Event Storming with Alberto Brandolini (Alberto Brandolini)

I started coding uh uh more than 40 years ago I'm one of the uh sincl Spectrum uh kids at that time uh then

later uh coding became a way to make uh money uh just before going to the university the little weekend jobs were about coding uh then I graduated in

computer science and and started a profession in in that field a few years later I started my own company so I joined the business side I'm now I'm

paying developers to write soft for me I still open the ID for a kind of a meditation style and um yeah that's more

or less uh the the the story then I also ended up in uh in the first agile Consulting and then large scale modeling

facilitation that was not exactly my uh traditional stuff uh but then it became uh it became interesting 10 more than 10

years ago now this idea of even storming popped out the that that is the original blog post and the date is actually November 18 uh

2013 also the platform is uh now is kind of uh outdated but I still still kept it but the original article is telling an

original story which is not not the story I would like to tell today so in 2013 first blog post the original idea was to get in one

single Workshop from business to aggregate Discovery and still it worked in in some places but then the more we move to maturity we discover a few other

things so uh 2019 the more or less uh mid Midway of the even storming Evolution uh the book

was in the making I had a community of practitioners idea but it was o split the format into uh three uh different recipe big picture process modeling and

software design but we also had chances to play it in very different scenario from corporate retrospective to uh induction of new hires designing

startups modeling critical processes uh redesigning architectures and many many different opportunities people ask me to do something that I've never done before

I was warning them like I've never done that but I'm happy to try and yes that was great then of course everybody had

this little problem in 20120 and uh and everybody was uh forced to deal with uh with remote I already had discussion about remote before this and my

discussion were something like uh gra get a plane ticket you cannot do it remotely blah blah blah and uh and of course I had to rethink some of this

constraint because in 20 20 we didn't even know if we could ever get back to normality of what was the actual concept of normality put in another way the very

idea of putting all the key people in the same room was illegal at that time and most of our businesses that were running conferences workshop and training classes and and even events and

and so on well that's not possible anymore and uh so we had a lot of free time to R invent and rethink the whole

story so well let's spice up our mirror license and get things working um but now we are in 2024 and in person is an option again uh

I don't know if we ever got to normality I think the pendulum is is swinging still uh a lot many organization actually turn to remote as a default way of working and some headquarters are

kind of empty uh but I think the pendulum is still swinging a lot and uh what is the

resulting shape of a company of an organization is still yet to be assessed correctly I was actually talking about it in the last uh uh XP conference in

Bano uh just three weeks ago and if you're interested in this one I I have a very uh um kind of sad and pessimistic

perspective over the consequences of uh remote working like I think organization didn't fully understand the consequences and it would align all the connect all

the dots it's not exactly an exciting picture uh but I'm not going to be talking about this just if you're curious or maybe spacing the Q&A uh

that's a that's a talk that I uh that I wrote for this specific thing end of the intro I would like to get into big picture even storming well

um the big picture Workshop is probably the most uh interesting because it's uh it's defying some of the rules of traditional

worship just uh start with the idea of inviting all the right people part of the recipe is providing an unlimited

modeling space this is vital because so many problems we are dealing with just get discussed or

artificially chopped divide and conquer this type of thinking many times because one we don't have enough space to model this we are constrained by the size of a

whiteboard for example um and we think that we have to focus we we we we are missing the tools for dealing with

complexity in the large which is a very dumb starting point if we think about it if you're dealing with midsize problem let's l Lage our ability to visualize

sketch a diagram visualize the problem in a few things discuss pointing finger to things but if we dealing with large problem oh let's get back to the less efficient let's talk about it around the

table which is not going to help so we are having Advanced strategies for midsize problem and dumb strategies for large problem so it doesn't take a genius to think what if we have more

space maybe we could deal with larger problem but the idea of unlimited modeling space is actually larger is not only the paper roll given the illusion

