do a basic web app. You gotta be able to stand something up. - So this is the product, right? This is one of the products. - This is one of the products, right. So I mean, the software
product itself in this case was very simple. It was a website that you would go on, and you would type in, if
you were a parent in Boston, you could go on there, you
could type in your address, the age of your kid and
whether you had any other kids in a Boston public school,
and it would tell you which public schools your kids
were allowed to attend. But the background of this
is that they had recently changed the rules about
which public schools you would be assigned to. And they did it for a great
reason, which is they wanted more kids to be walking to school. But the way they conveyed
this was to print a 28-page brochure in six point type, and mail it out to everybody. And that doesn't tell you which
school your kids can go to. And there was this human
cry amongst parents. There were, obviously, video I remember on the BostonGlobe.com of a parent crying, so frustrated that school
selection time was coming up, and there was really no
reasonable way to figure out which school your kids could go to. And so one of our fellows, Joe
Mahoney, did this full time and grabbed a developer half time. And I think we calculated
it was about eight weeks, yeah, eight to ten weeks
between the time the mayor asked him to do this and
the time he stood up a working prototype. It evolved after that. But it was when Nigel
Jacobs, who was our sponsor and is also now on our Board,
who's our partner in Boston came to us, and he said,
"You realize that if this had "gone through regular government channels, "it would have taken two
years and cost at least "two million dollars to do this?" And I thought, OK, this
is something valuable. And I don't think we even
realized that when we started doing this. So then you see how the
product is not just something that the parents can use. The product is the way that
everybody around that goes, why does that need to
take two million dollars and two years? - [John] Right, so we had a
little bit of a group of us had dinner on Tuesday, and
Jen was going to articulate what the real product is,
'cause it's easy I think when you're kinda wandering
around trying to figure out what's happening, you can
say, well, here's the thing we make, that's the product. Or here's a thing. And we were talking about
this, and trying to get people to say, well, if the outcome
you're trying to get to is people asking why, why
can't it be different, that's an important and powerful outcome. And trying to figure out how
to get to there is kind of key. And it kind of resonated
with me from Mozilla 'cause like one way to think
about Mozilla's product was Firefox web browser, but
another way to think about it is that we were trying to
put pressure under the system to try to force Apple
and Google and Microsoft to behave differently
'cause our mission was a little bit more than
a commercial mission. It was a behavioral mission,
and I think that's in a lot of ways what CFA is, too. So this is I think in real
time we're trying to articulate this now, but that's a nice articulation. - Can I add to that a little bit? The other thing, moment
that came to mind when you asked me that was in the second year, this woman that brought us in Detroit, which is a crazy place to work. Had a great experience with the fellows. I can tell you about the
products, but the point is at the end of the year she said, "You know, I've spent 25
years in public service, "and I gave up on feeling
like I could serve the public. "And now with these tools
I feel like I'm serving "the public again." And that felt somehow
like product-market fit. - [John] Yep, yep. So play it forward to
where we are now in 2015. So how many people are, how do you count how many people are in CFA? - So that's a really good
question, and we have about 45 staff. We currently have 24 fellows. There are 34,000 people who
at least somewhat regularly go to Code for America Brigade Meetings. So brigades are ongoing volunteer groups. They meet up through meet up. They often meet weekly in
city halls around the country, and they're the kind of
volunteer fire department, but for technology essentially. And some of them very much
consider themselves, you know, I mean if you say how many
people are at Code for America, there's a whole bunch of
those people who would say, I'm in Code for America. - [John] Yep, that's
also common themes from what you hear with
Mozilla, too, which is like you turn a weakness is, you
can't hire enough people, to a strength, which is
you've become inclusive, bringing everyone. So you said staff about 45, fellows 24. That's a nice effect, too, right? Because you have people come in, and then they automatically leave. - Yeah. - And then they come back in,
and they automatically leave. So you have alumni too, right? An alumni network. - Yeah, that's right. And there's so much crossover now. We've got, we also have,
part of our extended family is we have six start ups
that have spun out of Code for America. Some of those start ups were founded by, one of them was founded
by one fellow and one former city partner. We've got former fellows that are now our, then go into city
government and then hire us. We've got fellows that have
come from city government, and it's just now everybody
sort of is in this square dance of being at some place in
the Code for America family. - [John] And I guess for
a little while you had an accelerator program, too,
where people from the outside would come in, and you'd
give them hugs, more or less. - Yes, yeah. - Could you talk about how that worked? - Well, sort of round about
the end of the first year of Code for America we
