►
From YouTube: 2020 08 26 DataTeamDemos
Description
Data Team demos towards FY21-Q3 Great Team - Self-Service Data.
https://docs.google.com/document/d/17Xjk5pS7MfARWEKAKThhLlCUnt-G3Q1aGVMWeTYwWT8/edit?usp=sharing
A
B
We,
starting
with
the
handbook
page
that
we
put
together,
as
we
mentioned,
we
talking
through
a
solution
for
product
geolocation
analysis
for
our
self-managed
customers
and
the
high
level
context
for
this
is
that
understanding
understanding
where
our
product
is
used
around
the
world
is
an
important
step
towards
developing
more
complete
understanding
of
our
customers.
What
our
reaches
and
related
location
insights.
B
A
lot
of
this
is
derived
from
usage
ping,
which
is
set
at
the
instance
level,
and
it's
a
way
that
customers
can
send
us
payloads
of
summary
statistics
back
on
their
usage
data
and
this
instance
level
data
will
allow
us
to
understand
country
level,
statistics
and
trends,
and
instance,
adoption
version,
adoption
rate
and
details
around
our
instance
life
cycle,
as
mentioned
in
the
agenda
doc,
we're
excited
to
start
building
out
knowledge,
assessments
and
certifications
for
both
the
dashboard
solution
and
the
underlying
sql
data
models.
We've
developed
for
self-service.
B
The
telemetry
team
introduced
a
number
of
improvements
to
usage
paying
around
this
state,
and
so
this
is
kind
of
the
first
recommended
date
where
we
see
stable,
reporting
around
usage
data
metrics.
So
it's
a
good
thing
to
keep
in
mind
for
the
rest
of
what
is
presented
in
this
dashboard
and
for
each
of
these
high-level
metrics,
we've
included
little
blurbs
around.
What
exactly
do
we
mean
by
something
like
paid
instances,
and
this
will
help
set
some
of
the
context
and
reduce
any
confusion
around
what
is
explicitly
being
shown
here.
B
Curiously
enough,
we
do
have
some
pings
from
antarctica,
there's
a
few
other
interesting
tidbits
in
there,
but
we
do
see
in
this
time
period
a
pretty
stable
trend.
Most
of
our
usage
data
is
coming
from
china
and
the
us
and
germany,
and
so
one
interesting
thing
of
note
here,
is
that
this
is
actually
reporting
based
on
the
host
of
the
instance.
B
It's
not
necessarily
the
physical
location
of
our
customers
and
so
there's
a
lot
of
interesting
detail
that
we'd
like
to
dig
in
further
to
here
to
understand
the
underlying
dynamics
within
that,
but
even
at
this
level
we
can
kind
of
see.
You
know
we
have
the
most
growth
across
china,
russia,
south
korea,
regions
like
france
and
the
netherlands
are
actually
contracting
a
little
bit
and
so
we'd
even
like
to
go
even
more
granular
than
country
to
see
where
adoption
dynamics
may
exist.
B
B
So
there's
a
really
great
opportunity
here
to
to
communicate
the
value
of
of
what
git
lab
can
understand
from
usage
ping
and
try
to
encourage
more
of
our
user
base
to
stay
opted
into
this.
We
know
there's
there's
complications
with
ultimate
customers.
Many
of
them
are
air-gapped,
but
this
is
definitely
an
interesting
area
for
improvement.
B
B
And
one
final
thought
on
that
same
topic
is
that
there
is
a
lot
of
variance
in
the
versions
that
are
reported
among
our
instance
base.
Even
if
we
look
at
the
top
10
versions
based
on
the
number
of
instances
reporting,
this
only
accounts
for
about
60
of
the
usage
data
that
we
see
so
there's
a
lot
of
diversity
in
the
usage
data
that's
available
to
us,
based
on
just
the
diversity
in
our
customer
base.
B
So
the
last
thing
that
I'll
touch
on
here
is,
if
you're
interested
in
producing
your
own
analyses
and
you're
familiar
with
with
these
data
models
and
have
worked
with
them
in
the
past.
This
hopefully
looks
like
a
pretty
significant
improvement.
We're
really
excited
at
how
simple
this
final
data
model
has
turned
out
to
be.
It's
a
lot
quicker
to
iterate
through
analyses
with
this
new
solution
that
we've
come
up
with,
with
that
I'll
jump
back
over
to
the
handbook
page
real
quick.
A
Very
nice
derek
I
caught
a
couple
at
least
four
insights
for
me
about
how
gitlab
is
being
used,
that
I
wasn't
aware
of
so
thank
you
and
your
team
for
that.
