Retrospective on Steroids - Toyota Kata

Håkan Forss

Recorded at GOTO 2015

Get notified about Håkan Forss

Sign up to a email when Håkan Forss publishes a new video

so this is the first time in Denmark
doing the slides for the Lego that's a
really nice to be here so i will talk
about how to put retrospectives on
steroids using to your ricotta how many
of you in here do regular retrospectives
how regular is that like every month
every two weeks every week every day
multiple times a day hmm okay so let's
meet the hopefully this is the nigel
team they've been doing agile for quite
some time they regular do they
retrospectives and their ceremonies they
do things like pair programming TDD and
everything but it seems like they are
kind of becoming a little bit stale
they're not really improving anymore but
they do they're retrospectives they look
at what works good and what doesn't work
so good they collect ideas of what is
working what is not working what they
want to improve and they create long
lists of things they take and improve
but most of these suggestions of how to
improve actually points in many
different directions so how do they know
that I actually improving or not if we
will think this as physics and we are
moving in multiple directions with this
equal force what where will we be more
or less in the same place maybe moving
around a little bit depending on the
forces right so this is many times I've
seen in agile
teams that this is the case we can
collect a lot of stuff we have many
different suggestions on how to improve
but are we actually improving a lot and
many times I cannot come to a situation
where you get to an improvement whack
Amole meaning that you are having lots
of different suggestions you try a lot
seems to just pop back up again and you
have to whack it with another solution
is this something that you can recognize
a few nods so for me I think stop kind
of collecting problems and start
improving this is really what we want to
do so Who am I I'm an agile coach at a
company called King yeah we do different
types of casual games or games in
general yeah we are present in a lot of
countries so here's some facts about
King yeah we have developed more than
200 games we are present in more than
something like 340 million average
monthly unique users and this is from q2
2015 we are currently the global leader
in cross-platform casual game play a few
of the tightest you might recognize here
we have candy crush pet rescue farm
heroes and bubble witch something that
you might not know is that the company
was actually founded in 2003 so it's
really not a very new and a kind of a
startup company anymore we are present
in Stockholm which is the the first
office a London Barcelona am Alma just
across the water here in Bucharest
Berlin Singapore and in Cielo we were at
least in end of q2
and here's some interesting other facts
so I kind of mind-boggling kind of
things so we on every year we have 1.4
billion average game plays a day that's
quite a bit of game plays each day how
many of you play candy crush your candy
crush soda on a daily basis okay I won't
tell you boss and this year we actually
had the ones trilliant level played in
one of our games and we kind of moved
from web based games into facebook games
and now we are also present on mobile
here are the few of the main titles that
we have enough about king and let's dive
back into and to your ricotta stuff I've
been doing this for quite some years I
think I was exposed to Cara maybe 2011
or 2010 and we met at my often and
talked about this in 2012 I think yeah
so I've been doing it for quite some
time I'm really a terracotta geek I
think this is just like yes besides
something that I tried to use on a daily
basis even if my the ones that I'm
coaching is not really aware of it yeah
i'm an avid cyclist and i also like
barbecue and if you haven't guessed it
yet i'm a huge lego fan and just to make
this super extra clear now when i'm here
in denmark yeah lego is a trademark of
lego group which does not sponsor
authorized or endorse this presentation
I wish there were at least sponsoring me
but that's another question maybe you
to your ricotta so how many of you know
what a cada is do we have any suggestion
raise your handler did I miss you so
something you repeat yeah karate yeah
why do you do it so you do it you repeat
something a pattern to learn it any
other suggestions to make it without
thinking hmm so Cara at least what I've
learned is really comes from martial
arts or a kind of Asian tradition of how
to convey a knowledge between
generations indifferent for in different
areas and especially in martial arts and
is that you create a set of movements
that you repeat over and over so it kind
of becomes muscle memory and when it's
much muscle memory you don't have to
think about it anymore and not the big
kind of part of the movement but you can
start to hone in on the details and
really think about the content of how
you use it either in combat or a when
competition would that be an appropriate
kind of description for the guys doing
martial arts I see a few nods so a guy
called nanaka who research on f how we
convey information and knowledge sharing
and so on in his excellent book book
managing flow he describes Cara as a
synthesis of thought and behavior in
skillful action a meta cognition of
reflection in action
and what he describes the Cara compared
to a routine is that the car itself it
might seem on the surface to be quite
static but internally when you practice
it a lot it will actually change and
adopt to the context that you do it and
I think it was this morning when we had
from Lego and the Henrik you talked
about the shoe hurry so this is the same
kind of idea that we each shoe level we
do we follow a script but as we learn we
can focus on the content of what we're
actually doing and therefore we can put
all our efforts and all our focus on the
