►
From YouTube: UX Showcase: Thinking Big Shipping Small With Metrics
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Hi
everyone,
I'm
nadia,
I'm
a
product
designer
at
monitor,
apm
and
together
with
amelia
from
monitor
health
I'll,
be
sharing
some
updates.
What
we're
working
on
at
monitor
ux.
A
So
I
will
talk
about
our
new
feature
in
metrics
for
adding
new
panels
and
how
we're
trying
to
think
big,
while
also
shipping,
small
and
providing
very
quick
value
to
our
users.
A
So
just
for
a
bit
of
background
on
metrics
metrics
dashboards
are
dashboards,
that
site
reliability,
engineers,
devops
engineers
and
also
just
developers,
used
to
monitor
the
health
and
performance
of
either
their
application,
or
also
the
infrastructure
that
the
application
is
running
on.
So
here
in
the
screenshot,
you
see
a
metrics
dashboard.
A
A
Users
don't
have
access
to
the
files
of
the
out
of
the
box
dashboards,
which
leads
to
a
bit
of
complexity
where
some
of
our
dashboards
have
one
set
of
features
and
other
dashboards
don't
have
the
same
features
just
because
they're
different
in
implementation,
which
is
a
constraint.
So
it's
not
really
the
desired
user
experience
that
we
want
to
provide.
A
So
some
of
the
dashboards,
you
can
only
add
a
metric
to
a
dashboard
for
other
dashboards.
You
can
edit
the
dashboard
fully
in
a
file
where
you
can
define
all
the
different
properties.
A
Each
dashboard
contains
building
blocks
called
panels.
So
here
you
see
this
is
a
panel.
This
is
a
panel
and
each
panel
can
contain
different
types
of
data
visualizations,
so
it
can
be
a
chart
or
it
can
also
be
not
data
visualization,
but
just
a
text
block.
For
instance,
you
can
also
add
a
text
blocks
in
your
dashboard
if
you
want
to
add
any
instructions
or
any
description
for
this
particular
dashboard.
A
So,
as
I
mentioned,
the
features
are
different
between
different
dashboards,
whether
you
can
edit
the
dashboard
or
add
metrics
to
it.
It
depends
on
what
dashboard
you're
looking
at,
which
is
not
ideal,
and
furthermore,
you
can
only
add
certain
types
of
panels,
so
you
can
add
a
metric
to
the
dashboard,
but
you
can't
add
a
text
panel,
for
instance.
So
this
is
a
limitation
that
we're
trying
to
solve
by
introducing
a
feature
to
add
a
new
panel
to
your
dashboard.
A
Also,
you
will
be
able
to
edit
the
panel
delete
the
panel
and
so
on.
So
this
is
our
attempt
to
kind
of
bring
this
experience
to
the
common
denominator
and
also
make
sure
that
users
can
set
up
their
panels
fully,
not
just
add
a
metric
but
add
multiple
metrics
set
alerts
on
those
metrics
all
within
one
view.
Basically,
so
what?
What
does
it
mean
to
add
a
panel?
It
means
a
lot
of
things
actually.
So
there
are
many
different
things
that
we
need
to
work
on
and
consider.
A
For
example,
when
adding
a
panel
users
may
want
to
set
up
alerts
on
those
metrics,
they
may
want
to
add
multiple
metrics
to
a
panel.
They
will
need
to
tweak
different
panel
settings
and
then,
depending
on
the
panel
type,
they
will
have
to
tweak
settings
for
the
chart
like
a
line,
chart
area
chart
or
a
single
stat,
visualization
and
so
on.
So
those
settings
will
be
conditional
based
on
the
panel
type.
A
So
it's
a
lot
to
handle
it's
a
very
big
workflow
that
we're
looking
at
and
currently
at
metrics
we're
really
trying
to
focus
on
some
of
the
quick
improvements
to
the
user
experience
because
we're
on
the
way
to
dog
food
metrics
with
our
infrastructure
team,
so
approaching
something
like
this.
That
will
take
many
many
many
milestones
to
build.
It's
not
easy!
So
how
do
we
balance
the
long-term
vision
with
the
nbc
approach?
A
Because
we
don't
want
to
end
up
with
the
same
kind
of
disjointed
experience
that
you
know
we
came
to
this
point
by
implementing
lots
of
small
features
very
quickly,
so
we
want
to
consider
the
big
picture,
but
we
also
want
to
make
sure
that
we
can
ship
something
very
quickly
because
not
being
able
to
add
a
panel
to
your
dashboard,
it's
a
big
obstacle.
