►
From YouTube: 2023-08-23 Product Analytics UX/PM Sync
Description
Today we talked through some issues that came up in implementation of the Dashboard Designer re-design and found a path forward.
A
Hey
this
is
the
product
analytics
pmux
weekly
sync
for
August
23rd
2023.
We
have
one
agenda
item
today
discussing
a
demo
of
the
proposed
pivot
on
the
dashboard
in
BC
yeah.
Do
you
want
to
kick
us
off
and
share
your
screen
to
talk
through
this.
B
Cool,
let
me
start
off
with
the
issue.
That
is
the
context
for
this
jet
item.
That
is,
the
new
custom
dashboard
flow,
specifically
the
dashboard
designer
MVC.
So
this
is
the
first
iteration
that
we
want
to
launch
to
customers.
It
is
currently
being
tested
by
some
internal
users
that
have
this
feature
flag
enabled,
and
with
this
MVC
we
have
a
new
design.
B
The
design
looks
pretty
slick,
but
while
implementing
this
like,
we
ran
into
some
like
practicality
issues
in
terms
of
just
how
the
grid
works
and
call
to
action
is
being
shifted
around
and
also
some
of
the
edit
experiences
a
bit
of
a
like
has
like
a
question
mark
on
it,
for
example,
and
I
think
the
designs
include
anything
that's
like
tells
us
how
this
title
should
actually
be
edited
or
what
the
edit
state
looks
like.
B
This
HTML
content
there's
like
an
attribute
you
can
put
on
for
that
or
switching
it
to
an
input
box
and
eventually,
while
talking
through
this,
we
just
ideated
and
took
a
bias
for
action
since
we're
lacking
a
ux
designer
or
a
dedicated
ux
designer
for
our
team
to
try
and
come
up
with
an
approach
that
solves
the
issues
that
we
were
like
talking
about
in
the
concert
and
yeah
I've
gone
into
detail
here
on
the
comments.
B
I'm
I'm
sure
everyone
in
this
call
has
already
seen
it,
but
basically,
what
we
saw
is
that,
like
we
can
maybe
do
like
a
iteration
between
what
we
have
now
and
our
vision
and
use
some
of
the
existing
pajamas
components
and
use
some
of
our
existing
forms
in
settings
and
like
the
new
MRN
issue,
screen
as
a
inspiration
to
come
up
with
something
practical
that
works
for
now,
and
it's
like
good
enough
for
our
users
for
fish
release
and
basically,
what
we
came
up
with
was
this
design
that
I
can
demo
here.
B
So
this
is
the
create
your
dashboard
screen.
We
took
the
dashboard
title
added
from
this
is
how
it
looks
when
you
create
a
new,
Mr
or
issue
which
I
can
just
quickly
show
here.
As
you
can
see,
it
has
a
similar
look
and
feel
to
it.
You
have
like
your
page
title
describing
the
action
that
you're
taking
and
then
you
have
your
fields
that
you
need
to
input
so
that
you
can
create
issue
similar
if
I
don't
have
the
title.
B
So
what
I
also
have
here
is,
as
you
can
see,
you
get
error.
This
could
probably
be
on
that
form
group
instead
of
like
jumping
up,
but
the
idea
is
there
we
added
an
empty
State
here
to
just
show
like
hey,
so
it's
basically
guide
first-time
users
into
there's
two
things
you
need
to
do
like
one
is
title.
B
Think,
first
time,
users
probably
going
to
ignore
that
filter,
but
it's
still
useful
to
to
show,
and
similarly
our
edit
experience
and
that's
the
other
part
that
this
design
kind
of
neglected
a
bit
is
that
we
currently
have
an
edit
experience.
It's
not
just
new
dashboards.
It's
also
an
added
experience
that
we
need
to
support
and
that
whole
step
flow
doesn't
fit
well
in
with
the
edit
experience.
B
So
what
we
have
here
is
pretty
much
more
or
less
the
same.