content itself instead of the form
anyone remember this movie so basically
what he were in karate kid yeah this was
the first version I saw the second
version with my son a few weeks ago he
didn't do this on a on a cordon yeah but
kind of wax on wax off he was practicing
on polishing the car but what he didn't
realize what that he was actually
building up a pattern of blocking
punches and kicks and what we really
want to do with Toyota kata is to kind
of create an organization muscle memory
for continuous improvement so we want to
ingrain this is something that we will
do without thinking on a daily basis and
the two Yorick Arabic it has two main
parts we have the improvement cutter and
we have the coaching Cara today I will
mostly focus on the improvement color
that's the actual doing that would be
the equivalent of a the retrospective
but in my point of view as I've learned
and kind of kind of made this part of
what I do I actually think the the
coaching Cara might be an even more
important part which is to describe how
a manager a leader
behave on a daily basis to build the
learning organization but we will focus
mostly on the improvement color today so
what is the improvement cutter well the
improvement Carter has to kind of main
phases we have the planning phase and
then we have the execution phase and
okay so we are not a droid conference a
planning might planning and face and the
execution phase and so on might not seem
so agile but bear with me yeah we will
see after the session what you think so
in the planning session we start by
trying to understand the direction what
are we trying to achieve where do we
want to strive towards so just like we
had a yes / say presentation just
recently talking about creating the
vision for a product or a project here
is the same we focus it on how we do
work more than on a product or a project
but it's how we can improve how we do
work so we really know what need to know
where we're going then the second part
is to truly understand how are things
working right now we know what we're
aiming for but we really not need to
know where we are right now otherwise
how can we know what steps to take to
try to strive towards yeah where we want
to go and then the lost part of the
planning phase is to set the next target
condition so this is kind of setting the
next sprint goal or maybe the next
release goal what do we want to achieve
within a short time period so
understanding direction a little bit
longer setting the next target condition
might be the Sprint goal or the release
go and then we come we step into the
execution phase and here we try to
iterate in small steps towards our
target condition that would be the next
step that we have said
the next target that we want to achieve
and then either when the time for when
we sure we're supposed to reach our
toric condition and the sprint end of
the release if we have either reached it
at that time or before then we exit out
and we repeat so this is a pattern that
you repeat over and over and it actually
is something that you basically never
stop you always try to strive for
something better in another place
describing is we have a vision we need
to understand where we are then we need
something that we ascribe would be like
one year or into llores case they are
kind of aiming ten years into the future
that might be a little bit big elephant
grasp in one bite right so we want to
set a challenge that this is a little
bit closer and that maybe is half a year
or a year out but it's still not
something that you can achieve
immediately and it's typically set in a
way that it is definitely outside of
what you think that you can achieve but
you want to strive for it and then we
set a number of these target conditions
until we move come up to the challenge
and then we reset the challenge and
again okay another example a vision we
should boldly go where no brick has been
to explore space and so on maybe this is
this Morse maybe so the current
condition think of yourself stepping
back in the beginning of the 1960s we
have rockets yeah potentially they can
lift off without exploding and so on and
then someone set a challenge we choose
to go to the moon and so on did they
know at that point that they will
actually achieve it probably not it was
really a bold challenge that they set
out they definitely didn't know how to
reach it but they broke it down to small
steps that would be Tory conditions yes
we need a rocket I could lift off we
need a rocket that can actually go out
to space maybe Circle it earth we want
eventually we ship we send a rocket off
and circle the moon but not land and
then we have a target condition of
actually landing on the moon itself and
coming back so this is kind of the idea
of how the improvement cutter itself is
framed in another context then let's
take a quick look at the coaching Cara
and the coaching Cara is supposed to
have the leaders coaching the learners
and a company that has implemented this
fully that means that every leader in
the organization has a number of
learners or someone that reports to them
that would do improvement work so you
have a the leader is supposed to be the
teacher or the coach for the rest in the
organization working on improvements and
the culture the leader should be someone
that gives a helping hand kind of help
I got a kind of learn and a I grow as an
individual the leader should be helping
that out but also be someone that kind
of pushes you in the right direction
well if we are an agile company maybe we
shouldn't do longer planning and bigger
batches and so on we actually want to
have smaller batches and we want to have
more of a continuous flow and to do this
we can follow a coaching script that is
a basically on this cord and it could to
kind of highlight what you go through
and you go through this in every
coaching session and this code you can
download from micro ders a website is to
