►
From YouTube: What does it take to transform a federal agency to a product-focused enterprise? -GitHub Universe 19
Description
Presented by Kelly O'Connor, Principal, Boston Consulting Group, and former Product Manager for the US Digital Service
By shifting the focus away from traditional, plan-driven “project management” (e.g., cost management, schedule management, etc.) towards a more user-driven product management approach (e.g., user research, minimum viable products, etc.), the VA has been working to create the best digital experience for Veterans and business customers. Learn how VA went about this massive cultural shift -- to making human centered design (HCD) a requirement, not a “nice to have.”
There’s no point in delivering something on-time and on-budget--if it’s the wrong thing.
About GitHub Universe:
GitHub Universe is a two-day conference dedicated to the creativity and curiosity of the largest software community in the world. Sessions cover topics from team culture to open source software across industries and technologies.
For more information on GitHub Universe, check the website:
https://githubuniverse.com
A
So
the
US
Department
of
Veterans
Affairs
is
a
very
large
organization,
with
a
budget
request
of
more
than
220
billion
dollars
and
three
hundred
and
sixty
six
thousand
employees
VA
is
the
size
of
a
fortune.
Seven
or
eight
company
VA
provides
health
care
services
and
benefits
to
9
million
veterans
and
their
families.
This
includes
things
like
education
benefits,
home
loans
and
burial
benefits.
A
And
in
2015
this
was
Veterans
digital
experience
with
VA.
So
if
you
wanted
to
compare
schools
to
use
your
GI
Bill,
you
had
to
download
raw
data
on
a
spreadsheet
from
va.gov.
If
you
wanted
to
find
a
VA
facility,
you
had
to
use
this
outdated
map
to
bring
up
free
text
of
locations
for
VA
facilities,
and
if
you
were
a
blind,
veteran
or
sight
impaired,
you
couldn't
even
use
this
tool.
A
The
primary
way
veterans
interact
with
VA
digitally
is
through
va.gov.
In
2015,
there
were
250,000,
mostly
static
pages
of
content,
not
written
in
plain
language.
46
percent
of
those
pages
were
mobile,
responsive
and
just
10%
were
transactional.
So
if
you
actually
did
get
to
a
page
where
you
could
do
something,
some
of
the
highest
demand
veteran
features
had
availability
of
76
percent
or
less
on
the
backend.
The
release
process
at
VA
was
90
days
or
more
generally,
and
there
were
multiple
logins,
with
varying
levels
of
security
to
protect
veteran
data.
A
So
today
we
think
there's
a
big
improvement
in,
but
in
veteran
experience
interacting
with
VA,
we
still
have
a
long
way
to
go,
but
you
can
see
VA
gov
is
a
modern
interface.
It's
hosted
in
the
cloud
with
much
much
better
technical
performance,
its
transaction
based,
so
veterans
can
come
and
do
something
they
can
apply
for
something
they
can
check
their
status.
They
can
schedule
something
you
can
see
on
the
right.
We've
launched
more
than
40
products
over
the
past
four
years,
so
in
government
that's
a
lot.
This
is
this
top
product.
A
Here
is
the
VA
facility
locator.
So
that's
that
old
map.
So
now
you
have
a
modern
online
map,
not
only
with
directions
to
VA
facilities,
but
veterans
can
also
see
what
services
are
offered
at
different
VA
medical
centers,
which
is
pretty
cool,
and
this
is
our
GI
Bill
comparison
tool.
So
I'll
talk
about
this.
A
little
later
we
were
also
able
to
reduce,
on
the
back
end,
release
cycles
to
anywhere
it
was
more
than
90
days.
A
Now
it's
between
one
and
30
days,
so
some
teams
at
VA
are
shipping
code
every
day
and
we
also
launched
a
login
that
meets
the
highest
level
of
NIST
security
requirements.
So
it's
a
level
of
assurance
3,
which
is
great
to
protect
VA
data.
Oh
and
the
nice
thing
is
on
the
facility,
locator,
blind
and
sight.
