►
From YouTube: 13.9 release post Marketing retro
Description
Brian Glanz from Marketing looks at results from the 13.9 release post, which had both high amounts and quality of traffic. Brian traces that to having highlighted wider community contributions and to collaboration across Marketing on that and other themes.
Data referenced are in https://docs.google.com/spreadsheets/d/1d9qzzjDHcqKMTcrQB8GE1FQXCDIc0tdDWn93age7QHQ/edit?usp=sharing
Learn more about GitLab 13.9 at: https://about.gitlab.com/releases/2021/02/22/gitlab-13-9-released/ and in related coverage such as: https://devops.com/gitlab-releases-updates-to-ci-cd-platform/
A
A
You
can
see
the
most
recent
20
release
posts
and
that,
from
this
perspective,
13
9
isn't
quite
the
monster
that
it
looked
like,
but
it
is
easily
double
the
results
of
recent
posts
and
still
something
worth
explaining.
Each
of
those
other
great
posts
have
their
own
stories
on
the
lower
right.
You
can
see.
I
looked
at
just
a
couple
versions
of
quality
of
the
traffic,
so
bounce
rate
and
percent
exit
and
13.9
all
the
way
back
to
11.00,
in
fact,
actually
has
the
best
ever
bounce
rate
best
ever
percent
exit.
A
That's
amazing,
13
8
is
creeping
on
that
position,
though,
and
it
makes
me
think
maybe
something
structural
came
on
in
january
and
lifted
the
results
for
the
whole
site
there.
So
the
next
time
I
had
time.
I
validated
the
first
10
days,
views
by
looking
at
15
days
and
also
by
looking
site-wide,
didn't
learn
anything
from
the
15
days,
except
it
just
validates
the
10
days,
but
the
site
wide
on
the
lower
left.
A
We're
looking
at
in
this
case
all
the
page
views
not
just
the
release
posts,
but
for
the
same
spans
of
time,
and
you
can
see
13,
8
and
9,
both
with
a
significant
lift
but
roughly
equivalent
lift
and
this
maps
to
what
nile
was
talking
about
in
that
video
about
our
seo
changes.
So
thank
you,
team
seo
and
team
webb
and
niall
for
the
video
on
that.
A
Though
you
would
expect
13
9's
release
post,
which
you
see
in
purple,
to
have
done
similarly
well
as
13
8,
not
very
differently,
particularly
because
the
rest
of
the
graph
site-wide
roughly
maps
onto
the
graph
of
the
results
for
release
posts
instead
in
purple,
you
can
see
13.9
actually
did
just
eyeball
and
50
better
than
I
would
have
predicted
based
on
these
site-wide
results
in
black.
So
what
about
the
quality?
A
Well
in
gray,
and
the
upper
right
charted
are
the
site-wide
results
for
bounce
rate
and
percent
exit
same
15
days
and
then
in
gold
on
the
lower
right
again
from
this
range
or
in
this
level
of
being
zoomed
in
or
out
13
9
and
13
8
standing
apart,
so
nothing
structural
on
bounce
rate,
we
just
had
two
good
quality
release
posts
in
terms
of
bounce
rate
in
a
row,
good
work
teams,
13,
8
and
13
9..
Now
why
right?
Why?
Usually
you
get
high
page
views?
A
You
get
also
a
high
bounce
rate,
which
means
lower
quality,
but
not
this
time.
Why
did
we
get
both,
in
fact,
but
also
the
results,
our
views
being
our
main
kpi
there?
A
Looking
at
the
sources
was
the
one
thing
in
google
analytics
that
felt
like
it
was
teaching
me
something
organic
search,
social,
direct
and
referral,
and
then,
if
you
just
scan
across
back
to
13.2,
you
can
see
13.9
doing
better
in
every
regard
and
when
you
have
a
multi-channel
success
like
that,
this
is
the
first
note.
It
just
means
you
have
a
hit.
It
means
that
the
messaging
the
content
were
just
better.
