►
From YouTube: Progressive Delivery UX/PM meeting 2020-06-09
Description
Meeting to improve alignment between Product Management and UX for progressive Delivery
A
Thanks,
let's
jump
right
into
the
issues
I
see,
nadia
is
not
attending
a
gta
meeting
conflict.
So,
let's
move
on
without
her
for
now
ux
ux
research
issues
in
progress
right
now.
First
up
is
this
custom
post
solution,
validation
for
add,
cancel
rollout
button.
A
A
B
So,
following
all
the
research
we
did
for
post
deployment
monitoring,
we
decided
to
add
alerts
to
the
environment
page
to
show
someone
that
something
is
wrong
with
their
deployment
and
then
they
can
take
action
on
it,
so
they
can
halt
the
role
out
or
they
can
roll
back.
This
is
all
done
from
the
environment
page
and
when
I
thought
about
it
was
like
hey,
wait
a
minute,
we're
leveraging
the
alerts
in
order
to
do
this.
Why
not
do
the
reverse?
B
Why
not
go
to
the
alerts
page
and
if
there's
an
alert,
do
the
same
action,
stop
or
hold
the
roll
out,
and-
and
I
suggested
this
to
sarah
who's-
the
pm
for
monitor
and
she
thought
that
was
a
great
idea
and
then
there's
just
started
this
whole
discussion
inside
the
threads
with
the
designer
from
there
who
said:
maybe
this
isn't
the
right
place?
Maybe
we
should
put
it
somewhere
else?
B
Maybe
we
and
there's
some
specific
alerts
that
we
don't
take
action
on,
so
maybe
it
needs
to
be
in
a
global
place.
So
this
kind
of
needs
solution,
validation
as
to
where
the
right
place.
To
put
it,
it's
definitely
a
smart
idea
to
do
that.
Just
the
question
is
where.
A
All
right
sounds
good
in
terms
of
how
we
wanna
how
we
wanna
solution,
validate
that.
A
Like
like
laura,
do
you
think
we
can
set
this
up
as
an
asynchronous
ux
research
like
as
in
like
that?
We
can
just
push
this
out
without
you
know
having
to
be
there
in
synchronous
interview
sessions.
C
C
As
far
as
I
know,
qualtrics
doesn't
record
a
video,
so
you'd
have
to
hope
that
they
typed
in
an
answer
to
the
question,
because
you
can
show
them
a
screen
and
call
tricks.
You
can
show
them
a
picture
and
you
can
ask
them
a
question
and
you
can
hopefully
hope
that
they
get
an
answer.
We
can
try
it
and
see,
because
that's
the
critical
part
you
want
to
say
like
which
one
do
you
prefer
and
why?
Because
that's
where
you're
gonna
get
the
reasoning
behind
putting
it
wherever
we
decide
to
put
it.
C
B
You
can
go
into
an
internal
group,
it's
just
figuring
out
what
people
would
press
and
why
and
where
they
expect
things.
So
we
have
some
smart
people
in
the
company.
C
Well,
it's
so
smart
for
not
smart.
It's
just
like
preferences.
I
think
at
this
point
really
is
what
we're
trying
to
understand
right.
So
we
can
totally
do
it
with
internal
folks
as
long
as
you
feel
like
they're
not
going
to
bring
more
to
the
picture
than
a
regular
customer
would
who
doesn't
work
internally
here.
A
Expectations,
what
do
you
think
for
it
like
do
you
think
internal
users
are
gonna,
be
reflective
of
our
target
audience
here.
B
I
do
think
it's
better
to
do
sync
like
personally
synchronous
sessions,
so
people
can
walk
through
and
voice
what
they're
doing.
I
haven't
done
many
solution
validations
here
at
gitlab,
but
the
few
that
I
have
it
was
really
valuable
and
we
changed
a
bunch
of
stuff
based
on
the
fact
that
people
have
said.
Oh,
I
expected
it
to
be
here.
Oh,
I
thought
that
this
would
do
this,
so
I
think
that's
something
that
you
don't
gain
in
asynchronous.
C
Yeah,
you
can
try
to
ask
them,
like
I
said
like
in
the
survey
or
whatever,
and
when
you
present
the
idea
of
the
design,
you
can
ask
them
and
hope
that
they
answer
you
or
hope
that
you
that
you
get
enough
information
from
their
answer.