Impaired
veterans
can
use
this
tool.
A
A
Here's
a
little
bit
about
my
professional
background
and
some
things:
I
care
about
so
I'm,
very
passionate
about
veteran
issues.
My
dad's,
a
Vietnam
combat
veteran
82nd,
airborne
I,
try
to
be
a
patient
advocate
in
the
opioid
crisis
and
I'm
very
excited
to
spread
the
word
and
help
large
organizations
shift
from
a
project
management
focus
to
a
product
management
focus,
especially
in
government.
A
A
That's
me
in
the
middle
and
we
are
still
going
strong
in
this
administration.
This
is
the
best
professional
experience
I've
had
in
my
career.
So
if
you
are
interested
in
working
on
an
issue,
you
care
about,
the
government
has
every
issue
possible,
come
and
work
at
the
digital
service
and
have
an
immediate
impact.
A
So
today,
I
want
to
talk
about
six
plays
that
we
ran
at
VA
I'm.
Also
going
to
tell
you
about
tons
of
mistakes
we
made,
so
you
can
you.
Hopefully
you
can
avoid
those
in
your
organization's.
The
one
caveat
is
as
I
go
through
the
success
we've
had
at
VA.
Please
keep
in
mind
that
our
digital
service
team
flexed
from
about
25
to
30
people
at
any
point
over
the
past
four
years.
So
everything
we
did.
A
So
the
first
play
is
think
big,
but
ships
super
small
and
fast.
So
this
is
actually
a
screenshot
from
our
digital
service
strategy
in
2014,
so
marina,
Martin
and
three
other
digital
service
people.
We
literally
worked
out
of
a
large
closet
at
VA
headquarters
on
Vermont
Avenue,
but
we
developed
this
document
called
the
21st
century
VA
and
it
talked
about
how
to
transform
this
huge
organization
to
improve
veteran
experience,
benefits,
delivery
and
back-end
operations.
When
it
comes
to
digital.
We
took
this
document
to
every
meeting.
A
We
went
to
every
elevator
ride,
we
handed
it
out
and
one
day
Sloan
Gibson,
the
deputy
secretary
of
VA
read
the
document
he
came
to
Marina
and
he
said
I
want
that.
Va
and
I
tell
that
story,
because
it's
a
pivotal
tipping
point
when
our
small
team
became
a
small,
empowered
team
and
that
made
all
the
difference.
A
So
once
we
had
some
power,
which
was
pretty
fun
because
you
can
do
stuff,
it
only
takes
a
few
change
agents
to
start
building
momentum.
So
these
are
just
a
few
people
who
really
helped
us
digitally
transform.
Va
we've
got
drew
here
today,
who
leads
our
awesome.
Api
program
he's
also
our
chief
foodie
at
VA,
but
I
want
to
talk
about
Curt,
coy
so
Curt's
over
in
the
corner,
and
he
was
actually
the
business
sponsor
of
that
GI
Bill
spreadsheet
and
he
knew
in
2015.
This
is
a
horrible
experience
for
for
veteran
students.
A
I
want
to
fix
this,
so
he
goes
out
to
the
private
secretary.
You
know
puts
a
propose.
He
gets
proposals
in
and
the
most
competitive
proposal
he
receives
is
for
a
digital
product
for
14
million
dollars
and
it
would
take
a
year
to
launch
so
what
Curt
decided
to
do
is
go
to
marina
and
her
team
of
two
engineers
and
a
designer,
and
he
has
testified
to
this
to
Congress.
He
tells
he
has
told
Congress
Marina
and
her
team.
A
A
So
now
we
have
a
small,
empowered
team.
We
have
some
power
and
authority.
We
have
some
champions.
We've
identified
some
problems,
we
know
we
want
to
solve,
but
I
don't
know
if
people
out
there
who
work
in
large
organizations
if
this
resonates
with
you,
but
we
had
a
decision
to
make.