always kind of reiterate where is the
net what is the next target condition
that we want to reach and then what how
are things really working today and then
we ask yourself so what was the last
step that we tried to do what was the
last experiment that we want we were
trying to achieve and what did we expect
by running that experiment alright or
taking that step and then comes the
really important parts here in in the
trio ricotta and the coaching side of it
is to what actually happened so we
should be able to really understand what
happened and now we get to the really
exciting part that's why we're dimming
the lights right what did you learn so
to your ekkada is really an improvement
cutter is really about how do we
optimize for learning it's not about
necessarily achieving a certain result
but what did we learn by trying to do
this and then we step back and we ask
yourselves what obstacles do we actually
think now is preventing us from reaching
our target condition what is your next
step what experiment do you want to run
here's a really important part what you
expect will happen
when we run it and we define this before
we run the experiment and then we decide
define when can we go and see what we
have learned from taking that step so
for me this coaching Cara is really what
a true lean and agile leadership style
is all about if we go through a similar
kind of pattern or following this in a
shoe kind of level approach of how to do
coaching in continuous improvement it is
really really what i have found to be
lean and agile leadership style okay a
story so this is not from king but from
another customer that I had before a
joint king and how we had an agile team
that went from cannabis scooter to a
race bike air and they were doing kind
of regular retrospectives but they
didn't really see any great results so
not so fast and they were not so
motivated in terms of doing improvements
on how they did work and by introducing
to your ricotta they become much faster
and much much more motivated in actually
trying to improve their way of working
so what were they actually done trying
to do well the problem was that they had
a big kind of proprietary hardware that
they loaded in their own proprietary
software into and they wanted to try and
see if that software worked in that
hardware as often as possible and this
process might be familiar to some of you
this is called continuous integration
and it's basically about trying to fit
two pieces and if they fit it should be
green and maybe we get a very nice green
indicator that says wow it's actually
working or when you try to fit it
it doesn't fit you want to have a red
signal or some warning lights going off
telling you that it's actually not
working so when it works it will it
should be green when it doesn't work it
company was that quite a few times we
when it actually was working we got an
indicated indicator that it was not
working and when we actually got it to
work sometimes it was indicating that I
didn't work so this we wanted to kind of
avoid so we sat down we started to look
at what do we actually want to achieve
and what we want to achieve is when it's
green when continuous integration works
and the software works in the hardware
we want it to be green and when it's not
working it should be red sounds simple
wait is this how you do continuous
integration at your company obviously a
few notes okay yeah but the problem is
this it was actually really hard in this
case because they were loading it into
the hardware and so on it was easily
that it could kind of fail on the way so
instead of aiming for this that we
actually thought would take some time to
meaning that whenever something showed
that it was not working it should be
turning red but we couldn't guarantee
every time it was Green according to our
build it was actually working because
that was a little bit out of the control
of this team and quite hard to do so we
put up this Toyota kata board yeah we
started to look at so what's the actual
condition right now how is the the build
system actually working and the team
we don't really know a few and the
organization is complaining that when
they build something sometimes it shows
red and sometimes it shows green so they
had to actually find out how it was
working so they kind of stepped out of
their protected environment and met the
customers started to look at how was it
actually working and they went and as
it's often expressed in lean to go and
see where the real action really happens
so they went there they gather data and
they came back and what they could say
see is that about seventy percent of the
Reds should actually be green meaning
that for some reason the build failed
when the software was actually working
so of course that meant that no one
really listened to the build system
because whenever they rebuilt it could
be that I was working or not and they
identified a lot of different problems
or obstacles of why this was the case
but instead of trying to make a long
project plan of how they can actually
achieve this yeah we took the to your
car approached and we focused on only
one of the obstacles and instead of
saying we just gonna tack an obstacle
instead we define how we wanted it to
work when we have fixed the obstacles so
we set the target condition and we said
of the seventy percent read that should
actually be green in a month's time we
want forty percent of the red to
actually be green they started to work
one problem at the time we picked one
story to work i'm at the week later and
they worked a little bit more and after
they only reduced from 65 sorry from 70
to sixty-five percent so they haven't
really achieved the target condition but
I worked on quite a few of the obstacles
so why was this well we'll take a look
at that the team really still didn't
feel so fast or so motivated but why
well in this case the biggest optical