That's
probably
the
better
statement.
Okay,
so
yeah.
What
we
can
do
is
we
can
try
the
survey
tactic
first
and
see
what
we're
getting
back
and
if
we're
not
getting
good
data
or
we're
getting
questionable
data,
we
can
then
switch.
Stop
that
and
switch
and
just
talk
to
people.
C
A
What
do
you
think
for
it.
B
So
again
my
personal
preference,
but
it's
not
up
to
me
because
this
is
you're.
The
dri
would
be
to
do
like
pull
up
several
options
and
have
people
go
through
them
and
see
what
the
easiest
one
was
to
follow.
A
Yeah
all
right
sounds
good.
Let
me
let
me
get
this
information
into
the
into
the
dedicated
solution,
validation
issue
and
then
I'll
ping.
You
there
both,
I
think,
to
be
honest
for
a
quick
internal
investigation
planning.
You
know
a
few
synchronous
sessions
just
for
a
few
minutes
like
10
minutes.
10
minutes
increments
is
very
doable
like
the
way
that
I
was
thinking
about.
A
This
is
that,
if
you
want
to,
you
know
bring
in
participants
like
external
participants
just
for
like
a
five-minute
task,
and
that
is
kind
of
like
not
worth
the
time
and
like
it's
like.
We
gave
them
like
a
huge
bounty
for
just
five
minutes
of
work
like
yeah.
It's
like,
I
don't
think.
That's
that's
a
good
idea,
yeah.
I
agree.
A
Okay,
cool
and
work
this
into
the
issue
and
move
forward
most
likely
with
internal
quick
sync
sessions,
cool
a
b
testing
guide.
Again,
I've
been
saying
this
before
this
will
be
the
week
I'll
complete
it.
I
have
been
delaying
this
because
of
other
priorities,
so
I
want
to
have
a
quick
discussion
about
prioritization
of
ux
research.
Work
versus
feature
work
that
is
targeting
the
immediate
milestone
or
ahead.
I
know
we
want
to
move
towards
that
ideal
state
where
we
do
not
have
to
worry
about
that
anymore.
A
A
On
the
other
hand,
if
we
never
focus
on
ux
research,
then
that
does
not
get
done
so
I
want
to
see
like
hey.
What
is
a
good
way
you
would
like
me
to
think
about
you
know
now
it's
the
time
to
focus
on
your
research
versus
the
others.
This
is
a
work
in
progress
for
me
as
well.
I'm
trying
my
best
here,
but
there's
a
lot
of
things
that
you
know
on
my
plate.
So
I
think
it's
interesting
to
have
at
least
your
opinion
on
this.
B
The
priorities
we're
all
scaling
a
bunch
of
juggling
a
bunch
of
things
in
the
air
and
there's
always
the
important
things
and
the
urgent
things,
and
you
don't
want
to
be
in
a
situation
where
you're
only
doing
urgent
things.
So
it's
it's
how
you
balance
your
time,
and
I
don't
think
that
I'm
the
right
person
to
answer
how
to
do
that.
Maybe
take
that
with
other
designers.
C
My
advice
is
that
everybody
has
to
stop
working
at
100
that
they
were
before.
You
have
to
start
cutting
back
on
both
pieces.
So
then
they
both
add
up
to
100,
because
if
you're
doing
100
percent
of
research
and
100
percent
of
design-
that's
200
percent,
that's
in
you
can't
keep
it
up.
Nobody
can
keep
it
up
product
can't
design
can't.
C
So
the
velocity
overall
velocity
has
to
change
to
accommodate
the
additional
activities
that
have
to
happen,
and
that's
the
hardest
part
that
I've
seen
from
everybody
who
I
talked
to
at
this
company,
because
they're
like
I
used
to
be
able
to
do
x,
y
and
z
in
a
milestone,
and
now
I'm
struggling
to
do
that
and
a
b
c
d
well
yeah,
because
you're
doing
a
whole
bunch
of
other
stuff.
So
it's
it's
kind
of
trying.
You
have
to
reframe
it.
You
have
to
say
I
cannot
do
this
much.