Do
we
try
to
fix
the
big
existing
thing
and
chip
away
at
that,
or
do
we
just
stand
up
a
brand
new
small
thing
on
the
side
that
lets
us
work?
A
The
way
we
want
to
work
and
there
are
pros
and
cons
to
both
of
those
just
as
decisions
I,
think
we
made
the
right
decision.
We
decided
to
stand
up
a
modern
tech
stack
in
the
cloud,
so
we
could
work
the
way
we
want
it
to
work,
but
think
about
that
veteran
experience
right.
We're
sending
veterans
to
a
brand
new
website
called
vets
gov.
It
doesn't
look
like
a
government
website
right.
A
A
Since
that
time
we've
launched
about
45,
modern
digital
products.
In
2016,
we
had
a
pretty
big
milestone
where
we
launched
a
new
logon
that
was
more
secure.
That
was
really
really
hard
in
terms
of
change
management,
and
then
you
can
see
the
evolution
and
what
was
exciting
this
past
Veterans
Day.
Is
we
folded
in
the
big
thing
to
our
small
thing
on
the
side,
so
we
kind
of
swallowed,
VA
gov,
and
you
can
see
the
interface
here.
It's
a
much
more
modern,
better
technical
performing
site
for
veterans.
A
So
it's
a
fun
story
to
tell
these
are
just
two
of
the
40
products.
We've
launched
I'm
going
to
talk
a
little
bit
more
about
that
later
in
the
presentation,
but
that
was
the
first
place,
so
we've
got
a
vision.
We've
got
some
champions
and
we've
got
like
a
way
we
can
deliver.
We've
got
a
modern
tech
platform
where
we
can
start
shipping
stuff.
A
So
the
very
first
thing
we
had
to
do
at
VA
and
I.
Don't
think
this
is
unique
to
VA
I.
Think
many
government
agencies
conflate
these
two
disciplines,
so
product
management
is
not
the
same.
Everybody
in
this
room
probably
knows
this,
but
product
management
is
a
completely
different
discipline
from
project
management.
A
So
this
is
funny
because,
if
you
think
of
some
of
the
most
innovative
ubiquitous
technology
we
use
today,
it
came
out
of
government,
but
if
you
try
to
get
a
job
on
USAJOBS
as
a
product
manager,
there
are
no
job
posting.
So
this
was
I
searched
this
about
a
month
ago,
you
get
Lifeguard,
but
there's
thousands
of
project
and
program
managers
right.
So
this
is
a
big
root
cause
problem
government
needs
engineers,
designers
and
product
people
working
in
houses
feds
to
partner
with
vendors.
A
A
So
project
management
is
all
about
delivering
to
your
plan
right
delivering
on
time
on
budget
watch
out
for
scope
creep,
product
management
is
all
about
users,
you've
solved
the
right
problem
and
you've
delivered
something
that
provides
value,
and
this
really
resonated
with
VA
leadership
like
they
got
it
like,
there's
very
little
value
in
delivering
something
on-time
and
on-budget.
If
it's
the
wrong
thing,
so
we
started
to
go
out
and
talk
to
users.
Over
the
past
four
years,
we've
talked
to
more
than
4500
individual
veterans,
so
this
is
in
the
top
corner.
This
is
Marvin.
A
This
is
a
Seattle
homeless,
shelter.
We
did
tons
of
card
sorting
with
veterans
in
the
bottom
corner.
This
is
a
site
and
paired
veteran
did
tons
of
work
with
blind
Veterans
of
America
and
those
of
you
at
large
organizations.
You
know
when
you
go
out
and
start
talking
to
users.
You
start
hearing
about
a
lot
of
problems
and
you
can't
fix
everything
so
do
what
Marty
Kagan
says
find
out
what
makes
your
users
the
most
miserable
and
start
there
and
that's
what
we
try
to
do.
A
So
we
had
a
hypothesis
that
applying
for
healthcare
is
making
many
many
veterans
very,
very
miserable.
