►
A
Hi
there,
this
is
lee
matus
senior
support
engineering
manager,
america's
east,
with
an
update
about
our
support
response
crew
trial
after
our
first
week.
So
let
me
dive
in
starting
us
off.
I
just
wanted
to
share
with
you
where
you
can
find
more
information
and,
if
you're
an
audio
visual
learner,
then
this
is
the
spot
for
you.
If
you
prefer
to
read.
We
have
a
copy
of
this
slide
deck
and
our
other
materials
listed
here.
A
If
you
want
to
see
more
of
a
video
of
our
prior
intro
video,
you
can
check
that
out
and
otherwise
I'll
write
a
summary
on
the
issue,
so
you
can
check
up
there.
So
if
your
audio
visual
right
spot,
if
not,
we
have
the
resources,
you
need
to
do
what
you
want.
Thank
you,
diving
in.
I
want
to
take
a
minute
to
talk
about
feedback
that
we
got
positive
and
negative
and
some
questions
so
starting
off.
A
It
was
really
important
for
me
to
try
and
find
a
theme
here,
and
I
highlighted
things
that
I
think
were
a
theme
across
positive
and
negative,
so
it
was
nice
to
chill
and
grab
near
breach
tickets.
As
a
team
felt
like
we
were
actually
resolving
issues,
rather
than
only
babysitting
tickets.
That
one
was
really
strong
and
that's
something
that
I
want
us
to
continue
to
carry
with
us.
We
should
be
resolving
issues
and
not
babysitting
tickets.
So
I'm
happy
to
hear
that
someone
felt
this
way
way
better
than
doing
sla
hawk
all
week.
A
Someone
else
found
that
the
structure
that
it
gave
them
helped
them
to
communicate
when
they
could
help
and
when
they
weren't
able
to
help-
and
that
was
helpful
to
them-
also
collaborating
felt
more
smooth
and
natural.
So
that's
another
great
thing
that
I
heard
going
on
to
the
negative
side
of
which
I
heard
plenty
of
negative
feedback,
and
that's
great
thank
you
for
sharing.
We
had
11
people
write
in
for
the
whole
survey
and
I
wanted
to
share
some
of
the
things
that
people
thought
we
should
or
could
improve
on.
A
A
We
also
need
to
take
some
time
to
figure
out
how
u.s
fed
interacts
with
this
and
how
they
could
respond
to
a
surge
of
tickets
there
if
they
were
in
the
crew.
So
someone
thank
you
for
bringing
that
up.
I
really
appreciate
that
that
was
very
helpful,
because
it
was
something
I
had
not
considered
as
a
factor.
A
The
next
thing
here,
a
bit
conflicted
in
how
we
should
handle
some
tickets.
I
responded
to
prevent
a
breach,
but
I'm
not
familiar
with
the
subject
enough
to
provide
a
path
forward.
I
already
posted
to
see
if
someone
could
take
it.
No
one
did
I
felt
like.
Sometimes
there
was
a
friction
due
to
competing
priorities
of
working
on
my
assigned
tickets
versus
focusing
on
next
to
breach.
A
Someone
else
said
I
felt
like
I
was
the
only
one
working
towards
the
end
and
found
it
a
bit
overwhelming
to
juggle
four
breaching
tickets
and
one
of
the
last
bits
of
feedback
that
I
saw.
I
joined
the
zoom
call
for
an
hour
and
announced
it
in
the
chat,
but
nobody
joined.
So
taking
a
look
for
me,
the
theme
was
so
clear.
A
Collaboration
is
key,
starting
here,
people
loved
when
collaboration
was
happening
and
the
ability
to
work
together
and
how
this
structure
was
enabling
that
people
did
not
like
when
they
felt
alone,
when
they
felt,
like
others,
weren't
able
to
help
or
they're
with
them,
and
the
collaboration
was
not
happening
right.
That
to
me,
I
think,
was
the
theme
that
I
saw.