We
just
don't
have
that
empty
state
that
might
come
back
in
some
form
when
we
add
the
ability
to
remove
dashboard
or
panels
from
the
dashboards,
because
you'll
maybe
end
up
with
something
that
has
no
like
yeah,
like
users,
may
want
to
start
from
scratch
and
remove
all
panels.
That
does
bring
up
the
interesting
question
about
whether
we
should
let
users
save
an
empty
dashboard.
B
My
immediate
gut
feeling
is
yes,
like
I,
don't
know
how
you
what
they're
going
to
use
it
for,
but
we
probably
don't
want
to
limit
their
action
without
good
reason,
yeah
and
then
I'll
I'll
stop
first
there
for
any
questions,
because,
like
from
here,
I
can
talk
about
like
a
potential
second
iteration,
but
first
I
want
to
stop
here
for
any
questions
or
feedback
about,
like
specifically
change
what
you
think,
whether
we
should
like
take
advice
for
action
and
go
ahead
with
us
or.
A
A
The
other
question
I
have
is
about
the
filter.
So
if
we're
rendering
data
and
showing
the
filter
it's
helpful,
but
if
I'm
editing
a
dashboard
and
I
change
that
filter.
My
first
instinct
is
that
that
Now,
the
default
filter
on
the
dashboard
and
I,
don't
think
that
that's
true.
B
I
see,
that
is
a
good
point
like
I
think
the
filter
is
kind
of
necessary
here,
because
it
gives
you
that
same
what
you
see
what
you
get
experience
in
terms
of
like
you
can
already
play
around
with
your
visualizations
that
you've
added
and
give
you
a
good
idea
of
what
it's
actually
going
to
look
like.
At
the
same
time,
you
raise
a
good
point
in
that
users
may
think
that
this
is
like
a
static
element.
A
B
A
There
for
the
data
exploration,
part
of
it,
but
I
wouldn't
get
hung
up
on
that.
I
I,
like
this
I
think
this
is
looking
good.
It's
consistent
with
the
edit
experience.
As
far
as
the
list
of
things
are
there,
the
title:
is
there
I'm
saving
the
dashboard,
yeah
I,
don't
know
about
letting
users
save
an
empty
dashboard,
but
I
think
that
again,
that
shouldn't
hold
us
up.
I
appreciate
the
bias
to
action.
I
think
that
this
is
a
good
path
forward
for
the
for
the
NPC.
B
C
We've
got
a
separate
issue
to
discuss
the
visualizations
right
hand:
sidebar
yes,
based
upon
Kevin's,
work
previous
work
and
trying
to
iterate
towards
that
as
well,
which
we've
been
beyond
discussed
when
we
sat
down
and
explained
about
the
omvc
structure.
There's
it's
kind
of
just
to
wanted
to
ask.
There's
conversations
going
on
in
this
issue
with
sir
Justin
Mendel
yeah.
Probably,
is
that
in
relation
to
something
we're
going
to
be
working
on
based
off
these
designs,
or
is
that
in
as
a
secondary
augmentation
like
a
next
iteration.
B
That
is
probably
a
next
situation,
so
what
Justin
and
I
have
been
discussing
when
James
has
also
been
involved
in
this
discussion,
is
about
the
call
to
action
to
open
up
the
available
visualization
side
drawer.
So
we
have
these
other
parallel
issues
to
this
dashboard.
Designer
MVC,
which
is
like
I,
said
to
Implement
our
new
ux
designs,
which
transport
migrates
the
existing
dislike
temporary
sidebar,
fixed
thing
that
we
have
here,
which
looks
pretty
ugly
into
a
side
roller
component.
B
The
issue
with
the
side
roller
component
is,
you
need
a
call
to
action
to
actually
open
it
and
the
good
news
is.
We
now
have
two
options
which
I
think
are
viable,
but
I've
asked
Justin
for
his
feedback
in
terms
of
like
which
he
prefers
like
we
can
either
have
this
elongated
button
slash
pseudo
panel
at
the
top.