of a giant size whiteboard but also the abundance of markers stickies whatever is going to make you feel that there is no scarcity don't think about the market

don't think about oh we won't have time to discuss this just write it down no problem we're gonna find the space for for this and then the most natural shape

for a for a big picture Workshop is to model an entire business line uh every single Department involved every single step all the story from the beginning to the end left to

right so the question about hey well shouldn't we focus uh yeah that's the answer is no most of the time uh because

there is so much advantage in seeing the whole at least once in your uh career inside an organization then of course this is not

a Dogma you can can run and even storming instead of a big picture midsize picture but you will miss a lot of the potential like there's something

special that happens when you're looking at the board you are exhausted and and uh yeah you feel like that is that is actually everything I cannot think of

anything more that should fit into this picture there's a there's a a bonus get gets unlocked at that time so okay so how do you run a worship like is because

looks like a big mess well let's start with a mess actually making a mess and sorting it out is exactly a key mechanic

for for a big picture even storming uh we we provide some Foundation just a paper roll and the timeline left to right I don't make long introduction I

really would like to uh give the message of we can start very very quickly um and so my introduction is two or three minutes or maybe maybe even less there's

a little bit of extra facilitation tricks like making sure everybody has a marker um being intentionally rude uh so

maybe not giving markers to everybody but and markers and stickies uh but making sure they are accessible but they have to stand up so the moment you are

already standing okay well now it doesn't make sense to get back in the table in right I will go straight to the surface the other the other main trick

is I'm leveraging the power of imitation like I'm asking people to do something for the first time and and of course they don't know

how to do it perfectly what do human uh human beings do in this this scenario they look around they look to other people that seems to have a a little bit

more comfort in in this scenario and they imitate they repeat what others are doing they mimic they don't know exactly what you're doing but the moment somebody starts putting orange sticky

nose to the wall and the facilitator doesn't correct them like like oh your your event is wrong or anything like this please don't do this um if one is

doing it and it gets approved oh well that's easy I can do it too so you got to zero one two everybody really really

quickly the grammar we're using for the exploration is uh is very simple it's intentionally simple we have domain

events for historical reason we started using them in in Orange uh at the beginning I thought the core was going to be the aggregate so I kept the common

yellow for the aggregate then I needed colors for avability and event so out to be orange then I kept it because yeah I had already pictures in this way um and

the grammar is simple this is orange this should be in the past St put it on a timeline and even kids can do this um it's not entirely true somebody will

make it wrong but just let it go for the initial stages the reason why this becomes interesting is because inside the organization Nobody Knows the entire

story uh you will uh you will have different sources of truths areas where the truth is lacking you only have secondhand information but everybody

espe in very side of the organization will have only a partial representation of the story and you don't want to run separate interviews especially when

kicking off a project you want to see and you want also to show that individual stories do not make a coherent whole then of course if you

force people to um work on a single surface while you're constraining the discussion to lead to lean towards a a here at all because well this area

should fit with this other one this event should happen right after this except an exception that I should be documenting on the board as

well um a little warning uh nouns in every exploration will fool you and if you're trying to skip the learning part

using llms or any other new fancy tool they will be full as well but let's move on Second Step is enforcing the timeline so the the first moment was called

chaotic exploration basically everybody was free to do whatever they want it okay now I see clusters which are locally organized but the hole is not consistent let's try to make the hole

consistent at this moment people start to talk there is no way you can sort 200 sticky notes without having a conversation with your neighbors and the neighbors are exactly the people you

should talk to so one of the brilliant thing of storming you have localized the parallel conversation happening exactly between the people that should talk to each

other that's that's pretty cool when when you see it happening you realize I'm saving a couple of weeks for the for

the entire project lifespan this conversation is is so precious of course not every conversation is uh peaceful or