C
I
can
do
this
much
because
I
also
have
to
do
this
and
it's
hard,
because
the
business
is
depending
on
the
assumptions
that
we
made
eight
years
eight
months
ago
about
how
much
people
can
get
done,
and
we
just
have
to
shift
that.
We
have
to
start
saying
well,
I've
got
to
do
problem,
validation
or
I've
got
to
do
solution,
validation
and
that's
going
to
cut
down
on
the
amount
of
design
work
that
I
can
get
done.
Amount
of
product
work
that
I
can
get
done
as
well.
B
But
I
also
want
to
emphasize
that
we
can
like
everything
in
gitlab
iterate.
So,
yes,
you
know
you
if
you're
unhappy
with
the
guide
for
a
b
testing,
maybe
there's
just
too
much
of
it
and
we
can
cut
down
scope
because.
A
B
Yeah,
so
I
just
think
we
need
to
scope
things
out
because
I
it's
it's
really
hard
to
plan
these
discussion
guides
because
there's
you
know,
there's
plan
and
then
there's
reality.
So
it's
a
really
good
base
for
when
you
start
the
interview.
But
then
your
every
interview
takes
its
own
course
and
it's
just
there
as
a
guideline
and
the
interview
itself
takes
its
own
life.
So
I
think
well,
but.
C
B
You
need
to
see
that
you
have
like
the
the
questions
that
you
want
answered
or
answered,
but
sometimes
someone
will
give
you
something
that
you
didn't
think
about,
and
you're
gonna
like
jump
on
that
and
you'll
learn
like
a
bunch
of
things
that
are
off
script.
So
that
sense,
I
think
you
know
the
discussion
guide
is
a
great
tool.
B
B
C
B
C
Badly
and
that's
okay,
like
my
discussion
guide,
template
is
a
template
to
start
from,
and
I
feel
like
discussion
guides
are
very
personal.
How
people
I
like
a
good
outline,
that's
how
I
live,
but
some
people
don't
and
it
you
just
have
to
put
them
in
there,
so
you
can
reference
it,
like.
You
said
for
your
guide
to
have
that
conversation
with
that
person.
So
yeah.
A
C
B
It
was
like
such
a
mess,
but
nah,
but
like
I
saw
I
totally
went
off
script
all
the
time,
so
that
was
my
take
on
it
and
yeah.
I
think
it's
really
good
to
for
a
starting
point
and
like
take
the
first
interview
as
a
learning
point
and
then
yeah
like
build
on
it
iterate
like
everything
we
do
and
give
them.
C
Yeah
yeah,
the
only
danger
you
might
face
if
you
keep
iterating
so
much
every
time
is
that
you
don't
get
a
consistent
set
of
questions
answered.
So
it's
harder
to
see
the
trends
or
the
patterns
across
people.
That's
that's
the
only
thing
you
might
run
into
depending
on
how
much
you
vary
between
your
your
customer
interviews.
A
Yeah,
I
totally
understand
thanks
for
that.
This
is
a
good
feedback.
C
A
A
Going
another
half
an
hour,
so
it
should
be
done
in
two
hours.
Just
been
I've
been
de-prioritizing
it
because
there
are
other
issues
that
I've
been
asking
more
of
my
like
how
I've
been
prioritizing
them,
gotcha.
Okay,
with
that
we
kind
of
roll
into
the
next
one,
which
I
want
to
have
a
quick,
a
quick
look
into.
I
originally
put
this
in
the
one-on-one
between
us
or
it.
I
don't
know
how
much
time
we
have
left.
A
Let
me
see
I've
got
a
little
bit
a
little
bit
more.
This
should
be
kind
of
quick
to
just
get
an
a
little
bit
of
a
picture
on
like
hey.
What
actually
is
up
for
me
to
be
worked
on
currently
and
in
13.2,
and
how
that
kind
of
relates
to
you
know
the
interviews
that
are
going
to
come
up
in
the
next
two
to
three
weeks,
and
I
know
I'm
in
the
process.
It's
a
work
in
progress.
Iteration
part
of
part
of
my
process
improvements,
you
know,
waits
to
be
added.
A
In
that
case,
you
know
make
it
a
little
bit
more
comparable
month
to
month
what
I
can
take
on,
but
as
we're
not
there
exactly
yet
currently,
when
looking
at
the,
let
me
share
my
screen
for
a
quick
bit
currently
when
we
look
at
the
pd
needs
ux
review
epic.