were realizing that some of these things that were coming
out were products that could sustain at least a small company. And we started thinking about, if we're gonna achieve our
vision, which is that government can work for the people, by the
people, in the 21st century, if everybody helps out, then
it's not just governments that need to act differently. It's not just the people
who need to act differently, as evidenced by the
34,000 people who do this, but the government ecosystem
has to really change as well. And so we knew that there
was a lot of entrepreneurial activity that, potential
entrepreneurs that were not understanding the value
of the government market. So we did two things really. We started incubating some
of the fellowship projects that could be sustained as companies, and those are the six spin outs. But we also hosted three
cycles of the first ever government market focused
accelerator program. And we thought nobody would apply. The first year I think we
had 194 companies apply to the accelerator, probably
because we were sort of walking around there going,
you know, I don't think anyone, does anyone understand
the size of the government technology market? - Oh yeah, how big do people
think, like do you know? Do you have a number? - I have a number, they're disputed. - How big do you people
think the government technology market is? Like what are some guesses? Two hundred billion? A trillion? Ten billion? - [Voiceover] Five hundred billion. - Five hundred billion. What do you think? - Your first guess was the
closest, just a little bit under two hundred, and
that's just domestically, and it's just technology, right? So the ability to use
technology to capture back cost savings in what we
call service delivery is not even captured there. If you wanna talk about size
of government, you know, we're spending about a
trillion dollars a year on government safety net
programs in the United States, and most of that is mediated
by very, very poor technology. - [John] (laughs) Right. OK, so OK. And then now you've expanded
your quote product lines from, you got the brigades,
you've got the fellows. What other products do
you have now in 2015? So what we're gonna do
is we're gonna talk about the products for a minute, and
then we're gonna talk about your customers for a second. We're gonna go back and
talk about why you did this, and then we're gonna talk
about how you operate for a little while. - OK. - So talk about what other
products you have now. - Well, what we're doing now
is focusing in four areas. And so we've got teams of
folks that work outside of the fellowship in health
and human services, in safety in justice, which is
basically community policing and reducing incarceration in
the criminal justice system. Economic development and
what we call communications and engagement. So there are products within
those that aren't necessarily part of the fellowship. We've been working on an
easier way for city governments to publish information to
their citizens and hear back from their citizens, for
instance, that are more like a company would have sort
of a real set of products that we sell to folks. - [John] Why those areas? Why HHS, why safety in justice? How did you find your way to those? - Well it was somewhat emergent
and then somewhat strategic. So we had been asked over
time to do projects in each of those areas in
the first four years of Code for America. - [John] Asked by your customers. - Asked by our customers,
so the way we used to do it basically was we would say,
it was very driven by mayors. So mayors wanted to have
Code for America come. Frankly there is a very nice PR bump. Our fellows wear these
blue American apparel track jackets, and they'll
have a press conference, and they'll show up in this city, and they look like techies, and the mayor gets to say like,
we're bringing tech talent to the city. And partly because of that we
would basically do the project that the mayor wanted us to do. I mean, in the case of
Discover BPS with the school selection thing, it was the
thing that was like driving the mayor crazy. I mean, he was getting angry
letters from constituents. But we found that as we were
doing stuff across the board, every area of government. And when we looked at where
we wanted to get deeper and have not just be doing
sort of demonstration projects that got people excited about
a new way of doing things, but actually building on that
work and being able to get real outcomes in each of those areas. We said, look, we're gonna have to focus. We've got, you know,
experiments already in those four areas, really
interesting experiments. We've got products in those
areas that we can build on and reuse, and it happens
to be that those four areas are things that, you used
the term the other day, hair on fire. Like there is hair on fire
in this country around community policing and over
reliance on incarceration. And if you can use data
and what we call service delivery service redesign that
basically making the systems that people use to interact
with government better to reduce those things, it's
something that's really, really important to government,
and it's really important to the people. And that's true also in
the delivery of services to people who need food stamps,
to the child welfare system, all of these things are really,
really, really huge problems for government, so we get
attention 'cause they're really big problems. - [John] Let's go back to
the beginning and talk about why you did it, about why. So why did you do this? Why did you start it? What, why did you do it? Did you have to do it? Did you say, this needs to
be done, and like nobody else is raising their hand? - That's basically it. I had been working on another
event that was a sister event to those Web 2.0