U you will have S ASM you you will have friction and contradiction this is where a new color start to help you um magenta probably but this color is named

differently according to Nations and and and culture and also to availability of a 3M sticky Supply but I call it a hotspot like whenever there's there's

something relevant popping up in the conversation I captured this so the model is not only for the beti full part actually is a model of all the possible

contradictions in the system and uh and this is more or less the result this is a result for a conference organization scen scenario and it start with event

building blocks of our story telling the hotspots then pop up then we start adding the people little yellow doing things and the systems uh the where the

things can happen it might be technological like I see awf while there was also printers but it might be external organization like tax

agencies governments and so on an interesting definition for systems is also whatever we can blame that opens the gates of possible negative thinking

initially the workshop starts with very optimistic thinking the first thing you get described is the rosie scenario uh for example uh but then we keep on the

conversation and during the conversation some boundaries are starting to emerge the little vertical yellow lines so whatever is the color of the uh paper

tape you can uh find and then in this scenario we also uh went uh adding some extra toppings like ideas or opportunities to improve the systems in

Green in this scenario and also voted for the areas that could the be changed first or the problems that needed an

improvement more more massive in general um a big picture format has something like a standard recipe these are the key steps like chaotic

exploration enforcing the timeline kick off the conversation and a little bit of sorting plus hotspots but not good enough because it's localized

distributed and so on then we add more ingredients with people in systems people in system also forces us to uh add more event because I might have an

event which is specific to an external system or the potential failure of an external system will open a Side Story okay now we have enough information for

what usually happens in the afternoon which is called the explicit walkthrough where a narrator will try to tell the story in front of the audience with the

support of a scribe narrator will tell the story the audience will say like oh wait a minute Alberto it doesn't happen this way we have also this and this scenario or sometimes we do this and

this and the poor scribe is writing all the possible notes in form of events so that we can adjust the model to make

sure that what we see has a one-on-one relationship with what we just said it should follow the grammar but also should reflect the um conversation that

just happen so progressively the board starts getting tidier and reflecting all the key conversation of

the people you can see people started to nod in approval like okay yes now I see that my objection is visualized and the model is getting better what we get at

the end uh this is usually the moment uh where we are one hour before closing the day but this is the moment of okay we

see the whole thing we we have the clarity of okay we sorted out this mess and this is also very powerful way to

learn in the physical world but also one of the little bonuses that I was mentioning before it's just especially in large organization one of the

blockers for change is that people are afraid of breaking invisible things think about being in a bank uh you're always going to find the Departments that you were not aware of and you're

not going to change anything of the default behavior of because because if you break something that you're not aware of you're making the bank lose

money po are bankers and uh but this attitude is is absolutely there the solution of even storming is just let's

just make everything visible so that this can eventually enable the confidence of what if we skip this step

nobody's actually interested or what if we replace this sub process with these other alternative we have to manage this and this consequence and we see all of

them there is a backup plan for each one of these scenario confidence is actually very very uh uh interesting as an outcome of this

Workshop other bonus outcomes like uh once we complete the work through and the validated narrative we are in a great place and uh and we could decide

what to do uh uh at at this level let me let me be more explicit about the outcome of the exploration uh we are not solving the

issues during the workshop there is no time for this we are trying to time boox it in one day but we are trying to visualize everything um this format is also really scalable just because we try

not to solve stuff like and there's a lot of parallel activity going on we can make it scale the problem is one side the facilitation but mostly is

lack of oxygen meaning like room size if we have a very large room we can make it work well 35 and even 50 more more

people um we might have a map of the current situation or a plausible map of future state it works uh in terms of notation exactly in the same way in

terms of Dynamics is a completely different beast but that's about yeah in the future you don't have skeleton in the closup and in terms of outcomes I

like to separate official and unofficial outcomes there's massive learning we can extract bounded context we can have corporate alignment but in terms of unofficial

outcomes the business side might be learning the business in a way that would never admit and and also uh for for me as a

potential uh um person involved in a software project spot in the enemy meaning that um Enterprise projects do not fail because of Technology they fail

because of politics and in this Workshop we see politics in uh in a very clear way once we get the validated narrative

every other layer becomes chap so basically in the last hour we run downhill and uh so the the approach that