B
This
solves
the
thing
of
we
have
a
panel
that
moves
around
and
it
also
breaks
the
what
you
see
what
you
get
like
experience
in
terms
of
like
having
a
panel,
that's
sticky,
to
the
top
left,
because
this
is
small
enough
where
I
think
like.
If
we
remove
it,
if
users
aren't
going
to
be
like,
oh,
this
isn't
what
I.
A
B
Know
set
up
in
the
edit
experience
and
similarly
we
could
make
it
in
line
with
the
filter
and
add
it
to
the
right
hand.
Side
here,
I
think
both
Works
I
think
it's
also
something
we
can
iterate
on,
but
so
far
I've
pinged
Justin
for
his
feedback,
depending
on
what
he
says.
I
think
we
can
go
either
direction.
B
We
could
also
do
one
and
then
implement
the
other,
but
this
is
for
a
follow-up,
so
I
think
by
the
sounds
of
it
from
what
we
just
discussed
like
we
are
unblocked
to
carry
on
with
this
first
iteration
of
design,
which
I'm
like
demoing
here
and
all
we
need
to
do
in
terms
of
like
feedback,
is
just
updates.
The
error
experience
in
terms
of
highlighting
the
the
field
that
users
need
to
to
input.
A
On
when
the
dashboard
is
empty,
we
have
the
caller
action
to
add
a
visualization.
Is
there
an
indicator
in
that
sidebar.
C
I
think
in
the
next
iteration,
if
we
have
got
that
call
to
action
somewhere,
then
we
can
add
a
call
to
action
to
that
empty
State
like
we
normally
would
so
that
way,
users
can
just
click
the
button
and
it
will
open
that
side.
Drawer.
A
C
It
looks
like
plain
text
or
a
link.
It
doesn't
look
like
something
actionable
in
the
same
way
like
the
current
designers
yarn,
showing
off
with
the
check
multi
select
checkbox
that
looks
actionable
like
I
click.
It
I
do
something,
so
we
did
discuss
this
and
ways
and
things
that
we
might
be
able
to
do
to
make
it
more
obvious
visually
that
you
want
to
do
something
with
this.
A
Since
we're
already
talking
about
an
iteration
out
of
that
view
into
something
we
know
is
more
actionable
I,
don't
think
we
need
to
spend
a
lot
of
Cycles
trying
to
make
this
iteration
of
it
more
actionable,
so
I
I
think
we
can
move
on
and
if
we
get
user
feedback,
hey
I
didn't
realize
that
that
was
a
thing.
I
could
drag
over
I
click
to
create
visualization
and
then
I
recreated,
something
that
was
already
there.
We
know
that
this
iteration
should
solve
for
that,
and
we
can
more
quickly
move
to
it.
A
So
not
not
too
worried
about
it.
It'll
be
something
for
us
to
look
at
in
feedback
or
kind
of
the
hypothesis
would
be.
This
version
tell
user
that
they
should
or
how
to
add
something
where's,
the
other
one
didn't.
A
B
A
Is
looking
good
great
work,
guys.
B
C
I
did
want
to
say
one
little
thing
about
the
two
designs
she
showed
both
are
lovely
and,
in
fact,
I'm
kind
of
leaning
to
the
first
one,
just
because
it
looks
a
bit
more
snazzy,
but
the
one
thing
I
did
want
to
say
is
relating
to
James's
previous
point
about
the
filter.
Looking
almost
like
a
preset
when
you
do
add
that
button
in
between
it
makes
it
even
look,
it
looks
makes
it
look
even
more
like
it's
a
preset
value
from
a
you
know.
C
This
is
right
next
to
the
title,
which
is
a
preset
and
it's
looking
in
a
similar
vein
visually
in
a
similar
vein.
So
yeah,
that's
the
only
thing
I'd
say
regarding
that
is.
We
may
want
to
make
it
even
as
James
said
to
clearer
that
this
is
for
just
the
current
iteration
that
you're
looking
at
it's
not
being
saved
somehow
don't
know
how.