conferences called Gov 2.0. And through the research for that event, and we were trying to
sort of define the space. So Gov 2.0 back in
2007-2008 kind of meant get government agencies on social media. And Tim O'Reilly, who I was
working with at the time, had a notion that it meant
something much more powerful. It meant like all of the
disruptive nature of Web 2.0 applied to government, and the
ability to really think about government as a platform
for more generative work in the same way that at that
time the iPhone was coming out, and it was going, oh,
platforms make it possible to do all these amazing things. So in the course of that
work we actually went to D.C. quite a bit, and were talking
to people about how they did technology for government. And it was just so shockingly bad. So much money gets spent on these things. Things are built from the
ground up that just should be reusable components, that it
was very eye-opening in a way that like only the ultimate
nerd would be sort of outraged about it, but I was
pretty outraged about it. And the idea to do a service
year program was actually came from a friend of mine
who worked in local government and who wanted me to bring
people from the Web 2.0 world to Tucson, Arizona
to do apps for Tucson. And I kept saying it wouldn't happen, but the service year program
seemed like a really good idea. And when we thought of it,
I did spend a fair amount of time thinking who should do this, and realizing that no one else
was gonna do it, and that... I'm not technical, actually. I've written HTML, but
I'm not really a coder, but that I had spent a long
time building a network of people who would probably
be interested in this, and that that network would
be something that I could bring to bear on trying
to solve the problem. - Yeah, a couple of concepts
that Michael Dearing and Anne Miraco talked about
is that they like start ups to have authentic (mumbling). The start up's inside them,
and they can't keep it in. That feels a little bit
like what happened here. And then it's interesting
'cause your background until the Gov 2.0 stuff wouldn't
obviously say to me that you were gonna go try to fix
all of government with nerds. - I also wasn't entrepreneurial. I mean, I worked for
companies until this time. It was completely like
throw caution to the wind, quit your job, have no income. It was totally insane. - [John] And the framing was broad. It was just let's try to
make government less stupid by applying nerds, more or less. - Yeah, that was it, that was all we had. - It's not such a bad strategy. OK, so Chris? - [Chris] Yeah, why didn't
you do it as a for profit? - [John] The question is, why
didn't you do it as a for profit? - Yeah, and it's a good
question, and people ask us all the time why don't we
become a for profit now? The answer at the time was
that we were asking people to come do a year of service. So if you're gonna work for
market rate I just don't think people would have done it if
they felt that they were just contributing to someone else's profit. The answer is different today,
but it's not that different. There may be a some point
pieces of this that, I mean, we've already
spun out six companies, so it's not at all that we
believe that non-profits will fix this. I think it's we wanna play
a certain role in creating an ecosystem them will
be largely driven by for profit companies. But right now I think it's
still the right thing for us, and I think we're able to
attract people who wanna be part of it. - Yeah, and this is what,
two comments on that. So one is similar to what
Mitchell and Brendan did with Mozilla, which is they said, look, we can't compete symmetrically here. We have to compete in a new
way, and we're gonna do this thing in a open source way
that involves community. But as the quid pro quo
we're gonna commit to this, that we're not gonna become
super rich from this. We're gonna pour all this
back, so it was the covenant between Mitchell and Brendan
and the community at some level that they wouldn't profit
extraordinarily from it. So I think that's part of it. You have to figure out
what your advantages are, and how to play them,
and what makes that work. Non-profit status is part of that. The second thing I should have
mentioned a few minutes ago, which is I think you're
gonna find themes throughout the course, Eric Schmidt
for sure, me and Lou Reed and Jen, that we're all
pretty flexible about how we think about organizations. So I think that like the
model of the organization's changing quickly now, which
is there are spin outs, there are affiliate organizations,
there are arms length affiliations, there are groups
of people who just hang out and do stuff together. Like the brigade is just
some people who do stuff, but they feel affinity. And so I think that our
traditional notions of association and work are changing quite quickly now. I think you're gonna see that everywhere. So I think that the most
agile organizational thinkers will be the most innovative over time. - [Voiceover] Where did the
early funding come from, and (mumbling). - The question is where did
the early funding come from, and how long did it take to raise, yeah? - I got three $10,000 checks
that helped us get started, one from a friend on my birthday. That was nice. Head of a foundation. She bought me a wallet,
actually, for my birthday, and she said, "Open the wallet." That was nice. One from Jean Case of Case Foundation. - [John] Steve Case. - Steve and Jean Case, yeah. And one from the Sunlight Foundation, who also were our fiscal
sponsors for the beginning. And that was enough just
to like pay legal fees and be able to pay for FedEx's to
city governments and stuff. - Happy birthday, here's $10,000 dollars. Now go work for a hundred hours a week for the rest of your life.
