►
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
Okay,
recording
cool,
so
yeah
I
wanted
to
get
this
group
together.
Eli
you
and
I
actually
haven't
met,
but
it
was
my
understand
that
you're
working
on
improved
usage,
metrics
and
I
want
to
make
sure
I
was
disambiguating.
For
my
team.
We
tend
to
use
the
word
usage
being
exclusively
to
talk
about
usage
metrics,
but
there's
some
disambiguation.
A
B
A
So,
let's
start
with
this
disambiguation,
because
we
even
had
this
a
little
bit
on
the
product
call
I
think
today
yeah
that
we
there's
a
current
set
of
data
that
we
don't
get
from
both
get
lab
comm
and
self
managed
instances
called
usage
Bing.
It's
pretty
dumb
single
increment
counter
as
I
understand
it.
That
just
adds
a
counter
to
a
table
and
then
gets
shipped
to
us
once
monthly.
But
we
all
recognize
that
that
is
not
sufficient.
A
A
Let's
look
at
the
types
of
metrics
we
get
from
comm
and
see
if
we
can
reconfigure
what
we
want
the
Smout
to
be
for
each
one
of
these
stages,
so
that
when
we
get
access
to
those
improve
metrics
through
whether
it's
an
India
or
pendo,
or
some
other
instance
on
self-managed,
that
we
can
have
this
new
definition
of
smell
available
in
working
tonight.
Capture
that
correctly
Eli
yeah.
B
B
So
yeah,
that's
I!
Think
it's
a
bit
unclear
how
well
this
will
scale
to
so
post.
It
I
mean
snow
cloud,
wouldn't
be
a
problem
on
the
front
end
like
it
should
be
the
exact
same
either
way,
but
right
now
for
Caleb
calm
we're
just
like
each
yelling,
the
database
that
we're
hosting
and
putting
that
into
our
warehouse.
We
obviously
can't
do
that
for
self
host
the
businesses,
so
much
circles
can
be
part
of
the
usage
Bing
still,
but
we
have
to
make
queries
and
send
it
in
combination
with
snowplow
it.
B
That
part
seems
up
in
the
air,
but
is
tricky
from
my
perspective,
like
on
the
data
team.
We're
not
really
responsible
for
data
collection
like
her
are
sort
of
handoff
point
is:
maybe
it
a
bit
further
downstream.
Then
that
would
be
another
company,
so
I'm
not
exactly
sure
what
the
status
of
data
collection
is,
but
I
think
it's
some
sort
of
combination
of
snowplow
and
usage
ping
still,
okay,.
A
B
I
think
usage
thing:
yeah,
I
wouldn't
use
those
terms
because
usage
pain
still
might
encompass
front-end.
Somehow,
where
like
they
would
compile
off
the
front-end
events
and
then
send
them
out
from
the
instance
with
usage,
pain,
that'll,
be
one
design
pattern,
so
I
think
maybe
Fernan
back-end
is
keep
those
separate.
Okay,.
C
Iiii
was
just
going
to
add
there
that
we
measure
a
lot
of
the
front
end
stuff
with
the
the
usage
thing,
and
we
did
that
for
a
long
time,
because
it
was
the
only
mechanism
that
we
had
so,
for
example,
when
you
install
a
kubernetes
application,
the
only
way
to
do
that
is
through
the
GUI
right.
That's
all
a
front-end
process
right
that
you
go
on
the
page
and
you
have
to
click
the
button
and
stuff
like
that.
A
B
B
A
D
It
seems
like
there's
potential
duplicates
sometimes
because,
if
I
choose
to
create
a
project
through
the
web
interface,
I
have
a
front-end
tracker
and
a
back-end
tracker
for
create
project.
But
then,
if
I
do
it
through
the
API
there's
only
a
back-end
tracker.
So
if
you
take
the
number
of
back
in
actions
minus
the
number
of
front-end
track
actions,
you
can
figure
out
how
many
people
did
the
action
through
a
non
GUI
interface
versus
the
GUI
interface.
Whereas
the
back
end
is
a
completed
request.
D
A
And
sorry
a
word
eight
minutes
and
I
want
to
spend
the
whole
time
to
pay
him.
So
maybe
you
I,
if
there's
some
rough,
you
said
the
words
front
and
back
in
and
so
I'm
attaching
to
them.