right now for the team was that they
were too busy to improve they were just
keeping everything running and couldn't
really focus on the improvements so what
we really tried to do then was to
address this point as the main obstacle
and in a lean and agile kind of fashion
when something is really not working
what is the first thing that you will
try to do any suggestions yes try to
learn from it that's a good point yeah
and how do you learn from it the easiest
way I found in the very pragmatic way in
the lean and agile fashion is to do one
thing is to have the batch size always
have the batch size or double the
frequency of the things that you do so
this would be your first gut response to
almost any problem when you're not
improving enough why because if you
practice it more the more you can learn
the more opportunity that you have to to
actually learn from it so instead of
meeting once a week we decided to meet
every day so we will work on the same
thing the same problems but we will meet
every day that was the only thing we
changed we still at 65 sixty percent and
we're taking one small step at a time we
came down to fifty five percent
still not at 40 team is a little bit
more motivated well it was still not
really reaching the target so what did
we do well we decided to actually meet
as soon as the task itself was done we
meet instead of once a day we met
multiple times a day and in this case it
was kind of making the changes done even
smaller even something that you can do
in a few hours instead of something that
you can you need an whole day to fix and
what really happened then was that we
could move even faster with the same
amount of actual time spent on doing it
and moving much much most faster towards
the forty percent goal and at this third
attempt we reached the forty percent
goal and the team they were much more
motivated and they improved much more
than I did before and basically they
moved from a scooter to a race bike in
terms of doing improvements in this team
and they did this through smaller
batches we met very often instead of
meeting once a week or every two weeks
or whenever we have our retrospectives
we met as soon as we could minimal
viable change and we do them as fast as
we can and in the beginning try to focus
on some of the quick wins that you can
do even if you don't think that they
will actually be the main route course
sometimes fixing the small things will
actually make the big problems go away
and of course have fun and this team
they felt well to ricotta rocks so then
let's step back and kind of look a
little bit more in detail of how the
improvement cutter works so we have
understanding the direction graphs the
current condition
set the next target condition and then
iterate towards the target condition
understanding direction we talked about
it before setting a direction or a
vision where we want to aim for is
really really crucial if we don't know
where we're aiming for we can come up
with million different types of
solutions but we might not solve the
problems that we need to solve and as
Bruce Lee set a goal is not always meant
to be reached this off it often serves
simply as something to aim at so if we
put a goal for in the future is not
necessarily that we actually need to
achieve that goal because it's so far in
the future so we probably on the time
when it takes to get there we kind of
have changed a little bit direction on
the way but it's something to aim for
and everyone can rarely around it and
when we talk about to your ricotta it's
really about a process focus and not an
outcome focus so to your acara itself is
not focused on delivering a product or a
certain service or anything like that
it's about how we set up the system of
work so we can achieve what we want to
achieve getting a product out to the
market or having a certain service level
or whatever it can be so when you talked
about the safe this morning kind of
describing what you want to achieve and
having this cadence yeah being able to
do the replanning every eight weeks is
that the cadence that you use right now
yeah gathering everyone together be able
to focus on doing that that is a
capability as of an organization to be
able to change direction and have a true
focus moving forward and to your ricotta
is about setting up a system helping you
improve that system so that can happen
all over again
company vision vision it's about how we
want our processes to behave so we can
achieve the business vision and the
bitch the company mission so how much
this look like if we would look to to
your Toyota when they produce course
they would have something like this I've
never worked for toyota myself so this
is only second or third-hand information
but you will see something similar to
this their vision for product operations
is zero defects one hundred percent
value add that means that everything
that you do in that process should add
value to the customer one-piece flow in
sequence on the mound and security for
people do you think toyota is close to
this anyone here has a toyota or they
siri defects and it affects in your
Toyota's I see if you notes so there to
my understanding they are very high on
the short of the number of defects the
low number of defects on course but they
still have I think it's a 100 defects
per every thousand cars or something
like that so it's still not zero defects
do you think that everything they do in
the factories is one hundred percent
value added no not even close because if
this is where the raw material is stored
and the core on the line is over here
the movement from taking something up
here moving at the integral car here is
considered to be waste so for Toyota
they are far from being at this vision
and they will only been working on it
for 70 years
so how could this look like in software
m well yeah looking at the kind of
long-term perspective for a software
company that might be CEO defects and in