- Exactly. And then I think that
actually Stacy at Omidyar, Stacy Donohue at the Omidyar
Network is the first foundation to give us a check that
enabled us to pay salaries, which was really good
'cause I was losing my mind. Yeah. - In back. - [Voiceover] You said initially you had the governments apply? - [Jennifer] Um hmm. - [Voiceover] Like how does that work? How did you first of all get (mumbling). - If you're a fake organization? The question is how did you
get cities and governments to apply to your fake
organization for this fake prize? But it actually was a real prize. - Yeah, it was a real prize,
and they actually got a lot, I mean, they were mostly
really happy with it. Seattle not so much. You know, it's actually
a pretty common thing, and I don't know that I realized
that we were stumbling on something that was quite
common in government. A lot of like federal grants,
for instance, are administered by like here's the challenge, apply. Tell us why you're the right
folks to get this assistance or whatever, so we were kind
of accidentally just played right into a thing that they
already like knew to do. I mean, again, I think
the issue of us not being a real thing at the time
was just looking good. You know, having a decent
website just makes you look legitimate, or at
least it did in 2010. But you know, you make 'em sorta I say dance for grandma a little bit, and somehow they take you seriously. - [John] There's a lot
of stories like this, like many, many start ups, and
they had enterprise customers come in, they'll go rent
employees, or you know, rent space or borrow space from people. I think mostly it's about
how do you communicate. It's not lying exactly. You're trying to figure
out how to get them enough confidence that you're gonna deliver. You really have to believe
that you can deliver. That's pretty key. - And the folks that
were actually applying, like Nigel and Chris in
Boston, absolutely knew that we were making this up. It's the other people
in City Hall that they had to convince, and that's
where the website came in. - [John] Alright, there in the back. - [Voiceover] (mumbling) - How difficult was it to
close the first big deal? - [Voiceover] What was that like? - Big is kinda probably in
quotes because the first big deal was probably $20,000 dollars
from a city or something. - No, we got $250,000 out
of some of them, yeah. How difficult was it? It was difficult, but another sort of mechanism that we had,
or sort of a tactic was we made it a contest, so
if you won you were able to put out a press release
saying Austin's been selected. And we had a deadline, so
we had fellows starting on January 1st, and we said
you're not in if you don't get the contract signed to
us by whatever it was, probably September 1st. And I've learned this since in government, it's funny that you
actually sometimes relearn the mistakes you've made
before, but we were working with another city government, whom
I love but will not name here, on a contract that didn't have a deadline, and it took forever. So having forcing functions
is just incredibly valuable, especially in a government context. - Yeah, forcing functions
are important I think in all sales cycles. I think some sense of
scarcity, some sense of FOMO is a word we talk about a lot. - Yeah. - Fear of missing out,
that if they don't sign up, like some other stupid
city is gonna get it. Fresno's gonna get the
space or whatever it is. So let's keep moving a little
bit because things were going pretty well. So you play it forward to 2013. It's the third year of the fellowship. Things were going pretty well. We had 12 cities or something like that the third year?
- Yeah. - And so then Jen calls me in
April or something like that and says, "OK, great,
things are going great. "I'm thinking about leaving for a year." So can you talk about, and
to go work in Washington D.C. And so can you talk about what
it felt like to contemplate leaving this thing which
you clearly found product market fit, cities wanted
it, there were pulling on it, and you said, well, I
wanna do this other thing for a little bit, but I'll
totally be right back in a year. Can you talk about what that was like? - Yes. - And why you did it? - So I called you in yeah, March or April, but the whole thing had
started back in November. And Todd Park, who is the
CTO in the White House, really had created a program
modeled after Code for America called the Presidential
Innovation Fellows Program, which I would recommend
to all of you as well as the Code for America program,
if you're interested, and wanted me to come help him run it. And it was sort of his play at innovation, shaking things up in the White House. And so the election was just
over, and he was waiting to talk to me about this
until they knew we had a job for the next four years. And initally I was, I mean
it was a good four weeks after he asked me before I
even began to consider it. He said, you know, needed to come. I was like I can't move to D.C. I have a start up. I mean, I had like all these
completely solid reasons why that was never going to happen, but I really, really, really
liked Todd, and I really, really loved the idea
that they were doing this. So I kept saying let me... I'll help you find the right
person, and furthermore, I really think you should think beyond the Presidential Innovation Fellows Program, and you should think about
creating a core service within federal government
that does good technology. And that was because when he
called me I happened to be visiting the Government
Digital Service in the U.K., which had done that,
and done it very well. And I had my mind blowed by
that, so I used the opportunity of being recruited to just
pitch him back on something, but had no intention of doing this, and certainly not of moving to D.C. I mean, I have a now 12-year-old daughter, who was not going to move to D.C. So this was not just the start up. There are things even
bigger than the start up. - [John] Todd's a pretty
good recruiter, it turns out. Todd's one of the best
recruiters in the world. - He's the best recruiter in the world. That night Nick Sinai, who
was the Deputy CTO there at the time, who was
with me in London said, "I heard you got a call from Todd." And I said, well you know,
I'm not gonna do this. And he looked at me and he