If
I'm
over
rotating
them,
we
can
move
on
and
if
there's
some
reference
to
how
the
data
team
thinks
about
those
two
terms,
I
would
love
it
okay.
A
So,
but
it
sounds
like
in
general,
we
are
working
towards
a
goal
where
we
have
more
robust
of
front-end
and
back-end
data
and
there's
an
effort
you're
doing
with
manage
to
kind
of
like
precede
what
us,
having
more
robust
data
from
soft
manage
instances
by
using
data
sources
available
on
comm.
Oh,
yes,
I
was
directed
that
this
would
be
a
good
exercise
for
my
team
to
do
with
you.
You
can
also
respond
and
be
like
actually
I'm
overloaded,
there's
enough
with
manage.
Let's
let
that
play
out
before
we
start
talking
about
the
other.
A
B
No
I
think
it's
good
exercise
that
thinks,
like
the
this
tricky
part,
and
this
is
a
big
project
to
do
like
a
dashboard
for
every
single
stage
and
I.
Think
like
the
hardest
part
for
me,
is
like
being
new
to
this
company
and
like
the
product
and
there's
this
huge
gap
between
product
and
data
and
like
product
to
find
what
they
think
Mau
is,
and
then
it's
like
this
handover
process.
B
The
data
where
data
has
to
build
the
dashboard
and
there's
just
a
ton
of
domain
knowledge
in
that
gap
where
it
was
like
I'm,
not
really
sure
what
they
mean
by
this
and
then,
instead
of
just
one
stakeholder,
there's
like
you
know,
eight
or
ten
different
stakeholders
for
each
stage.
So
it
was
like
tricky
for
me
to
figure
out
like
what
what
do
they
mean
when
they
say
this,
and
this
of
this
so
I
think
just
having
like
a
face
to
put
to
each
stage.
B
B
I'm,
trying
to
you
know,
focus
on
the
growth
stuff
when
possible,
and
we've
been
seeing
if
the
data
team
could
get
more
support
for
like
a
proactive
analyst
would
be
able
did
want
to
fill
eventually,
which
would
be
separate
from
growth.
But
in
the
meantime,
growth
is
stepping
in
to
help
with
these
I
think.
B
They
probably
spelled
out
a
little
bit
just
because,
like
I'm
saying
like
it
was
hard
for
me
to
be
actionable
on
them,
so
I
was
kind
of
just
working
on
the
manage
one
first,
which
was
the
plan,
but
I
think
now
that
the
managed
one
is
wrapped
up.
I
shared
it
in
the
same
chapter,
if
you
have
a
periscope
access,
but
now
that
that
one's
wrapped
up
I
think
we're
going
to
start
planning
to
fill
out
the
other
ones.
A
And
I'll
see
like,
if
you
call
it
increased
urgency
on
my
part,
but
we
are
I'll
speak
for
the
team.
I
feel
like
we're
flying
a
little
blind
when
it
comes
to
usage
data,
and
so
it
would
be
great
to
have
additional
insight
into
actual
usage
of
our
product.
I
would
I
would
argue
that
secure
is
the
place
where
I
would
wants
to
know
the
most,
because
it
seems
like
the
place
where
we
have
the
kind
of
hardest
questions
to
answer
that
are
obviously
heavily
revenue
impacting,
but
then
I
also
know
that
secure.
A
We
don't
have
as
many
secure
users
on.com
so
that
might
might
not
you
one.
We
want
to
prioritize
I
would
prioritize
it
as
secure
configure
monitor
if
you're,
considering
taking
them
one
at
a
time
to
be
the
process
that
is
helpful
and
then
yeah.
If
it's
more
appropriate
for
us
to
have.
You
then
follow
up
individually
with
assigned
PMS
on
those
issues
that
works
for
me.
A
A
E
Yeah
I
think
that
would
be
helpful.
Okay
doesn't
have
to
be
I
mean
the
dashboard
looks
fairly
self-explanatory,
but
it
would
be
helpful
to
know
we
walk
through
how
you
guys
put
together
the
queries.
Maybe
we
can
do
some
of
that
legwork
up
front
and
reduce
the
load
on
you.
I
have
some
sequel
background.
I,
don't
know
about
Danielle,
Danielle
and
Nicole,
but
whatever
we
can
do
to
help
you.
Let
us
know
yeah.
A
B
B
B
B
So
I
ended
up
with
sorry.