I
would
love
to
hear
from
you
if
you
see
or
feel
that
there's
a
different
theme
but
coming
out
of
the
positive
and
negative
collaboration
is
key.
So
thank
you,
everybody
for
that
feedback.
A
I
want
to
dive
in
now
to
some
specific
questions
that
came
from
that
feedback
survey
and
take
a
chance
to
try
and
address
them.
How
is
support
response
crew
materially
different
from
how
the
working
on
tickets
workflow
should
be
working?
So,
if
you're
new
to
gitlab
in
our
handbook,
we
have
the
working
on
tickets
workflow,
which
covers
how
you
pick
up
tickets
and
how
you
should
work
them
and
how
you
can
ask
for
help,
and
things
like
that
now
in
a
perfect
world,
every
ticket
would
have
an
assignee
that
would
help
shepherd
it
along.
A
That
would
help
get
the
ticket
resolved,
and
when
that
happens,
there
would
never
be
any
breaches
and
we'd
be
set,
but
we
don't
live
in
that
perfect
world
and
because
of
that,
what
I
want
us
to
do
is
make
sure
that
we
build
a
fail,
safe.
So
response
crew
is
sort
of
two
things
at
once:
right,
it's
a
replacement
for
frt
hawks,
which
is
a
form
of
fail-safe
that
we
need
to
scale
up,
and
it's
also
a
fail-safe
for
the
working
on
tickets
workflow.
A
If
everybody
is
working
through
their
tickets
and
we
have
no
breachers,
then
the
response
crew,
then
is
working
on
their
own
tickets
and
and
picking
things
up
and
maybe
solving
some
other
projects
great.
But
there's
always
going
to
be
some
failure
happening.
That's,
okay!
We
should
be
ready
for
it
and
response
crew
is
our
ability
to
have
and
fail
with
grace
now
I
want
to
be
super
clear.
A
We
had
in
the
past
frt
hawks
and
sla
hawks
as
our
fail-safe
and
now
we're
iterating
on
that,
and
if
this
is
not
something
that
works
or
delivers
value,
we
will
iterate
on
this
and
so
going
forward.
I
think
it's
really
important
to
realize.
We
will
always
have
some
sort
of
fail-safe
process.
We
are
not
going
to
try
and
build
a
process
that
doesn't
have
fail-safes
so
keep
that
in
mind.
But
this
was
a
great
question.
Thank
you
for
raising
that
the
next
one
that
I
saw
here
was
in
the
crew
documentation.
A
That's
going
to
take
you
two
or
three
hours
to
set
it
up
and
evolve.
We
have
complex
issues,
and
so
some
of
those
days
you're
not
going
to
be
collaborative,
but
on
your
crew
day
you
should
be
ready
to
collaborate
and,
as
we
talked
about
earlier,
we
saw
that
feedback
where
people
loved
collaboration
and
hated
when
it
couldn't
happen.
So
if
there's
only
one
day
a
week
that
you're
collaborating
with
our
team,
you
should
be
collaborating
on
the
crew
day.
A
I
would
love
if
you
collaborate
more,
and
I
think
that
that
will
happen
because
it's
already
happening,
but
with
intention
come
into
your
crude
to
collaborate,
and
I
think
that
that
will
be
the
difference
right
and
you
should
expect
other
people
coming
in
ready
to
collaborate
on
that
day
and
because
of
that,
it
may
feel
different
than
other
days.
We'll
see
and
that's
great
feedback.
I
would
love
to
see
in
another
few
weeks.
How
that
looks?
Does
it
look
different?
Do
we
still
need
this?
Is
there
a
different
style
failsafe?
A
We
should
be
exploring
we'll
find
out
so
next
question:
should
we
reframe
the
standing
zoom
room
and
make
it
so
that
anyone
seeking
to
work
synchronously
can
use
it
rather
than
it
being
just
when
you're
in
the
crew?