I like most is uh looking for a blocker highlighted by uh blue uh blue arrow this is telling me Oh we gather all the key people in the same room we discuss

about everything for the whole day providing uh data clarification and all the possible stuff

everybody decided uh voted stated that this is the most important problem why not solving it so the actual outcome for me

is having the analysis and the political momentum in the same Workshop that is massive like there is no way to stop a train like this this might not always be

achieved sometimes you have political reasons not to go in this direction but that's that's a perfect cherry on the pie for a good workship other possible

toppings on the standard recipe might be to explore the value we will see something more about it in the in the next step uh highlighting the steps that where we are

creating or destroying value problems and opportunities plus our voting is more or less what I just described um emerging boundaries looking for bounded

context uh in a separate taale not everybody cares about your bonded context but also um a a kind of like te

topology style looking for the team boundaries and or matching the team boundaries with the boundary context let me be more more explicit here U during

the workshop you might uh we are not circling the area but we might see this like there are blockers between phases which are giving us a lot of information

there are also horizontal swim lenss and other indicators I can give you something very specific for this for this pattern in the human behavior uh one first ingredient is you might have a

sticky note that has been reported by many people and uh so kind of like duplicates and is also a a key event in in multiple stories but the interesting

p is the expert on the left and the ignorant on the right so the person on the left might say something like an item needs to have an approved description of high quality image before

being added to the catalog while the ignorant to the right and say like just give me valid URL and I'll start tweeting about it do these people need

exactly the same model clearly not they are talking about the same uh product or item on

sale but uh they don't need to have the same level of sophistication in the model dealing with this and of course the ignorant on the right will be the

expert for somebody else Downstream so it's not about judging the uh the the personalities but what I wanted to say

is the moment you add individual to the picture you get a lot of information because people are taking place on the

board around the areas that matters and you will learn that these people are in a given shape you will remember it even if you don't consciously

know and this is one one of the greatest massive stuff you also we remember the shape of of the model we we we had a workship where we were forced to remove everything from the from the wall

after the second day but we had a third day in the wall in in the room and on the third day everybody was pointing to an empty wall remembered that problem over there and everybody understood it

even if the problem was not there because they unconsciously remember that this corner was exactly the place where the ingate problem was so the level of

ner learning was a lot deeper than we actually imagined now the tough question can we do it remotely for a big picture let me show you why this is not

exactly the best solution in in an inperson experience uh we we have some good consequences like we are iners that a stronger peer

pressure I'm not taking out my phone because there are other people in the room looking at me and they don't have phones laptop up uh and also the fact that we are in a low-fi environment

means like well the lap to and the phone are not necessary this is giving us a full Focus the people in the room might get tired together or hungry together

but they are completely focused on one single task this is giving us the possibility of delivering a lot of stuff in in one day not to mention the fact

that is the feeling of it's only today there are people flying back home is giving an extra pressure for a commitment to to finish uh Plus plus the

parallel conversation through boot not a single microphone is giving more bandwidth for conversation and so we're going

faster if we go remotely a few things start to fall apart so there's very limited peer pressure nobody knows if I'm playing Candy Crush on my phone when

somebody else is exploring a boring area uh there's only one conversation possible at the same time and this will slow things down the F that everybody's

doing it remotely and we have one more chance next week or next day so the commitment to finish goes away and uh we

also have to do shorter batches so it's not one single day I granted my um presence for for one day now I can think

about granting my presence for the next session but I'm deciding after what I saw in the first two hours and a half and yeah I saw something but not

groundbreaking while in picture we were seeing big things before people had the

chance to opt out so um let's say I know how a good Workshop should taste like we managed to make it work uh