said, "Have you ever heard "the term water on stone?" And I was like, "What?" And he's like, "We will get you. "It's like water on stone." And that's what it was like. It was just, he doesn't let it go. - Yeah, you can't give Todd any daylight. If you give him any
daylight at all, he's like, oh interesting. You don't wanna do it
because of X, got it. And then he moves X, and
it changes, and it's like, "Maybe if I have the President call you. "Let me see, let me see,
I'll be right back." And he's a very, very
persistent recruiter, and that's what he does
for the President now in this program that Jen started. - No, I would say in
some ways I felt like I didn't have a choice, but
that's not obviously fair. I did have a choice, but he
was extremely persistent. - Yeah, Todd started a
company called Athena Health, which is a big healthcare company. And then he started another
company called Caslight. And then he decided there
were bigger problems to solve, bigger fish to fry, so
he went to the government and worked as the CTO for
Health and Human Services before becoming the CTO
for the White House. - And I think in those two
jobs he got enough exposure to why the healthcare system is as it is, that you have to follow that train back to federal government. In the end I said yes because
he went and created the opportunity to create
an American Government Digital Service, and said,
"I did the thing you asked. "Now you have to come",
which felt like, OK, I didn't actually say that, but that felt, yes, it was
sort of manipulative, but it worked out well. And because I was running
this organization, and I was dealing with people
who work in government who have very specific challenges
that are hard to understand if you don't work in
government, and I had heard that over and over again, not
just from our partners, but actually from the fellows. It was the fellows coming back and saying, "No, no, no, no, no,
you don't understand." It's really, really, really
different the way decisions get made, the incentives, the culture is very hard to understand
if you're not part of it, and I did feel in the end
like if I was going to spend the rest of my life
asking people to come work with government, and I wasn't
willing to do it myself, A, did I really have the
courage of my convictions? And B, would I really
ever understand our client well enough? - Yeah, I think my observation
is they gave you an insanely powerful tool to use now. It's like, oh I've been there,
I've been in your shoes, I understand how it feels. So tell me about what
it, so that was a year, and you stepped away, and
you put a leadership team in place to bridge, and
really you burned the candle at both ends realistically. You were also on Board calls
even while you were away. So but tell me what it
was like to come back. So you started this organization. You let somebody else
run it for a little bit, although you were around. And then you came back, and
how did your lens change? How did you look at it
differently than before? And what did you not see
right away that had changed? Just talk about coming back a little bit. - Well, I think it was a
profoundly strange experience in some ways. I had been through not just
a year in the White House, but a year of healthcare.gov
crashing and burning, being around that, trying
to start the United States Digital Service, many
times sort of wanting to stick a pencil in my eye. I actually felt pretty
demotivated at the end. Not demotivated, I felt when I
left that I hadn't succeeded, and I was just, I look back it now, and
it was a wonderful year, and so much good stuff has happened, but it's really hard when
you're in the middle of it. So I was sort of coming back
to Code for America saying breath of fresh air. Now I'm in an environment
where I'm the CEO, I'm in control, and it's
gonna be so different from the incredibly difficult year
I had trying to convince a bunch of people to do this
thing that was obviously the right thing to do for
me, you know, in my view, but not obviously the right
thing to do in their view. And found that, in fact, the organization had changed a lot. It took me a longer time to
diagnose that than I thought. It looked the same, and people
were saying the same words, and so I very much wish I
could take back those first three months and do those
first three months differently upon return. And what I really needed to do, first thing I did when I
came back is I met with every staff member, and like
half the staff at that point was new 'cause we'd grown
so much while I was gone, and got to know them a
little, but I didn't ask the right questions. And I didn't understand that
this focus area strategy I talked about earlier that
we thought we were doing wasn't happening. And so it was a process of
really seeing past the surface and grappling with a
lot of good intentions, but a different direction that
I'd really wanted it to go, and sort of trying to steer
even a non-profit in a way that took some time, and
it was a wake up call to me I think about how organizations work. - Chris? - [Chris] What would the
right questions have been? - Chris always asks those questions. - That is a really good question. I mean, very specifically at
the time there were just things I missed about, when I
came back we were starting our recruitment cycle for
our next set of partners. And I didn't ask specific
enough questions about the sales offering, about how
we were framing it because I assumed that things
that we were talking about at the Board had sort
of translated down from the Executive Director to the staff, and I just should have
sort of from the ground up looked at more material
specifically, asked you know, OK, will we be, how are we working
in these focus areas now? And I did ask a number of questions. I mean, it was an awkward
time because I came back to a couple of programs that
had been started that were really, really had great
intentions, but they were what Eric Reese calls zombie
projects, like they didn't have any clear metrics, so
you couldn't kill 'em, or there was no reason for
them to live or die because you didn't know why they
existed in the first place, other than sort of people
felt good about them, so we shut down a couple of programs. So there was already sort
of heightened sensitivity to Jen's back, and things are different. - [John] Jen's back, and
she's killing projects. - And I'm killing projects. And so then, you know, I
kind of wanted to demonstrate that I trusted the team, and
I wasn't a micro manager. And I think there was a way