Otherwise
we
would
see,
oh
it's
because
of
this
referral
source
or
it's
because
of
this
influencer
or
this
tactic
or
something.
A
Instead,
we
have
this
multi-channel
success.
That
just
means
the
content
was
great.
It
hit
the
market
at
the
right
time.
The
messages
were
right,
all
those
things
together
and
so
great
job
team,
thirteen
nine.
Obviously
I
didn't
do
it
myself
as
messaging
lead.
We
had
a
lot
of
influence
all
the
way
up
to
and
including
sid
in
making
this
better
and
a
lot
of
influence
from
within
marketing.
Folks
twisted
my
arm
on.
Like
the
themes
you
know,
are
you
sure
we
don't
have
something
on
enterprise,
devops
or
hey?
It
would
be
convenient
for
us.
A
Given
this
event,
if
there
were
something
in
here
about
cicd-
and
we
didn't
do
anything
inauthentic
here-
those
are
exactly
the
the
dozen
or
so
features
mentioned
in
the
introduction.
You
know
the
features
that
should
be
mentioned.
It's
just
about
telling
the
right
narrative
and
in
a
way
that
is
going
to
lock
those
gears
in
with
other
things,
we're
doing
in
marketing,
and
that
makes
sense
for
us
overall.
So
this
introduction
to
the
release
post
takes
five
or
ten
minutes
to
read.
Everybody
should
read
it.
A
It's
messaging,
it's
content,
it's
not
docs
right,
it's
not
a
change
log!
Those
things
are
great
for
what
they
are,
but
but
this
is
marketing
and
it
was
great
to
see
it
as
a
release
post
being
useful
as
a
piece
of
content
and
to
see
it
more
tightly
coordinated
with
what
else
we're
doing
in
marketing.
A
A
The
3270
are
not
yeah,
that's
not
the
whole
of
the
difference
there
that
comes
from
a
lot
of
the
hustle
out
in
social,
but
it
definitely
helped
and
it
helped
fuel
the
hacker
news
spillover
onto
that
front,
page
I'll,
get
to
that
with
some
visuals
in
a
second.
Let's
look
at
the
lift
and
search
for
thirteen,
eight
and
nine
compared
to
the
immediately
previous
versions
there.
A
That's
again
what
nile
was
talking
about
so
just
awesome
to
see
this
from
another
perspective,
what
what
he
was
mentioning
in
his
video
and
so
that
is
a
structural
shift
while
again
to
the
first
point
of
the
multi-channel
success,
139
still
did
even
better
right.
So,
okay,
now
what
happened
in
email,
600,
page
views
sourced
in
email
for
13.2,
then
800,
1500,
1200,
and
then
it
dies,
and
I
don't
know
why,
like
did
we
kill
a
newsletter
here
that
was
four
months
ago.
Did
we
fall
out
of
a
newsletter?
A
But
if
someone
knows
that's,
five
or
ten
percent
of
our
release,
post
traffic
might
be
nice
to
know
that
hustle
and
social,
like
I
was
saying,
and
with
the
media
in
general,
like
I
had
this
nice
interview
with
mike
for
devops.com
fun
to
get
out
in
front
of
the
media,
and
there
was
some
great
follow-up
there.
I
can't
say
any
one
of
them,
even
though
they
all
made
their
contribution
back
to
us
in
terms
of
traffic.
A
None
of
them
were
like
wow.
We
absolutely
need
to
do
this
thing
every
month,
nor
are
they
going
to
do
it
every
month,
so
that's
harder
to
replicate,
but
having
a
dedicated
section
for
community
contributions,
wider
community
contributions,
important
note
there
is
something
we
could
do
every
month
and
I
think
would
be
worth
doing.
It
rolls
into
those
social
data
and
the
conversation
out
on
twitter
points
to
the
conversation
on
hacker
news
which
pushed
it
onto
the
front
page,
which
ultimately
led
to
a
good
portion
of
our
great
results
here
in
the
release
post.