And
I
love
this
question
and
it
sort
of
goes
back
to
that
collaboration
and
I
think,
for
the
crew
trial.
We
want
to
be
intentional,
as
part
of
our
crew
structure,
to
make
a
space
for
the
crew
to
collaborate.
A
Now,
in
that
you
can
go
ahead
and
join
in
if
you
want
to
go
and
and
on
a
non-crew
day,
if
you
want
to
do
more
crew
stuff
and
help
the
crew
jump
in.
If
you
want
to
collaborate
with
other
people,
the
crew
is
around,
so
they
should
be
able
to
possibly
help.
Now.
We
also
have
plenty
of
other
non-crew
engineers
and
you
could
set
up
whatever
type
of
collaboration.
You
want,
whether
that's
a
pairing
session
or
a
group
around
a
specific
problem
or
customer.
A
I
think
that
I
don't
want
to
force
us
to
do
one
thing
and
say
we
have
to
do
it
this
way,
I
want
to
say
collaborate
for
the
crew.
We've
made
a
zoom
room
that
should
make
that
easier
for
everybody
else.
If
you
want
to
be
in
that
zoom
room
more
feel
free,
but
otherwise
feel
free
to
collaborate
in
other
different
ways
as
well.
So
think
about
it
like
that,
and
I
hope
that
that
helps
this
next
question
here,
which
is
a
really
challenging
one.
A
How
do
we
balance
tickets
coming
in
while
you're
on
the
crew
and
tickets
you're
already
assigned,
and
this
one
I
don't
have
an
easy
answer
for
and
I'm
going
to
lean
on
the
engineers
in
the
trial.
Please
help
me
understand
how
we
should
balance
that
what
should
we
be
doing
differently?
What
things
should
we
take
into
account?
Is
there
some
sort
of
setup
where
we
should
try
and
make
sure
that
you
have
less
tickets
assigned
to
you
as
you
come
into
your
crew
day?
A
I'm
not
sure
start
to
think
about
that
and
start
come
up
with
ideas,
because
during
this
time
we
should
start
to
flesh
that
out
and
figure
out
how
we
make
sure
that
gets
easier.
So
great
question,
I
don't
have
the
answer,
thank
you
for
raising
it
and
let's
collaborate
to
figure
out
what
we
can
do
to
make
that
better
and
clearer.
A
And
I
think
that
I
have
to
be
really
blunt
here:
only
you
can
prevent
the
bystander
effect
and
if
everybody
comes
into
the
crew
day
with
the
intention
to
collaborate,
there
shouldn't
be
anybody
standing
by
right
and
we
saw
in
the
negative
feedback
when
people
felt
alone
that
led
them
to
have
a
negative
experience
when
people
felt
like
they
were
collaborating
that
led
to
a
positive
experience.
Now
we're
trialing
this
out.
This
was
our
first
week
doing
this.
A
I
don't
expect
anybody
to
know
what
we're
doing
right,
we're
figuring
it
out
and
working
together
to
make
this
work.
So
with
that
in
mind,
I
want
to
just
say
I
think
our
team
is
great
and
I
think
we
already
do
a
great
job
of
collaborating
and
as
we
try
and
adopt
this
new
process,
I
do
want
to
put
collaboration
front
and
center
based
on
the
feedback.
We
saw
because
it's
clear
that
is
actually
the
secret
ingredient.
A
A
I
want
to
take
a
minute
to
talk
about
some
data,
so
I
pulled
up
some
data.
This
is
the
last
four
weeks
and
frt
has
been
dropping
over
the
last
two
weeks,
which
is
a
little
bit
concerning
our
nrt
is
pretty
strong,
but
our
frt
dropped.
So
I
wanted
to
take
a
look
at
another
graph,
which
is
our
heat
map
across
the
last
week
and
where
we're
struggling
the
times
are
in
eastern
standard
time,
and
it
appears
that
our
hottest
spots
are
where
we
go
from
amer
west
into
apac.