We
already
looked
at
this
in
101
or
it,
but
if
I
look
at
like
hey
what
needs
to
still
be
worked
on
in
13.1,
it's
basically
these
two
issues,
which
is
fine
I'll,
get
to
this.
As
soon
as
I
can.
A
A
A
What
do
you
think?
What
do
you
expect
in
terms
of
like
like
what
is
reasonable
to
expect,
or
what
do
you
think
this
is
the
minimum
I
need
for
it
from
you,
for
you
know,
13.3
target
milestone,
but
for
you
to
be
worked
on
the
13.2,
but
because
I
feel
that
five
issues
might
be
a
bit
too
much
alongside
with
ux
research,
two
items
and
my
additional
work
in
ci
onboarding
zitika.
B
So
I
don't
I'm
not
going
to
give
you
an
answer
because
I
don't
know
the
right
answer,
but
but
we
can
like
briefly
discuss
the
issues
and
I'll,
tell
you
where
my
head's
at
so
implement
blue
green
switch
on
deploy
board.
You
basically
know
what
we
want
to
do
there.
I
don't
think
there's
a
lot
of
effort
in
order
to
complete
it.
I'm
presuming
like
I'm
not
a
designer,
so
you
can
say
you're
totally
wrong.
B
This
is
a
ton
of
work
and
and
then
you
know,
I'm
going
to
trust
your
word,
but
but
since
you
kind
of
know
where
what
I
want
just
to
make
it
more
get
lobby
and
in
the
correct
format
that
doesn't
look
like
you
need
to
invent
a
whole
concept
right.
So
I
don't
think
there's
a
lot
of
work
there.
But
again
I'm
going
to
be
cautious,
because
I
don't
know
what
I'm
talking
about.
Okay,.
A
Yeah,
I
would
like
to
say
like
how
how
I
have
been
looking
at
this
is,
even
though
we
sometimes
know
what
we
want.
It
does
require
to
have
a
discussion
alongside
with
engineering
in
terms
of
like.
Is
this
going
to
be
feasible
within
one
release?
What
are
we
going
to
scope
off?
I'm
going
to
detail
my
design
to
be.
You
know
towards
that
single,
like
it's
going
to
have
a
few
iteration
rounds
before
we
arrive
at
a
solution
which
can
be
implemented
in
one
release,
right.
B
Yeah,
so
this
is
a
top
priority
item,
so
I
would
say
definitely
this
should
be
stay
on
your
list,
the
next
one.
I
don't
remember
what
it
was,
but
if
you
can
go
back.
A
Set
deployment
traffic
weight
yeah.
Why.
B
Is
a
little
bit
trickier
if
I
had
to
choose
of
the
two
which
to
which
to
focus
on
it
would
be
the
first
one
and
the
reason
is
a.
We
have
a
way
to
do
this
through
the
api.
So
user
can
still
do
this
without
doing
it
in
the
ui,
so
they
still
have
the
functionality
and
b.
B
I'm
not
sure
what
the
right
place
to
put
this
in
is
we're
kind
of
introducing
a
new
concept
which
is
larger
than
this
single
issue.
So
it
needs
a
lot
of
thought.
B
A
Worries
I
I
got
a
thick
skin,
just
say
what
you
want
and
I'll
take.
It
no
worries
more
work
and
green,
we're
here
to
create
the
product
right.
I'm
just
like
I'm
gonna
say
this,
as
is
my
main.
My
main
thing
to
to
figure
out
correctly
for
myself
is
to
be
able
to
deliver
what
I
commit
to,
and
it's
really
a
hard
problem.
A
But
when
I
got
it
right,
you
know
it's
gonna,
it's
gonna
fly
off
the
of
the
of
the
band
without
a
problem.
We're
gonna,
you
know,
get
some
real
velocity
going
even
more
than
we
already
do.
B
At
all,
that's
like
the
definition
of
a
product.
Okay,
add
cancel
button
so
that
we
talked
about
this.
We
need
to
do
solution,
validation
here
before
implementation.
B
A
Is
then,
let
me
see
this
is
intended
for
so
we
are
now
only
discussing
work
that
will
touch
13.3,
so
we're
not
talking
about
work
that
targets
13.2,
but
it
will
be
worked
on
by
me
in
13.2.