Just
10%
of
healthcare
applications
were
coming
in
online
at
VA,
so
in
2015
to
give
you
some
context,
VA
received
582
thousand
applications
for
healthcare,
just
10
percent
came
in
online.
What
that
meant
for
users
is
they've,
got
a
call
a
call
center
and
we
don't
hold
for
hours.
They've
got
a
mail
in
hardcopy
forms
or
they've
got
to
drive
to
a
VA.
A
So
we
proved
that
there
was
this
pent
up
demand
my
dad
70
years
old.
He
wants
to
do
stuff
on
his
phone.
He
wants
self-service,
so
the
healthcare
product
became
one
of
our
foundational
stories
that
we
would
tell
so.
This
is
one
of
my
favorite
plays
because
I'm
not
technical
at
all,
I'm
an
English
major,
so
I
think
good
business
is
about
telling
good
stories.
A
So
one
of
the
things
I
noticed
when
I
came
to
VA
I
was
a
contractor
and
it
was
very
fun
to
come
to
VA
and
be
a
fed
with
some
power.
But
a
lot
of
our
vendor
partners
would
have
these
really
fancy
scoreboards
and
tons
of
KPIs
and
metrics,
and
you
have
to
remember,
there's
an
opportunity
cost
to
collecting
that
data.
You
don't
need
dozens
and
dozens
of
metrics
just
pick
a
few
that
will
help
you
prove
the
hypothesis
or
tell
your
story
so,
with
the
healthcare
product
we
picked
one
metric.
A
These
are
some
of
the
other
metrics
we've
collected
over
the
past
four
years
across
our
portfolio.
This
is
an
article
that
was
just
featured
this
past
Veterans
Day
I
do
want
to
call
out.
We've
got
5,000
people
at
VA
using
github,
so
that's
pretty
awesome,
not
exactly
sure
what
they're
all
doing,
but
it's
pretty
cool.
A
So
one
of
the
things
about
telling
stories
with
the
digital
service,
we
have
some
guiding
principles
and
you'd
be
surprised
how
many
times
this
gets
articulated
as
we
work
find
the
truth
and
tell
the
truth.
So
you
have
to
tell
true
stories,
be
transparent,
tell
data-driven
stories,
but
you
also
have
to
tell
human
stories
and
show
versus
tell
so
I
want
to
tell
the
story
of
Dominic.
Who
is
one
veteran
that
we
worked
with
in
early
2016?
So
Dominic
was
a
Navy
veteran.
A
He
wrote
a
letter
to
President
Obama
he'd
been
in
and
out
of
homelessness.
He
was
eligible
for
VA
health
care,
but
he
was
not
able
to
apply
for
it
successfully.
So
our
team
read
the
letter
and
we
reached
out
to
him
and
he
agreed
to
do
a
user
research
session
with
us,
so
I'm
going
to
show
that
two
short
videos
they're
two
minutes
each
and
we
are
in
the
parking
lot.
So
Dominic
was
working,
but
he
had
to
meet
us
on
his
lunch
break
and
you
can
see
how
it
went.
B
C
B
When
I,
when
I
stand
in
for
something
on
here,
it
directed
me
to
that
and
I
just
remember
it
popping
up
and
there
being
a
link
for
it.
I
know:
I've
done
this
with
finding
to
be
a
fault
for
everything
which
is
sometimes
very
evening.
Sometime,
not
I,
know
I
go
to
that
tent
anything
so
motile.
It's
like
ridiculous.
B
B
B
What
is
the
terminology
that
they're
trying
to
use
for
this,
because
this
doesn't
make
any
sense
and
that's
when
I
ended
up
calling
the
800
number
to
get
someone
to
walk
with
me
through
it,
because
I
was
like
look
I'm
having
an
issue
I,
don't
understand
what
you
guys
are
asking
for
and
I've
seen.
This
come
up
to
a
few
times
where
it's
like
you
need
to.
You
need
to
update
the
PDF
for
it.
I'm
like.