not to be a micro manager, but still understand that
decisions that were made in the first few months
when I was already back were on track, and if I had
dug in we would have been able to steer those because
the way Code for America works is it takes you nine months
before you start working with the government partner
to get that all set up. So you live with the
consequences of your decisions for a really long time. - So the staff about doubled, right, between when you were first,
then when you came back. - Yep. - So tell me about how you
think about what it takes to manage differently now,
now the staff's 50 people instead of 25 people, give
or take, and so as a CEO, how has your job changed? As that's grown, how do
you manage differently now? - Well, as you know, I don't
feel like we're entirely in a complement there. - I have to say this is
the part where Jen's gonna say something self-deprecating,
but she's doing fine. - Well thank you, but we
definitely need more operational strength in terms of managing. - What does that mean,
operational strength? - The day-to-day. You know, I'm here doing
this talk instead of... I mean, very literally at
this moment, and due to get OKR's back to our focus
area leads so that they can ship them to our functional area leads, and I'm here instead. (laughter and applause) - This will be a theme, too. People with day jobs don't
have enough time to teach, but thank you. - So you know, there's a
blocking and tackling of one on one's with your
key staff, you know, are we clear on our goals
that just sometimes get missed because I'm on the road,
I'm doing fund raising, I'm doing public speaking and
doing business development. And you know, we're hiring
for Chief Operating Officer so that there's someone
who's more consistently there to do those things and make
sure we're still on track, and like really read
through all the OKR's, and really edit them,
really make sure we're on the same page. But I did have a benefit
early on at having I feel like and I hope everybody has this experience, and I've had it previous
in my career, too, where the words you say
mean the same things to the people you're talking to. It's just such a beautiful feeling. And I mean I had one staff
early on, very young guy, just right out of college he came in. I think he could just read my mind, and there was very little
sort of friction in terms of translation from what I intended
and then what was executed. And when you have it team like that, it feels really, really great. And you just can't count
on that continuing. And so the thing I do just sort
of daily right now is try to remind myself, I know I said those words, but that's probably not
what that person heard. Let's double back and check. And where I fall down is if
I don't go back and check because I'm on a plane
elsewhere, and you know, it's gotta move forward,
and you're not going, OK, let's match between what
we thought we said we were gonna do and what we're
actually doing, you know. There's a great quote, the
problem with communication is the illusion that it has occurred. - That's a beautiful quote,
that's exactly right. So that's one of my favorite topics. I think of the role as a
CEO, so I think that as you get bigger CEO's start to
talk about more and more words like communication and alignment. - Oh my god, I hate
the term alignment now. - Yeah, I love the word. Alignment is almost all I
think about as you get bigger because what you're trying to do is get- - It's on our list of words
you're not supposed to say at the office anymore, and
I still say it all the time. - That's good to know. Noted. The reason you talk about
is what you're trying to do is create an organization
where the organization will roughly make the same
set of decisions whether you're in the room or not. So how do you do that? And especially if you're
growing, if you're doubling every year, if you're
growing more and more people? People will come in who don't
have a ton of time with Jen. The things she said, I mean,
it's an amazing experience to have somebody you work
with that your don't have to talk in complete sentences to. Like there's somebody that I
worked with for better part of 25 years, guy named Mike
Hanson who's at Stanford CS, he and I can finish
each other's sentences, and he's the smartest technical
person I've ever known. I can say three words,
and he codes a prototype, and like boom. And like that's an unbelievably
powerful relationship. And the net is if you ever
find a relationship like that, do not let it go. Invest, invest, invest. But as CEO what's happening
is you're bringing more and more people in, and you'll say words, and like Jen says, they're
perfectly clear to you. They're probably perfectly
clear to some percentage of the staff, and they are
probably like Greek to some of the rest of the staff. I remember in my start up,
Reactivity, we hired a CEO who was an HP guy, and
he's an amazing guy named Glen Osaka, and he used to
get so pissed off at me. Like the first year he was
CEO, and I had stepped down from being CEO, he had
almost fired me literally four times that year because
I was such an asshole. But we got into such fights. I was the founder, I was a kid, whatever. We'd get in these meetings
where he would say, well, in the near term we need
to do this and that and this. And I'm like, in the
near term we have to do like this and that. And what we realized is
like when he said near term, he meant like 18 months,
and I meant like Tuesday. (laughter) And we're like, holy
shit, we're talking about different things, even though
we're using the same words. And so then he gave me
like safe words, right? So then goes OK, OK, when we do this, instead of us fighting in
front of all the children, just say, let's take this offline. And we'll go talk about
it, and then once he figure that out, once we
figured that out together, we started issues when we
had vocabulary differences as opposed to intent differences, and that was a breakthrough. But it takes a lot of
fucking work when you've got a team that's growing, and
a team that's trying to do all their own work, and
you've gotta keep track of who knows what, and so as
a result, like at Mozilla, like as a result as we
got bigger and bigger I started... Oh, there's a second
problem, which is every day, especially in any learning organization, the CEO learns stuff. - Yep. - So you learn stuff today
that makes your thought process a little different than yesterday