A
A
A
No,
we
didn't
pick
him
because
of
perseverance,
although
that's
awesome,
great
com
contribution
on
the
kubernetes
agent
here
and
just
around
that
great
work
kev
at
zika
gg.
Here
this
one
is
important
for
lots
of
reasons,
first
of
all,
great
contribution
to
vulnerability
reports,
but
it's
also
important
because
it
came
in
at
our
ultimate
tier
and
a
lot
of
folks
have
a
sort
of
misunderstanding
that
the
community
contributions
are
just
about
the
warm
and
fuzzy
free
and
open
source
stuff.
It's
just
a
community
thing.
They
sort
of
put
it
in
that
box
where
actually
it.
A
It
means
we
have
a
collaborative
trustful,
mutually
respectful
relationship
with
our
customers
that
our
competition
doesn't
have
with
their
customers
and
can't
have
because
they're
not
open.
It's
it's
a
big
part
of
why
we
can
authentically
own
that
term,
even
outside
of
this
stuff,
that's
technically
open
source,
because
yeah
source
available
it's
open
core.
Now,
in
addition
to
the
quality
of
the
contributions
this
month,
we
also
had
a
really
good
quantity,
I'm
linking
here
to
this
dashboard,
which
is
publicly
available.
A
That's
cool
I've
looked
at
this
before
in
aggregate
for
other
reasons,
and
in
version
10,
roughly
speaking
so
10.x
we
averaged
about
100
community
contributions
merged
per
month.
In
version
11.x
that
went
to
150
in
12,
it
went
to
213.
We
have
enjoyed
an
average
of
250
community
contributions
merged
every
month.
That's
amazing!
A
We
are
on
track
for
300
in
14.x,
we're
almost
there
and
there's
13.9.
You
can
see.
We
came
in
at
actually
301
went
to
press
with
299.,
so
the
quantity
I
mean
by
itself.
That
sounds
amazing
and
it
is
every
single
month
right,
but
also
it's
not
surprising,
and
it
lends
more
weight
to
the
idea
that
we
could
have
a
section
like
this
and
follow
on
distribution
and
marketing
efforts
every
month.
A
Now,
let's
look
at
a
couple
of
those.
You
know
I
nudged
the
social
team
and
I
nudged
the
evangelism
team
to
say:
hey,
I'm
right
in
this
section.
We
should
find
these
folks
out
there
in
the
wild
as
it
were,
and
and
thank
them
out
there
as
well
and
in
that
regard,
here's
our
brand
account
finding
roger
from
siemens,
our
evangelism
crew,
mentioning
siemens
and
follow-up
awesome
work
here,
michael
and
everyone,
the
community
celebrating,
along
with
them,
here's
michael
finding
kev
other
folks,
just
saying,
isn't:
kev
great!
Yes,
he
is
so
is
michael.
A
So
is
this,
michael
all,
that's
to
say
like
stuff
like
this
celebration
happens
every
month,
but
the
coordination
between
what's
in
the
post,
what
other
themes
marketing
is
working
with
celebrating
with
the
community
and
between
what's
in
the
post
and
and
the
social
team
and
the
evangelism
team?
It's
not
always
quite
so
tight
and
I
think
we
really
put
those
gears
to
work
together
in
13.9
and
that
spilled
over
into
the
results.
A
A
Our
social
team
also
has
a
great
idea,
which
I
I'm
curious
to
hear
more
about
in
terms
of
the
data
from
their
end
of
the
pipe
these
little
product
release
updates
and
it
looked
like
they
were
getting
some
traction
again,
I'm
just
getting
the
public
view
of
these,
but
I
do
see
some
of
those
spilling
over
into
the
google
analytics,
and
I
would
just
love
to
hear
from
the
social
team
the
evangelism
team,
how
they
feel
about
the
collaboration
here
and
its
effect
on
our
results.
A
I
think
that,
plus
the
collaboration
on
the
major
themes
really
made
13.9
a
success
and,
of
course,
I'll
be
sharing
this
with
the
rest
of
the
product
marketing
team.