this case I added in production because
product development typically needs a
certain amount of a exploratory before
we actually go to production so we can't
have perfect quality the first time see
represent a one hundred percent value
added meaning that everything we do
should add value to the altitude a
customer and the product and maybe
highest value first on the mound with
you as a company be able to do this
today the companies I worked for and for
King we're definitely not close to this
but it's something to aim for grasping
the current condition so one of the
things that you really want to do as a
leader especially is to step out of your
corner office you need to go out where
the work is done to go and see and find
out how things really are at the gemba
you can't sit and look at things in and
I report or anything like that because
you don't learn all the subtle details
what is actually happening you need to
get out there and find out how it
actually works and especially for us
that is working with intangibles things
like software typically what you want to
do is to bring that intangible kind of
thing and make it tangible so
visualizing you work yeah making it
physical can really help being a lego
fan one of the things that I try to do
sometimes is to actually make those
visualizations in Lego of course and
having people to actually move around
Lego bricks instead of post-it notes can
actually make it more fun
I see a few people kind of thinking we
have bricks a lot of them maybe we
should do that so what do you want to
collect well you should try to collect
data and facts not gut feeling but
sometimes the only thing that we have is
gut feeling well then try to turn that
basis collect your gut feeling yeah so
it becomes data it should be described
your process how is your process
actually working how is the process
of running your processes as a kind of a
final gauge on if you're improving or
not so not diving too much in detail
here but what we really want to do is to
focus on how the process itself work so
we want to not focus so much on what we
actually deliver but how we deliver how
long does it take to have a finished
build how many times do we check into
trunk yeah how often do we pair program
and so on a lot of these things based on
how we actually work not on the outcome
itself that is much more important to
measure to be able to improve your
system than the actual outcome of course
we need the outcome to be able to have
some gauge if we are improving a lot but
it's only the gate a gauge for if we are
improving a lot it doesn't really tell
us what to do to improve and typically
if you look at lead time throughput and
quality would you say that they are
decisions or something that tell you
immediately if you change something if
you got a positive or a negative result
on it these are typically lagging
indicators so lead time for instance if
you change how
deployment cycle until you have new lead
times you can't see them a prompt so
what you instead we'd like to measure is
things like queue size working process
and so on there's a much more important
than the actual outcome and typically
this is something that we don't measure
we've measure how the velocity and stuff
like that and that is only the result of
how we do work not something that we can
actually take action on and we have
setting the target condition and here
comes kind of the coaching dialogue
between the leaders and the people doing
the actual work is to find what is a
target condition that is just right and
what we want to do here is to define a
target condition that is just outside of
our knowledge threshold meaning that
it's we don't necessarily know how to
actually solve it but it shouldn't be
kind of something that we have no idea
how to do it it should kind of feel like
fitting a square peg in a round hole
because then you can push yourself then
you can kind of think outside the box
and when we set these target conditions
it should be an hypothesis on the
journey towards your challenge condition
it should be based on your business
strategy or model for how to do process
improvements so if lean and agile is
your thing increasing the batch size is
probably not a target condition you want
to set if you don't want to prove a
certain point and then you should follow
the golden la cruel how many sweets do
we have in in the audience not so many
lower gum is a great word that we have
in Sweden do you have that in Danish the
gol de la cruel it should be not too
hard not to ease it should be just right
sorry ok I will never try to pronounce
that but that sounds good
in most other languages especially
outside scandinavia they don't have this
word so the goal de la cruel and this is
look of course really really hard and
that's why it's important to have a
coach or a mentor present all the time
so you can kind of try together to find
this and when we set these target
conditions we have our current condition
we kind of aiming towards the challenge
do you think we will set the target
conditions like this well no it's more
like this we have our current condition
and we are aiming for something what we
want to do is to just to take a step
into the unknown and when we are in that
position then we can kind of redirect
the next step so we only take one step
in and see what happens and many times
the problems that we saw here when we
took the first step a lot of them just
disappeared and we have a new set of
fine problems that we can address
instead some examples of target
conditions that I've used in the post
might not be the perfect ones but it
will kind of help you to think about
that target conditions is really about
the process how we run the process is
not delivering a certain number of a
story points or a velocity and so on
that is kind of more on the output side
we want to shift it more to how do we
actually work and then we comes to
iterate stored the Tauri condition and