I
have
too
many
and
she's
open,
so
I
turned
it
to
a
table
and
they're
just
trying
to
grow.
It
basically
fall
into
three
different
areas
captured
on
the
back
end,
captured
in
snowplow
and
then
not
captured
at
all,
and
that's
kind
of
I
think
what
would
cause
the
most
delays
and
then
they
made
the
discussion.
Look
back
end
versus
friend,
I.
Think
I
can
explain
that
better,
but
if
it,
if
it's
in
yeah,
there's
overlap
there,
but
I
can
figure
out
which
is
the
best
source
there.
B
B
A
E
E
Options
in
that
list,
on
the
left
hand,
side
as
an
event
come
up
differently.
If
you
click
on
your
groups,
vs.
explore
groups
cuz.
Those
to
me
are
going
to
be
kind
of
the
same
quote-unquote
section
within
that
page
structure,
and
the
answer
might
be
yes
and
I
might
be
digging
us
in
too
deep
I'm
just
curious
if
we
are
tracking
this
as
unique
events.
B
B
B
B
I
can
confirm
later
on
yeah
what
I
was
saying
about
user
IDs
right
now.
It's
just
this
one
which
is
coming
from
snow
cloud
and
it's
based
off
of
cookies.
So
if
I,
clear
out
my
cookies
or
if
I
switch
browsers,
it's
not
gonna
know
that
I'm
the
same
user.
So
that's
kind
of
a
blocker
for
this
right
now,
but
there's
a
ongoing
issue
and
debate
about
that.
Whether
we
should
be
turning
these
already
on
there
and
then
we
can
actually
because
this
way
would
be
counting.
B
D
A
Okay,
so
you
went
through
the
process
of
kind
of
how
you
created
those
dashboards
in
the
couple
minutes
we
have
left
sounds
like
Sarah
was
at
least
interested
in
knowing
the
basics
of
when
I'm
looking
at
periscope.
How
do
I
select
and
then
drill
down
into
a
snowplow
datasource.
You
maybe
should
show
us
that
real
quick
so
in
those
managed
dashboard
that
you
showed
there
were
some
audit
events
or
other
things
that
were
based
on
snowplow
data.
B
A
That's
something
that
Sarah
could
self-service
to
say
listed
it
injured
in
this
specific
event
and
maybe
see
sure
the
basics
of
getting
started
there.
Yep.
B
B
B
Everyone,
a
periscope,
should
have
full
permissions
here
and
then
it's
gonna
have
like
100
different
columns
and
I
could
open
up
permissions
on
what
I
was
doing
with
manage,
but
basically
every
documentation
here
as
well
in
DBT
Doc's.
So
it's
going
to
give
these
information
about
like
what
each
column
XP
is
but
yeah
we're
gonna.
Have
you
know
every
column
we've
gonna
need
here.
The
scrolling
is
not
working,
but.
A
B
D
B
B
A
Just
too
close
for
the
team,
I
think
I
guess
what
I'm
hearing
is.
There
is
an
effort
to
evaluate
other
data
sources.
It
may
or
may
not
be
snow
plow,
but
it's
I
think
valuable
for
us
as
a
product
team
to
start
getting
a
sense
of
the
data
that
we
get
from
calm
because
hopefully
we're
going
to
have
data
like
that
from
self-managed
instances
in
the
future.
It's
the
extent
that
we
can
kind
of
pre
refine.
A
What
we
think
is
important
for
smile
and
what
you
think
is
important
as
a
product
manager
for
knowing
how
your
product
usage
is
moving
forward.
We
can
work
with
UI
to
create
those
dashboards.
I
know
Daniel,
especially
having
been
here
for
a
while
we've
kind
of
consistently
said
hey.
If
we
can't
get
the
data
for
self-manage,
it's
really
not
valuable.
A
I,
don't
disagree
that
in
terms
of
making
broad-based
product
decisions,
but
in
terms
of
getting
a
sense
of
the
types
of
data
we
could
get
is
what
I
think
of
this
exercise
for
is
kind
of
like
proceeding
for
when
we
have
data
on
self
management
says
unnecessary
that
you
would
use
both
emotions
I.
Think
everybody
that
makes
it
cool
I
think
we're
over
time
again.
You
know
I
appreciate
it.
I
know
we
can
for
you
in
here,
but
it
was
super
helpful
for
me
and
then
yeah
if
you're,
if
you.