C
B
Is
annoying
that
you
can't
use
whatever
concrete
PDF
browser
that
you
have
like
if
you
got
Adobe
the
current
form
of
Adobe,
you
can't
open
it
it's
the
problem.
This
is
a
huge
problem,
because
some
people
don't
have
the
financial
to
upgrade
their
Adobe
or
to
even
upgrade
their
to
a
computer
that
carries
the
current
flash
drive
or
you
know
PDF
reader
on
it
and.
C
B
A
C
B
Is
definitely
easier
to
read
a
larger
font?
Okay,
it's
very,
very
much
straightforward,
which
is
great
because
the
other
website
is
like
takes
you
around
the
corner
and
over
the
meadow
and
tries
to
lead
you
in
a
back
door.
That's
blocked
with
spiky
IEDs
like
this
is
super
straightforward
click
to
enter
your
name
and
general
information.
The
other
site
is
not
like
that.
It's
like
looking
at
like.
C
C
C
B
A
Okay,
so
that
was
just
one
user
research
session
we
did
of
about
4,500
and
what
was
really
helpful
with
this,
the
Dominic
video
we
call
it
is.
We
had
all
that
data
to
make
the
case
that
the
form
we've
got
up
there
now
is
not
working,
but
when
we
showed
this
to
secretary
McDonald
and
Sloan
Gibson
and
Sloan
just
put
his
head
on
his
desk
and
he's
like
we
have
to
fix
this.
So
don't
underestimate
the
power
of
qualitative
data.
A
A
So
the
first
thing
we
learned
very
quickly.
This
is
a
painful
lesson,
was
that
old
back-end
systems
and
manual
workarounds
aren't
going
to
scale
when
you're
exponentially
increasing
usage.
So
don't
forget,
we
are
quietly
standing
up.
These
new
products
on
a
site
were
not
marketing,
it
says
beta
all
over
it
and
so
on
one
of
our
first
products
we
launched
that
had
about
62
transactions
a
day
within
the
first
week
it
went
to
500.
So
that
was
a
problem
we
screwed
we
screwed
up
on
that,
but
what
we
did
was.
A
It
was
really
interesting
because
we
got
a
lot
of
pressure
like
take
the
new
thing
down.
We
will
just
go
back
to
the
old
thing,
with
60
a
day
that
was
fine
and
what
marina
said
was
no
we're
gonna,
keep
it
up.
That's
a
forcing
function,
we're
gonna
the
digital
service
is
going
to
work
with
you.
We're
gonna
fix
it.
So
this
is
a
lesson
you
only
have
to
learn
once
but
back-end
systems
old,
back-end
systems
are
a
big
constraint
in
government.
A
When
we
were
launching
products,
we
did
as
much
as
we
could
to
optimize,
but
sometimes
you're,
just
stuck
with
that.
Those
old
back-end
systems
forms
are
a
really
big
part
of
user
experience
in
government.
So
you
can
see
the
first
form
VA
has
like
535
forms
veterans
can
use
to
access
benefits.
Our
first
forum
took
us
five
months
to
develop.
We
got
faster
and
today
it
takes
the
digital
service,
five
days
to
launch
an
online
forum.
A
So
here's
some
topics
of
some
of
the
biggest
mistakes
we've
made.
The
first
thing
I'll
talk
about
is
tools,
so
in
government
I.
Think
having
modern
productive
tools
is
one
of
the
biggest
constraints
that
we
face.
So
once
you
get
through
the
procurement
and
the
security
hurdles,
what
we
learned
is
you
can't
just
stand
up
some
relatively
easy
tools,
like
slack,
so
you
have
to
anticipate
training,
and
you
also
have
to
be
aware
of
a
network
effect.
A
So
you
have
to
have
a
core
of
people
using
these
tools,
so
so
new
users
can
kind
of
see.
Oh
I,
don't
use
this.