How
do
people
contribute
now
that
you've
shared
this
demo?
How
how
is
the
best
way
for
people
to
provide
feedback
to
what
they
saw
and
start
to
contribute
toward
this?
The
self-service
solutions.
B
B
A
C
I'll
get
into
it.
Let
me
share
my
screen
all
right,
so
the
echo
rob
and
what
derek
shared
the
idea
is
to
create
self-service
areas
around
subject:
sub
subject
areas
within
gitlab
analytics
areas,
so
the
area
the
lead
to
casting
focused
on
was
customer
segmentation
analysis
and
the
idea
is
that
our
gitlab
team
members
will
be
able
to
land
on
this
page
and
largely
be
able
to
self-service
themselves
without
the
need
for
immediate
questions.
C
Sure
questions
can
can
be
asked,
but
the
idea
is
that
this
would
give
you
a
lot
of
information
to
train
and
to
train
you
and
to
get
you
up
to
speed
on
viewing
our
customers.
So
with
that
context,
we
can
start
to
get
into
it.
So
customer
segmentation,
the
lens
we've
taken,
is
that
we're
looking
at
getting
to
know
our
customers
better,
getting
to
know
their
characteristics,
their
different
firmographic
attributes
and
to
see
how
our
book
of
business
what
they
look
like.
C
So
with
this
page,
what
you'll
be
able
to
do
coming
soon,
we'll
have
knowledge,
assessment
and
certifications
that
our
team
members
be
able
to
take
to
learn
how
to
use
the
page
and
the
the
different
tools
included.
We
have
a
solution-
ownership
section
so
these
folks
here
whenever
there's
issues
in
the
source
systems
or
people
need
access.
These
are
some
of
the
first
people
we
can
go
to
and
maybe
then
they
can
delegate
out
to
resolve
any
issues.
So
jack
brennan
for
salesforce,
andrew
murray,
he's
a
director
of
revenue,
jack's
director
of
sales
systems.
C
So
here's
some
key
terms
here
and
we
link
to
the
handbook.
So,
as
you
all
play
with
this
a
little
bit
when
you
link
here
and
you
click
on
it,
let
us
know
if
you,
if,
if
it's
clear
the
definition,
so
the
idea
would
be
somebody
could
come
to
this
page,
they
could
sale
segments
included
in
this
analysis
they
can
click
here
and
then
they
know
what
sales
segment
is.
C
So,
if
that's
not
the
case,
then
that's
an
area
for
improvement,
and
that
could
be
some
good
feedback
same
with
the
key
metrics
that
are
included
in
here.
So
with
the
with
the
customer
segmentation,
we're
taking
a
view
of
the
products
that
our
customer
have
the
delivery,
which
is
self-managed
and
sas
the
sales
segment,
the
account
owner
team,
the
the
territory
and
the
industry
so
we'll
be
able
to
slice
and
dice
our
two
of
our
big
key
metrics,
arr
and
customer
account.
C
So
this
solution
is
currently
configured
to
to
give
insights
into
those
into
this
version,
and
we
have.
We
have
many
more
ideas
of
additional
questions
that
could
be
answered.
That
can
be
some
other
feedback,
as
you
all
review
this
like
first,
can
you
get
insights
into
these
attributes
and
to
any
additional
things
you'd
like
to
know
about
our
customers
so
that,
as
you
kind
of
think
about
it,
maybe
think
about
it
from
that
lens
too?
C
So
I'll
get
you
oriented
to
the
rest
of
the
page
and
then
we'll
maybe
see
if
we
have
some
time
to
dive
into
some
stuff.
So,
within
our
solution
we
want
to
be
able
to
have
a
dashboard
that
somebody
can
go
in
and
just
use
filters
and
get
answers.
We
want
to
have
a
solution
where
somebody
can
develop
a
dashboard,
so
we're
looking
at
getting
some
different
data
spaces
within
within
sisense.
C
So
these
would
be
two
of
our
users,
one
user,
that
you
know
I
just
need
to
maybe
your
executive
user.
I
just
need
to
go
to
a
dashboard
and
just
get
the
information
I
need.
I
don't
have
time
to
be
building
stuff.
You
know
that's
what
we
have
a
big
team
for.
Maybe
then
there's
another
person
that
person
that
can
go
build
stuff.
You
know
dashboard
developers,
so
you
want
to
empower
that
person
that
can
get
a
little
more
technical,
a
sql
developer.