this is if we kind of skip with the
planning phase here it's very easy to
skip this and jump immediately into here
what you typically find on is that you
will have the same position as we had in
the beginning a lot of different actions
moving in many different directions but
we are actually not moving anywhere so
what we really want
to do is the planning phase first and
then we start to iterate and when we
take think about it rating think of
yourself being in the laboratory you
have the two vials and you want to pull
them tick putting together and what
happens wow I did not expect that hmm
what is that sound that's the sound of
learning because if we only do the
things that we knew in advance what
would happen we are just late
implementing it experimentation is about
trying not to throw mics on the floor I
did not expect that to happen so I
learned something don't put the mic on a
very blank table so what we really want
to do if we want to optimize this yeah
we want to place put down and hypothesis
when we before we run and there should
be text in here and let's see if we see
anything no so it should say hypothesis
prediction experiment observation and
then we have the learning because if we
don't write down before we run the
experiment what we expect to happen we
are actually reducing our ability to
learn because the human brain has a
dysfunction calling a confirmation
biased which means that if we don't
write it down and write it down in
details after the fact we will say well
I expected that to happen and you are
actually reducing your ability to learn
so if you want to optimize for learning
and improving your system of work go
through the scientific approach of
trying to phrase an hypothesis make the
prediction what you think will happen
closely observe what happens when you
run the experiment and then look at the
between what really happened and what
you expected and you should really try
to set it up so something like fifty
percent of the things that you do is not
getting the result that you were
expecting so in information theory this
is kind of the optimal where you can
learn as much as possible but still have
a result so to your ricotta for me it is
really about creating an organization
muscle memory and for continuous
improvements it is also for me a true
leadership style that will represent
what lean and agile is really about we
look at improvement as experiments and
by having these fixed routines that are
kind of defined basically like a script
in your shoe state when you try it over
and over again the same way which will
feel mechanically in the beginning it is
kind of a guiding rail kind of stepping
into the unknown because if you repeat
that pattern over and over again and
your energy on the things that is
unknown so are the core does the exact
car is important yes but having routines
is more important so if you find the car
as described here not to be exactly what
you would like define your own cutters
and run them over and over again until
they become muscle memory that is more
important than following exactly the
points that's been described here but
it's probably a very good starting point
if you haven't tried anything like this
before try to cut us for a couple of
months and you will see that there's
some depth in there that you didn't
realize in the beginning if you want to
learn more there's a number of links the
slides will of course be available and i
think the video will be available as
well at the top you have my blog with
the tag to your ricotta and you will see
a lot of Lego slides and quite a bit of
material of course micro ders site where
he has the source material and please
feel it oh please feel free to go there
yeah and I have links on my blog to the
air towards that as well and with that I
will close for questions and remember to
first a problem most appointment is a
secret coach turn on those damn and i
canna cross okay what you sometimes
management doesn't allow strandings
implementation alligator in general I
would say if you have a resistant to
change try to make the change as small
as possible and time limited so when I
work with team like with the Anna's team
here I would typically say yeah well
let's just try something and depending
on the buy-in you have yeah if you have
a very low by and try to make it as
short as possible and just make one step
because when you make one step and it's
not dangerous then you're typically
allow to take the next one so try to
make them as small as possible and time
limited one where you draw a line
between leaders and is going to see the
actual work away from that corner office
and then being a disturbing story scary
figure oh gee this is actually quite a
hard question to answer and what we did
with one management team well actually
you need to show respect as a leader and
to do this you need to be very safe as a
leader so we had a number of managers do
like a gamble walk on a regular basis
visit the teams look at the boards
asking questions about what was going on
and so on yeah but they didn't know how
to do it so they felt awkward in that
situation and therefore that kind of
projected to the teams and the team's
felt that they were kind of monitored so
what we did was actually write down a
script of how these managers should
behave and we added things like when you
step into the team room the first thing
you would do is to ask is it okay if we
enter the room right now is this an
appropriate time and if they say yes
well then they have invited you in if
they say no then you can ask when would
be an approved
time to do this we also made this script
actually available and we define for
each step of that script what was the
intent of the question so if they ask so
what can I see on your board it's not
about kind of controlling it's about
them trying to learn so you're kind of
being open and transparent what you want
to do thank you