in kind of a nuanced way. And then the next day you
learn a little bit more, and the next day you
learn a little bit more. And so if you're not
careful, your message changes and kind of moves over time,
and you don't get enough time with everybody for
them to notice this. - Right. - And so at Mozilla I had
to keep simple messages, a small number of messages,
and I had to repeat it over and over and over and over and over. I had to keep it the same very meeting until it was time to
change, and when I changed, make a really big change. Say OK, I'm about to change
this thing I've said for a year, but it's a hard communication problem. - Yeah, I suffer from that. - Sure. - [Voiceover] You said you
have this kind (mumbling). Did you have those guys
when you were (mumbling). - So the question is, were
the folks that I relied on- - [Voiceover] Yeah, were
they there when you went to look for government, or did
you have to hire those people? How did that work? - We didn't hire new
leadership when I left, but that was partly because
I made this promise to be right back. Let me just go set up a
service in the White House, and I'll be right back. - [John] One thing that
happens when you hire temporary leadership or
interim leadership is they tend to fill up the
slot that the person was in, and so it tends to not
exist when they come back. - Yeah, so yes, one of the
folks that I was referring to could sort of read my mind
was there when I left, yeah. - [John] Other questions for
Jen, and I'll keep asking questions, too, but do you have questions? - [Voiceover] I'm from the south. What's your strategy to
penetrate like a much more culturally less accepting of
I guess, change in general? What's your (mumbling) areas like that? - What are you saying about the south? - [Voiceover] I'm just
saying 'cause you studied Boston, Seattle, D.C., like
that's different than like rural Alabama, like all
of underserved communities that might not be so open to (mumbling). So what's your strategy dealing with that? - So the question is,
what's the CFA strategy for dealing with cities that
might not be as progressive as some of the early adopters? Which actually kind of a
common customer question, too. - Yes, it's a really good question. Before I answer it I'll
say, I mean, we've worked in Atlanta, we've worked
in Louisville, Kentucky. We've worked in, I mean,
we've worked in some smaller places. We haven't worked a ton
in the south, I'll agree, but we're also often in places
that are not particularly tech friendly. There's like one great
champion for change who's really into us, and then a
lot of folks that might be more of the culture that
you're referring to. So we've done some of
it, but I mean basically the strategy is you've
gotta just make this normal. So the more that they
can look at other cities, bigger cities, cities their
size, you know, cities that are like them in some way and say,
they're doing it over there, and the sky didn't fall
'cause they did it there, then you'll be able to get
into those places that are less friendly or less sort
of jumping on the bandwagon. I think that happens, I
mean, stuff that just seems absolutely nut balls, you
know, 10 years later it's like well, that's just how you do it. And that's what we're trying to do. We're trying to make
this just how you do it. And there was a great line
from our Code for America summit which was last week this
year where someone tweeted, every year this seems less and
less like this crazy thing, and more like, of course. - Yeah, part of start
ups is just showing up and just doing it, and then
people just get used to it. So I have one more question,
and then if we have time for a couple more questions. So how do you tell whether
you're winning or not? 'Cause it's not a financial
metric, it's not a user metric. How do you tell? - So I'll give you the official
answer, and then sort of non-official answer. We'd set really clear metrics. There's the one metric to
roll them all philosophy, and we have not done that. We have a number of metrics. We do a lot of different
projects, and so we have to have a way of looking across them. You know, we didn't go
with like users because some of these projects
have like three users, but they have very high impact. So we've gone for the
number of our projects that are sustained. We've gone for the number
of people who've decided to stay in government. We've gone for a whole set of
things, actually redeployments because users isn't a great
metric for us in all cases. When things are redeployed,
we consider them to be successful. Why would someone redeploy them
if they weren't successful? So redeployments are a big metric for us. So you know, we kind of live
and die by this dashboard where we tally up these wins
in these different categories, and we report that back to our funders. I know we're winning when
stuff happens as like happened last week where we're in the
right place in the right time to be able to tell somebody that, I won't name what it is,
but I will say it's the State of California is
about to put out an RFP for a computer system that
will run a social service. And they think that the system
will cost a billion dollars, and it'll take five years. And in the meantime of
building, I mean things that take five years and cost a
billion dollars have about a two percent success rate,
especially in government. So in that meantime, the
service that people and children rely on will continue
as is, which is terrible and it doesn't work very well. And we know there's a better approach. We know you can do this
modularly, iteratively. We know you can start
small and build and have users involved all the way. This is being done now
because of healthcare, because of the United
States Digital Service, because there's now all these examples, you know that you can
do this in government. And you know, I think we
were in a position to say there's a lot of political
reasons why this may move forward, but I think
we might be able to stop it. And when people are really
listening to you because you've done something else
for them that worked out really well, and they're
going, OK, maybe there's a way to make this come out
differently, and like that's a big win, then I think we're winning. Those things make me really happy. - [John] Sort of remind me
of the question a student asked I think, which is