C
This
is
someone
that
can
have
some
level
of
sql
knowledge
and
we
want
to
give
them
tools
for
that.
So
here
we
have.
We
have
the
dashboard
for
our
viewer
and
then
for
our
sql
user.
We
have
a
erd
and
this
is
the,
but
for
a
sql
user.
This
will
be
super
helpful.
So
these
are
the
data
objects
that
roll
up
in
our
enterprise
dimensional
model
to
create
the
arr
data
mart,
which
is
what
drives
our
customer
segmentation
analysis.
C
So
a
sql
person
will
be
able
to
come
in
here
and
look
at
it
and
say:
okay,
here's
how
I
can
join
all
these
things
together.
So
that's
the
idea
of
of
the
erd.
We
have
a.
We
have
reference
sql,
so
we
have
a
mark
level.
We
have
a
mart
layer
similar
to
the
product.
Geolocation
that
derek
shared.
We
have
a.
We
have
a
mart
level
table
that'll,
be
really
easy
for
kind
of
your
sql
1010101
user.
Where
select,
you
know
to
select
some
columns
from
this
table
very
basic,
very
easy
to
use.
C
We
also,
then,
this
erds,
maybe
somebody
some
more
advanced
sql
users.
They
can
come
and
play
with
these
objects.
We
even
have
some
ideas
to
make
these
objects
easier
to
work
with,
but
so,
depending
on
your
level
of
sql,
you
have
some
different
options
and
in
future
iterations
we'll
give
some
more
reference
sql
on
how
to
play
with
those
dimensions
more
similar
to
the
product.
Geolocation.
C
There's
a
data
lineage
chart
and
you
can
see
if
you
wanted
to
see
kind
of
what
all
the
objects
are,
that
roll
up
to
the
that
roll
up
to
the.
I
guess
it's
loading
here
that
build
that
model,
we'll
let
that
load
kind
of
similar
here
and
you
can
click
in
here
and
you
can
see
kind
of
what
what
what
objects
build
each
of
these
objects
so
that
that's
kind
of
cool
you
can
play
with
that
and
then
our
trusted
data
solution.
C
So
we
want
people
to
have
confidence
in
our
data
and
this
section
here
that
will
continue
to
be
developed.
Does
this
where
we
want
to
implement
tests
to
ensure
that
the
data
coming
into
our
warehouse
is
complete
and
accurate
and
that
the
quality
of
the
models
that
are
built
are
also
accurate?
So
this
is
what
this
dashboard
does
here.
C
C
C
So
let
me
orient
you
to
the
dashboard
one
there's
a
bunch
of
filters
here
that
you
can
filter
for
the
firm
of
firmographic
attributes,
so
you
can
filter
for
your
account
owner
team
industry,
your
territory,
delivery,
sas,
self-managed
your
products
and
your
product
tier
sales
segment
and
whenever
you
filter
by
one
of
these
every
chart
on
the
dashboard
will
also
filter
by
one
of
those.
So
the
idea
is,
if
you
click
us
east
here,
it
would
give
you
it
will
show
you
all.
The
charts
based
on
the
slicer
of
us
east
account
owner
team.
C
So
that
is
good.
So
that's
one
piece
and
then
down
the
side.
You
can
see,
we've
organized
the
dashboard
into
sections
to
kind
of
call
out
what's
being
done
in
this
in
this
dashboard,
so
you
have
product
attributes,
sales
segment,
charts,
delivery,
charts
account
owner
team
territory
and
industry
and
then
also
some
a
few
account
level
details
and
we
love
your
feedback
on
which
one
of
these
charts
pop,
which
one
of
them
kind
of
gives
you
some
some
insights,
which
one
of
them
answers
questions
or,
as
you
look
at
it.
C
What
questions
are
being
answered
in
your
mind
and
we
want
it
to
be
largely
self-service
with
with,
but
somebody
could
just
come
to
this
chart
and
then
it
would
they
would
get
information
out
of
it.
So
that's
what
I
would
love
for
you
all
to
do
is
just
to
go
play
with
this
dashboard
and
and
and
see
if
you
can
get
some
insight.
So
let's
do
an
example.
So
let's
say
I
am
I'll.
Take
a
sales
example.
C
Let's
say
I
wanted
to
know:
what's:
what's
our
smb
business
look
like
well
before
I
do
that.
So
when
you
land
on
the
dashboard,
what
is
showing
you
now
is
for
total
company.
So
if
you
view
any
of
these
charts,
take
care
take
product
category
for
the
prior
month.