The
way
I
use,
email
so
drew
is
one
of
our
first
champions
and
trying
to
get
his
team
to
use
slack,
and
it
was
a
ghost
town,
so
Drew
and
I
would
be
slacking
each
other
messages
all
day,
and
so,
if
the
tool
becomes
just
one
more
thing
to
check,
that's
a
problem
so
don't
underestimate
how
much
training
you
need
and
that
network
effect
for
these
new
tools.
A
The
second
big
mistake
the
digital
service
made,
was
around
handoffs,
so
we're
a
tiny
team,
we're
25
to
30
people.
So
it
was
always
the
plan
that
we
would
launch
Minimum
Viable
products
and
then
hand
them
off
to
our
VA
IT
counterparts.
So
there's
8,000
people
in
VA,
Oh
int,
supported
by
another
8,000
contractors.
We
did
a
horrible
job
in
the
in
those
early
days
of
anticipating
the
rigor
of
a
handoff
right,
so
that
resulted
in
this
kind
of
launch
and
abandoned
experience
for
some
users
our
first
year.
A
So
now
we
have
very
rigorous
kind
of
checklists
and
handoff
activities
and
we
also
have
integrated
teams
where
we
bring
our
teammates
along
throughout
the
development
of
the
MVP
building,
trust
and
discovery.
I'll
talk
about
together,
so
our
digital
service
team
knows
a
lot
about
technology
and
design
and
product.
A
What
we
don't
know
about
is
VA
health
benefits
delivery
and
we
weren't
a
very
humble
group
in
the
beginning,
so
we
we
got
better
at
this,
but
make
sure
that
you
have
integrated
teams
that
you're
bringing
in
experts
and
that
you
recognize
there's
a
lot
of
things.
You
don't
know,
and
the
last
thing
I'll
talk
about
is
the
release
process.
So,
during
our
first
two
years
we
took
a
very
waiver
based
an
exception
based
approach,
which
was
great
for
us,
so
we
could
use
cool
tools.
A
We
didn't
have
to
do
all
those
documents
in
the
SDLC
for
compliance,
but
that
didn't
really
help
the
enterprise
so
really
taking
a
more
enterprise
approach
and
kind
of
fixing.
Those
blockers
for
the
at
for
our
counterparts
was
really
important,
whether
that's
updating
policies
or
our
next
play
is
debunking
some
myths.
So
this
is
also
something
fun
that
we
did,
especially
when
you
have
a
leadership
support.
So
these
are
real
things
that
we
heard
when
we
started
working
at
VA.
A
So
if
any
of
you
work
in
large
organizations,
I'm
sure
there
are
a
lot
of
misinformation,
you
can't
use
github,
it's
illegal.
You
can't
go
to
the
cloud
you
can't
put
in
an
evidence.
Bag
like
none
of
these
things
are
true,
so
getting
the
word
out
and
making
sure
that
people
have
the
right.
Information
is
really
important.
One
of
the
biggest
blockers
on
large-scale
change
initiatives
is
this
mentality
of
like
we've
always
done
it.
This
way,
we've
always
done
it.
This
way.
So
again,
shifting
from
I
have
my
PMP.
A
This
is
how
I've
always
worked
to
know
we're
gonna.
Do
it
differently
is
really
important,
so
ask
why
but
disrupt
with
purpose,
don't
just
disrupt
to
disrupt.
So
when
we
started
asking
questions
like,
why
do
we
need
all
these
documents
and
meetings?
Why
are
we
spending
thousands
of
dollars
to
get
people
PMP
certified?
It
is
a
useless
credential
with
delivering
a
great
digital
product,
and
why
are
we
blaming
security
for
everything?
So
ask
that
second
question,
like
what
specific
security
requirement
is
preventing
us
from
working
the
way
we
need
to
work.
A
And
this
is
the
last
play.
I'll
talk
about
so
I
talked
a
little
bit
about
being
humble,
have
some
humility
and
don't
go
it
alone,
so
it's
30
people
in
a
three
hundred
and
sixty-six
thousand
person
agency
you
have
to
work
with
partners.