B
Yeah
I
think
the
the
end
Vision
will
definitely
solve
or
help
with
that
user
experience,
because
I
think
the
end
vision
is
to
have
no
filters
and
that
you
add
filters
to
this
kind
of
as
like
a
tokenized
search
that
you
have
a
button
here
that
says,
add
filter
that
in
itself
could
actually
also
be
confusing,
because
now
it
could
be
seen
as
like.
B
Oh
you're,
adding
a
static
filter
and,
like
you,
won't
be
able
to
to
edit
again,
but
definitely
yeah
some
copy
or
something
I'm,
also
fine
with
hiding
it
for
the
first
iteration
that
we
take
to
our
customers.
Maybe
we
can
get
some
feedback
from
our
internal
users
once
we
get.
This
live
like
out
there
on
like
gitlab.com,
and
we
can
just
like
propose
that
question
and
say
like.
Does
this
make
sense
to
you.
A
C
A
Yeah,
that
seems
fair
enough.
Leave
it
or
hide
it.
I'm
good,
either
way
and
we'll
get
some
feedback
from
users
thinking
about
at
least
for
our
internal
product
management,
Persona
and
developer
persona.
The
tool
that
they're
used
to
is
one
of
the
tools
they're
used
to
could
be
cisens.
It
behaves
the
exact
same
way
where
you
can't
you
can
change
the
filter,
that's
on
the
top
of
our
states
and
things
go
and
it
saves
it.
But
it's
not
in
the
edit
experience
it
more.
A
live
edit
of
I'm.
A
Changing
the
panels
on
the
dashboard
or
I'm
changing
adding
a
new
panel
to
a
dashboard
I
can
change
the
filter
there
and
that
persists,
but
that
persists
whether
you're
editing
a
dashboard
or
not.
So
it's
not
something
that
you
State
per
se,
so
it
kind
of
mirrors
that
experience,
so
it
shouldn't
be
like
it's
not
totally
foreign
or
totally
unknown
to
what
a
user
might
be
used
to.
B
A
Yes,
I
think
we've
done
the
agenda
then
as
well
any
other
ux
orientation
questions
we
can
go
through.
Overall,
here's
synchronously
I.
C
Guess
I
I
think
I
feel
like
I
want
to
like
voice
make
sure
we're
with
an
agreement
is
obviously
we
are
currently
looking
to
fill
the
design
position
within
our
team.
C
So
as
far
as
I'm
aware
and
how
I've
been
viewed
this
or
me
and
yarn
have
gone
back
to
just
covering
the
design
position,
we're
not
looking
to
create
a
whole
new
mpz
designs
or
create
new
entire
new
pages.
It's
just
about
taking
what
we've
already
got
and
just
iterating
towards
where
we
want
to
be
in
the
smallest
possible
steps,
rather
than
like
doing
big
full-blown
Vision
stuff
like
we
were
doing
originally.
A
Yes,
small
baby
steps,
iteration
the
only
thing
where
I
might
ask
for
you
to
to
do
some
design.
If
we
get
that
far
without
a
designer
is
going
to
be
that
new
usage
quota
page,
it
is
modeled
a
hundred
percent
after
the
existing
pipelines
usage
quota
page
and
that's
what
I
hacked
together
a
buyer
brand
based
on.
So
there's
not
a
lot
of
design
work
there.
It's
just
reusing
existing
components.
Unplugging
in
our
data
versus
pipeline
data.
C
A
Yeah
I
think
that
we'll
have
a
designer
by
then
anyway,
but
that's
the
only
new
page
running
through
issues
in
my
head.
That's
the
only
new
page
I
can
think
of
that.
We'll
create
they're
going
to
need.
You
know
all
new
design
or
new
implementation,
otherwise
we're
working
on
existing
Pages
iterating.
A
This
is
the
big
one
and
we're
going
to
wait
for
new
designer
to
be
in
place
before
we
start
working
on
a
new
vision.
First
visualization
designer.