One
of
the
biggest
obstacles
is
that
users
can't
preview
panels.
A
So
you
have
no
idea
what
you're
adding
to
your
dashboard
you're
just
writing
code
and
you
don't
you
don't
know
what
it's
going
to
look
like
until
it's
merged
and
boom
it's
in
the
dashboard.
So
how
do
we
approach
something
big
like
this?
We
decided
that,
as
the
first
step,
we
will
ensure
that
ux
is
collaborating
with
engineering
very
closely,
so
we
scheduled
a
technical
spike
and
also
design
exploration.
So
I
focused
on
questions
like
what
is
the
ultimately
desired
experience.
A
What
are
the
of
all
of
the
open
questions
that
I
have
around
this
feature?
What
dots
don't
understand,
what
questions
do
I
have?
What
might
it
look
like
in
our
wildest
dreams
and
the
engineer
miguel,
an
engineer
from
the
apm
team?
He
focused
on
questions
like
okay,
so
what
is
possible
today?
What
are
the
constraints
that
we
have
today?
How
much
effort
will
the
desired
user
experience
take
to
build
what
is
possible
within
the
next
one
milestone
or
two
milestones
and
so
on?
A
So
at
the
same
time,
I
started
working
on
some
visionary
wireframes,
just
trying
to
imagine
what
it
could
look
like
what
this,
what
this
workflow
requires.
What
are
the
different
features
that
need
to
be
part
of
this
and
so
on.
So
there
are
so
many
open
questions
right
now
around
this
experience
that
really
it's
just
it's
it's
very
visionary.
A
A
However,
it
turned
out
that,
even
though
I
did
my
best
to
strip
the
experience
down
to
the
bare
basics,
adding
the
panel
to
the
dashboard
will
require
a
lot
of
back-end
effort
and
right
now,
apm
has
limited
back-end
capacity,
so
turned
out
that
my
mvc
is
not
mvc,
which
I'm
sure
happens
quite
often
turned
out
that
it
will
take
at
least
three
milestones,
and
if
it
takes
three
milestones,
it's
not
an
npc.
A
So
I
decided
that
okay,
we'll
call
this
midterm
vision,
and
I
have
I
had
to
have
a
another
sync
meeting
with
miguel
and
go
over
all
of
the
details.
Make
sure
that
I
understand
all
the
constraints
and
together,
basically,
we
negotiated
a
more
nbc
nbc.
So
we
decided
that
okay,
we're
not
going
to
add
the
new
panel
to
the
dashboard,
but
we're
going
to
allow
the
user
to
preview
the
panel
and
then
copy
the
code
and
make
it
easier
to
go
to
their
dashboard
file
and
paste
in
there.
A
So
of
course,
it's
not
ideal,
but
we're
shipping
something
very
important
here,
which
is
a
preview.
So
now
you
can
see
what
your
panel
will
look
like.
You
can
tweak
all
the
different
panel
properties
and
then
once
you're
happy
with
the
results
you
can
edit
to
your
dashboard,
which
really
reduces
the
barrier
to
entry
for
users
who
are
not
really
comfortable
working
with
yml
files.
A
So
once
we
did
that
there
was
like
one
week
until
the
end
of
the
milestone
not
much
time
to
run
proper
solution,
validation
effort,
because
we
really
wanted
to
start
implementing
some
type
of
mvc
in
this
milestone.
A
So
I
conducted
unmoderated
internal
usability
testing
with
our
own
infra
team,
who
gave
us
really
good
feedback,
and
I
was
able
to
further
improve
the
the
prototype.
So
now
at
least,
we
know
that
the
mvc
will
work
and
then
we
can
further
learn
from
the
feedback
that
we
will
get
from
once
it's
actually
built.
A
So
the
next
steps
would
be
to
implement
this
mvc
solution
in
13.3,
and
then
there
are
multiple
follow-up
issues
I
linked
to
the
epic
in
the
agenda,
where
we
will
be
doing
more
discovery
and
more
solution,
validation
around
the
midterm
vision,
the
long
term
vision
and
so
on.
So
this
is
very
exciting
because
it
we're
starting
to
allow
our
users
to
do
things
within
the
ui
and
not
just
within
the
actual
yaml
files.
A
If
you
have
any
questions,
I'm
happy
to
answer
them.
Just
ping
me
on
slack
or
yeah
feel
free
to
schedule
a
coffee
chat
and
now
I
will
hand
it
over
to
amelia
who
will
tell
you
more
about
her
work
on
incident
management.