A
So
the
amer
response
crew
has
a
component
in
there,
but
does
that
mean
amer
east
needs
to
help
do
a
better
job.
Do
we
need
to
reshape
and
restructure
crew
days?
I'm
not
sure.
Yet
we're
going
to
keep
looking
at
that
and
I'm
going
to
ask
engineers.
If
you
have
ideas,
please
would
love
to
hear
them
if
you're,
seeing
things
that
you
think
are
not
working
well,
we'd
love
to
hear
about
it.
Thank
you
for
what
you've
raised
so
far.
Let's
continue
to
work
together
as
we
figure
this
out
now.
A
I
also
wanted
to
look
at
one
more
way
to
look
at
this
data
where
I've
merged,
basically
any
sla
that
was
not
lnr,
so
frt
and
nrt
as
one
right.
This
is
kind
of
our
total
sla
performance
across
the
last
six
weeks
and
you
could
see.
Pre-Americrew
is
relatively
flat.
We
had
one
outlier
day,
which
I
have
no
idea.
What
happened
there
and
I
didn't
spend
any
more
time
on.
A
We
look
to
be
in
and
around
the
range
that
we
traditionally
have
for
the
last
six
weeks,
so
at
the
moment,
no
strong
improvement,
but
that's
after
week,
one
coming
into
week,
two
of
this
process,
great,
let's
keep
an
eye
out
as
we
go
forward
and
see
what
things
do.
I
hope
with
that
refocusing
around
collaboration
that
that
is
going
to
be
the
component
that
leads
to
our
success.
A
The
last
slide
that
I
want
to
leave
you
with
here
today
is
our
action
plan
based
on
some
of
the
bits
that
came
out
of
here.
So
if
you're
in
u.s
fed
I've
created
an
issue,
come
there
and
let's
talk
about
how
we
figure
out,
is
there
a
better
way
to
align
the
crew
with
fed?
Do
we
need
to
pull
everyone
out?
Is
there
a
better
way
that
we
make
sure
that
we
do
it
every
other
week?
A
I'm
not
sure
I'm
leaning
on
u.s
fed
help
us
understand,
because
your
contributions
are
welcome,
but
I
don't
want
to
overstress
people.
So
thanks
if
you're
in
lnr,
you
shouldn't
be
in
the
crew,
some
people
were
that's
an
accident,
I'm
sorry
if
you
are
still
in
the
crew
and
still
getting
notified
about
it.
Let
me
know
I
can
fix
that
and
one
of
the
things
that
came
up
as
well
that
I
heard
was
around
alerts.
Can
we
get
alerts
back
in
pagerduty?
That
was
really
helpful,
I'm
not
sure
about
in
pagerduty.
A
Yet
maybe
amia
is
using
a
bot
to
do
it
in
the
response
crew
channel,
and
I
think
I
could
set
that
up
for
a
mirror.
So
I
have
to
create
a
schedule
and
then
set
that
up
so
keep
an
eye
out
for
that
and
if
we
can
get
it
to
pagerduty
in
the
future
great,
if
not
if
this
solves
the
need
great,
but
let's
keep
iterating
and
the
thing
that
I
want
to
end
with
is
we're
sort
of
in
this
holiday
mode
where
support
engineers
are
taking
time
off,
customers
are
taking
time
off.
A
A
So
the
next
few
weeks
are
our
time
to
experiment
with
building
that
trust
and
that
collaboration
as
a
crew
as
individual
crews
per
day
and
then,
as
we
come
into
mid-january.
I
think
that's
where
we're
going
to
start
to
see.
How
does
this
really
work
so
around
mid-january?
I
look
to
make
another
one
of
these
and
we'll
talk
more
and
I'll
ask
for
more
feedback.
So
thanks
already
and
look
forward
to
that.
Otherwise,
I
really
appreciate
your
contributions.
I
hope
this
was
helpful.
Thanks
all
have
a
great
one.
Peace
bye.