So
prior
month
we
had
exit
arr
of
130.3,
and
then
this
shows
you
the
mix
of
the
product
category
for
the
the
total
month.
You
know
probably
not
shocking
30
of
it
is
premium
coming
in
at
premium.
We
have
a
other
category
there.
C
C
Then
I'll
come
in
here
for
a
landing
here
in
a
couple
of
minutes,
there's
a
lot
of
charts,
so
maybe
it'll
take
like
maybe
15-20
seconds
to
refresh
and
while
it's
refreshing
so
now,
this
has
given
us
the
view
for
s
and
b.
So
now
we
can
see
the
product
mix
for
s
and
b
this
chart's
really
cool
you
can.
This
is
a
dynamic
chart
for
smb.
This
is
giving
you
the
the
total
arr
for
smb.
If
you
want
a
total
parent
counts
for
smb.
So
this
is
a
nice
chart
there.
C
C
You
can,
then
you
could
come
down
here
to
see
the
delivery
mix
of
smb.
You
can
see
for
this
one
here.
This
is
kind
of
the
prior
month
mix
with
number
of
customers
in
arr.
This
delivery
kind
of
shows
you
the
quarter
over
quarter,
growth
per
smb,
so
on
and
so
on
and
so
forth.
One
thing
to
note:
I
know
tim
will
like
this:
here's
a
net
retention
view
here,
where
it
gives
you
a
view
on
net
retention,
for
we
have
the
smb
filter
so
by
your
territory.
C
It's
giving
you
your
smb
net
retention
by
territory.
It's
giving
you
this
industry
view
here,
it's
giving
you
your
prior
month,
industry,
metrics
and
then
this
one
is
giving
your
net
retention
view
of
the
same
thing.
So
just
to
kind
of
orient
you
to
to
the
chart.
I
got
about
one
minute
left:
does
it
have
delivery?
C
Yes,
delivery
right
here,
yep!
So
here's
your
delivery
of
smb.
So,
let's
see
smb,
it
looks
like
you
got
a
nice
mix
of
sas
and
self-managed
in
smb.
One
thing
to
note:
this
dashboard
is
at
the
ultimate
parent
level,
which
is
our
level
we
use
for
external
reporting.
So
it's
interesting
at
the
ultimate
parent
level.
You'll
see,
there's
ultimate
parents
that
have
more
than
one
delivery.
So
you
can
see
this
here.
They
can
have
a
self-managed
delivery
and
a
sas
delivery.
C
You'll
see
here
that
there's
in
the
filter
here
there's
some
there's
some
ultimate
parents
that
have
more
than
one
product
more
than
one
tier,
a
different
product,
different
tier.
So
as
you
kind
of
work
through
the
customer
hierarchy,
just
keep
that
in
mind.
When
you
do
it,
I
know
it's
a.
It
could
be
a
little
bit
tricky,
sometimes
with
dart,
without
work
with
our
with
our
parent
hierarchy.
So,
let's
yep,
so
that's
that's
kind
of
what
we
have.
C
We
have
some
other
charts
down
here
that
we're
still
playing
with
and
and
and
still
validating.
So
don't
don't
mind
this
section.
Renewals,
I
think
we
we
know
renewals
is
a
it's
a
tough
one
to
solve.
I
don't
think
we
quite
solve
for
what
is
what
what
our?
What
do?
Our
renewals
look
like,
I
think,
we're
close,
so
we
still
want
to
work
on
that
some
more
yeah.
C
So
that's
the
customer
segmentation
and
I
spent
a
lot
of
time
going
over
what
what
you
see
in
the
charts,
but
it
would
be
great
if
it
was
like
self-explanatory,
would
be
the
idea
and
then,
in
terms
when
you
review
it,
we
do
have.
Where
is
the
fusion
team
demo
yep?
So
we
have
a
link
in
the
agenda.
I
added
some
questions
here
and
maybe
on
the
team.
If
we
could
add
more
questions
that
we
would
like
our
early
adopters
would
give
us
feedback
on
like
how
easy
is
it
to
use?
C
A
A
If
you
want
and
if
you're
so
inclined,
you
can
take
some
of
the
sql
examples,
we've
provided
and
write
your
own
queries
and,
as
you'll
see
over
time,
you'll
see
a
lot
of
these
subject.
Areas
start
to
blend
together
and
you'll
say:
doesn't
it
doesn't
make
sense
to
have
it
there
or
there?
Absolutely
that's
what
an
integrated
data
model
does.