B
So
when
I
open
this
issue-
and
I
told
you
we
added
first
of
all
to
the
environment
page-
to
take
action
on
alerts
when
I
added
this
issue-
I
just
thought:
maybe
this
one
would
be
quicker,
but
then
it
started
this
whole
controversy
so
yeah
it
wasn't
part
of
the
mvc.
In
the
beginning,
it's
a
nice
to.
B
It's
really
nice
cross
work
between
different
stages.
It's
great
collaboration,
but
you
know
I
assume
this
is
the
solution.
Validation
will
take
some
time
and
then
we
need
to
formulate
a
plan
plan
so
to
be
more
realistic,
it
will
probably
go
to
13
4
than
it
would
13.
A
I
mean
like:
let's,
let's
see
how
soon
I
can
get
those
those
interviews
scheduled,
I
think
internally,
like
it's
so
much
more
accessible
to
schedule.
A
few
calls
with
a
few
developers
internally
than
it
is
to
you
know,
reach
out
get
recruitment
going
then
get
responses,
then
schedule
them
and
then
have
the
meetings.
A
B
Yeah
well,
this
is
one
of
the
quarterly
goals
for
q3.
B
A
Yeah,
it
seems
that
way
like
it
seems
a
little
bit
like
a
little
bear
just
from
like
looking
at
it
right
group.
Deploy
keys,
on
the
other
hand
like
this,
is
similar
as
project
deploy,
keys
or.
B
Yes,
well,
we,
the
release
team,
took
on
deploy
keys
a
few
milestones
ago
as
our
responsibility,
so
we've
been
doing
a
lot
of
work
around
the
deploy
keys
and
we
did
group
deploy
tokens,
and
now
we
want
to
do
group
deploy
keys.
A
B
We
have
a
lot
of
work
in
get
lab
across
all
teams
to
do
group
actions,
so
this
kind
of
falls
under
that
umbrella.
A
Cool,
let
me
see,
and
it
is
state
to
cancel
pipeline
reason
in
the
jobs
page
in
case
incremental
rollouts
will
stop
due
to
an
error
rate
threshold.
B
Yeah,
so
you
see
my
beautiful
maka,
it
says
this
job
was
controlled
due
to
error
rate.
B
A
B
B
A
Oh
yeah
and
it
and
it
is
related-
let
me
see
to
the
post-deployment
monitoring
research
insights
right
like
this
has
been
validated
that
this
is
a
problem
and
that
we
need
to
do
this.
A
Problem
is
validated
so
quickly
going
through
the
others
like
from
that
perspective.
Ux
research
perspective,
which
of
those
issues
like
do,
have
some
validation
from
a
problem
perspective
already
there
do.
You
know.
B
They
all
have
been
validated
the
the
one
that
needs
a
solution.
Validation
is
the
cancel
to
alerts
page
now,
two
go
to
two
down.
No
another.
B
That's
the
only
one
that
needs
solution,
validation,
just
because
there
was
a
lot
of
controversy
of
where
we
need
to
add
it,
and
why
so,
I
think
it's
worthwhile
to
do
a
solution,
validation
here
and
in
terms
of
things
that
can
be
postponed.
The
one
above
you
set
deployment
traffic
via
ui
is
one
that
I
would
be
willing
to
postpone
and
the
group
deployed
keys.
A
I
can
do
let
me
see
which
one
was
that
group
deployment
keys.
This
one
is
that.
A
Needed
can
be
postponed,
but
right
or
not
all
right.
Thank
you
so
much
for
this
information.
I
can
work
with
this.
I
let
me
go
over
this
this
week
and
then
in
one-on-one
next
week
we
can
finalize
our
approach
with
this.
B
A
Let
me
see:
was
there
anything
else?
Oh
yeah,
quick
idea,
though
we're
on
time.
We
can
push
this
on
to
next
meeting
no
problem.
B
A
It's
more
of
like
just
an
idea
towards
direction
pages.
Did
we
have
one
of
those
little
meeting?
No,
we
didn't.
Where
did
you
copy
and
paste
that
laurie
the
other
time
I
see
copy
and
paste
like
end
of
meeting?
We
went
over
time
little
like
some
line
you've
put
in
like
in
other
meetings.
Now
I
know
we
have.