project forward 10 or 15 years, I'm not sure what the timeframe is. What's the happy case? Like paint the picture of what happens, or what government's like and what the interactions are like. - Well ultimately I think,
yeah so the interactions, you would not have this
completely like screech experience if you're having an interaction,
either digital or otherwise in government that makes
it so different from a consumer transaction
that you would have. - [John] Getting a
(mumbling) or something? - Yeah, and you know,
maybe a little different in some ways, but right
now there's this huge gulf, and so we need to span that gulf. There's a ot of ways that
you might express that. I think also it's not just
for us in this room, right? I care deeply about government
working in ways that if you've not been on food stamps, or you've not had your kid go
into the child welfare system, or actually if you've not
tried to do a start up without a lawyer, you
won't see that dysfunction. And so it's for everybody, right? But I think that the real
way we'll know is that people will have greater trust
and faith in government, and will support it. That's the problem. It's very hard for people
to support government and wanna invest in social safety net, trust government when
the experience we have kind of makes them look
imcompetent and frustrates us. So the ultimate goal
really is that we believe in government, we invest
in government and we use it in ways that are
consistent with our values. So if we're gonna spend
a billion dollars on any given service, it
oughta be in helping people, not in the computer system, right? We spend $70 billion dollars
a year on food assistance in this country. Let's actually have that be
in money that we give people to buy food, right? - [John] Alright, any
other questions for Jen? Sure. - [Voiceover] So a lot of
people in tech like to stay away from government because of
like bureaucratic red tape. What were some of the
biggest hurdles that you guys encountered with all the
different local governments? And how'd you get past them? - [John] So the question
is how do start ups, a lot of people in tech wanna
stay away from government. Seems hard, lotta red tape. What are some hurdles, and
how did you get around them? - There are hurdles for
sure, and the biggest one is procurement. I could talk all night
long about procurement. Let's not do that. - [John] It's more interesting
than it sounds like. - It is interesting,
it's very interesting. It's a system that accrued
over time to reduce risk which has accidentally
created enormous risk, right? It's created high, very
high failure rates. - [Voiceover] Can you
explain what you mean? I'm not sure what you mean. - So the system of
procurement is designed, there's multiple levels of
procurement systems basically, and it's a way that the
government buys things. We unfortunately buy software the same way we buy pencils or trucks,
which is we spec things out to a very, very high degree,
and then we take them out to bid, and there are so
many steps in the process. All of those steps in the
process are designed to reduce fraud and waste
because there's very low levels of trust. You don't want your taxpayer
dollars being spent because I decided to give a contract
to my cousin, right? So there's all these layers
and procedures that are built in to keep that from happening. Unfortunately that means
if you're a start up, you might experience nine
months of trying to go through a procurement process,
or trying to get a bid. The amount of paperwork is intense. The number of certifications
you have to make. You have to do things like
certify that your bathrooms are labeled properly, like
you have appropriate labeling on your bathrooms. I mean, there's all of
these sorts of crazy things. - But in nine months
technology may have changed. - Right. - And projects get long, people roll off. So like when you get long
projects, people roll off. That increases the risk
of the thing failing. That's why big projects
just don't convert. We know this. - Well the nine months would
just be to get the contract, and then they want you to
build the thing for five years because they're giving you
specs for seven modules of a system that's (mumbling), and you really wanna start
small and build with users. But I will say is that it's changing. We've got a number of
start ups that came through our accelerator, and we
had one in particular had a great rule. Anyone who complains about
procurement is fired. Just deal with it. And actually that's worked
really well for them. But you see procurement hacks. I mean, a lot of people kind
of originally kind of called Code for America for a
procurement hack because we did go through a procurement,
but instead of procuring software we were procuring
people who could do what we needed, what was needed
in a short period of time. There are a lot of things
that are happening to change procurement that
are too nerdy for us to go into right now. To be slightly over the top
on nerdiness, for instance, federal government's now issued something called an Agile BPA. It's a blanket purchase
agreement for Agile software development. As this starts to propagate
throughout government, and in fact we'll probably
use it with the state that we're working on to try
to get them off of this billion dollar plan which won't work, the problem is the tools at hand. So you have a ton of people
in government who know the right thing to do, and
they look around and they go, none of the tools that I
have will get me there. Oh well, let's just do what we did before. And when you give them tools,
like here's this thing. It's called the Agile BPA. It's legal, you can use it, do, then they can do that. And that's what's changing. It changes maybe not as fast as I'd like, but it's changing, and there
is now a fund exclusively for government technology start ups, and the guy who runs
it, this amazing named Ron Bouganim, who started
it because he worked with us for a couple of years as a
volunteer and saw the dynamics of the market and saw
that they were changing says that the average sales
cycle in his portfolio right now is 86 days, which if you told somebody
that a year ago in government technology,
they would have said you are crazy. But it's because people are
finding they have new tools, and they're finding new
ways of doing things. - Great, so we should wrap
up 'cause Jen's gotta get to the city, as do I, so thank you very much.
- Thank you, this was fun. (applause)