The
digital
service
is
certainly
not
the
only
group
at
VA
doing
really
cool
and
innovative
stuff,
so
you
have
to
find
these
pockets
of
innovation
partner
with
them,
help
them
with
blockers
and
celebrate
their
success.
A
So
this
is
one
of
my
favorite
things
that
we
did
at
VA.
That
I
hope
will
go
on
for
a
long
time.
We
started
so
I've
talked
a
lot
about
top-down
leadership
and
the
support
we
had.
But
that's
not
enough.
If
you
want
to
change
a
huge
organization,
you've
got
to
have
bottom-up
support,
so
we
started
a
product
management
and
human
centered
design
training
program,
it's
old
school,
it's
an
in-person
program.
We
run
it
out
of
headquarters.
We
can
do
50
people
in
a
class
and
the
first
day
is
a
classroom
day
in
the
classroom.
A
So
we
invite
we're
gonna
cycle
all
of
our
VA
IT
people
through
it.
In
addition
to
our
business
sponsors,
our
procurement
people
and
our
security
people,
so
during
the
first
day
you
sit
in
class
and
you
learn
what
the
heck
is
product
management.
You
learn
about
human
Center
design
and
agile
and
Minimum
Viable
products,
and
then
we
also
have
leadership.
So
we
have
people
like
drew
and
our
CIO
and
our
deputy
come
in
and
actually
give
people
permission
to
work.
This
way,
that's
really
important.
A
I
call
it
like
the
Merv
Griffin
Show,
which
no
one
on
my
team
gets
that
reference.
But
so
this
picture
on
on
the
right
is
the
day
in
the
classroom
and
then
these
other
two
pictures,
the
second
day
is
really
the
secret
sauce
of
the
training.
So
that's
when
we
actually
go
out
to
the
field
and
talk
to
veterans,
so
this
is
at
the
VA
Medical
Center
in
DC,
and
we
break
out
into
small
teams.
A
So
we'll
have
small
teams
at
you,
know
prosthetics
and
pharmacy
and
long-term
care,
and
you
can
see
this
picture
in
the
middle
is
in
the
lobby
with
what
I
think
is
a
pretty
scary,
dangerous,
popcorn
machine.
But
it's
basically
just
our
team.
Talking
to
veterans
and
building
empathy,
we're
not
collecting
requirements,
you're,
just
kind
of
an
empathy
building
and
my
set
shift
activity.
So
this
is
a
great
program
we
do
this
every
month.
A
A
They
also
teach
a
module
in
that
product
management
training,
which
is
fantastic
in
addition
to
teaching
a
module
at
the
Georgetown
program,
which
is
really
cool,
so
invest
in
the
right
training
for
your
team,
I'm,
biased
because
I
teach
in
the
Georgetown
program,
but
send
your
folks
to
marty,
Kagan's
class
or
Harvard
has
a
class
but
make
sure
you're
getting
them
the
right
training
and
not
the
traditional.
You
know
deliver
on
time
and
on
budget
be
humble,
so
I
wanted
to
talk
a
little
bit
about
as
a
product
person.
A
It's
easy
to
have
empathy
for
users
right,
that's
kind
of
built
into
the
fabric
of
product
management,
but
have
empathy
for
your
colleagues,
because
it's
a
scary
thing
trying
to
use
these
new
tools
and
I
discovered
that
when
I
started
at
the
digital
service
at
VA
and
I
come
from
some
pretty
good
companies
like
Gartner
and
Deloitte.
We
did
not
work
the
way,
engineers
and
designers
from
Microsoft
and
Google
and
Code
for
America
work
at
all.
A
A
So
hopefully,
some
of
those
plays
and
lessons
learned
helped
you
go
back
to
your
if
you
work
in
a
large
organization
and
affect
some
change
so
think
big,
but
chip,
small
and
fast
focus
on
delivery.
Tell
true
stories
that
matter
get
smarter
as
you
work.
Sometimes
that
means
failing
in
the
open,
be
a
mythbuster
and
have
some
humility.
Thank
you.