remotely during the pandemic where there were no Alternatives possible the moment we could reopen the door for a big

picture okay well let's get back to in person it goes a lot better um the the some people were happy I mean the the

workship can still deliver and the alternative is not understanding your business if you have remote um a remote

organization but still I felt like uh no that's not the way to uh um you could do this a lot better so can I make carbonado with cream how many

ingredients can I replace and still call it a carbonara well in carbonara the answer is known but yeah that's more or

less what happened in in general the summar is we learn new facilitation techniques and we improved a lot from the from the very beginning if somebody

is asking us can we facilitate the session uh distributed yes but it's going to take a few days and not only one and and uh we need patience and we

also need a lot more commitment from from the people involved so yeah still the recommendation is run it in

person and uh yes I would stop for a second and get back to a short Q&A maybe

if any question on this um on this part Alberto we got a question forp by the way um so how when you say big picture

how big a picture do you kind of go like I've I've had situations prior where it was you know like it's a really big organization

so how how much of our business process do you kind of capture you know like do you do you draw boundaries around because the you can get the whole

company in the room and put everything in there right that's massive right yeah like where do you draw the boundaries and also second part question is like do you draw timeline boundaries right hey

this is what typically happens for for for doing a job a job to be done kind of thing or is it like multiple quarters because you need to do like your taxes and things like that that could also be

part of your business process like where do you kind of draw the boundaries around this big picture concept I draw the boundaries later

after uh so there there is one natural uh Dimension which is an entire business line because the important area for for

me and for this format is to see where departments are not collaborating well um so I I'm a tremble Seeker I want to

see friction and I don't want to and I don't want to cut the scope up front in this way we we might get in a place

where we we get the illusion of consensus inside the inside the room okay this is everything that happens except it doesn't work because in

another department there is another person setting constraint of what the flow is supposed to be so the the

premise of big picture is let's try to see everything around the single business line and let's try to see the

areas where this illusion of uh uh flow is uh is is Vanishing where where sales is selling more projects than the

development team can actually deliver um what what are you looking at do you see the team capacity or what is the strategy for sales versus the strategy for delivery I'm I'm looking for massive

inconsistencies between departments and to do so I need to look at the whole and um so that that's uh that is the main

idea variation been like we've been working with the um product over platform so maybe having one of two

products that were supposed to work on a shed platform and the timeline were were kind of different it's a little bit bumpier it needs actually more space and

maybe a more sophisticated organization but the whole idea is I want to see the places where the story of uh theart

Department a and the story of Department B are colliding and uh that's why I don't want to cut it of course I can still do the Cut and Run smaller

Workshop but then there there's a political consequence which is if you if you talk with one group and you get a consistent story and then if you talk

with another group and you get another consistent story but then you realize that two stories Don't Hold On Together what then you are the bearer of

An Inconvenient Truth you might get shot as a messenger in the process by putting every same room you're basically forcing them to give you a story and you're

forcing coherence on that story they might not admit oh we never realize that our requirements have been conflicting in the past three years but now it's

there and even more is not your problem anymore now is an organization problem so it I'm I'm saying this because uh many times especially if you collect

information in in five six separate interviews you will have the smell of something is wrong but then you're going to be the politically weak person to

Bear the bad news in this way they will own the inconsistency or yeah the business side I'm I'm I'm using they probably in a wrong way I'm the business

side in many in many situation uh but but we need to there is no way that's one thing to remember there is no way to

align it to business if the business side is not aligned with the business and and that that's the most important thing we are forcing The

Narrative to be coherent then turning it into actionable requirements becomes a lot easier that that is a that is the main reason of course you can make tradeoffs

you can take take a leaders scope but you're giving up something one thing is this feeling of okay I can see the whole thing one thing that happened in my

organization was the moment we explore the whole thing with with new hires um they were the moment I also was giving the

feedback like there's nothing more than this this is all we do and then the new highest were oh why don't you change that why don't you do that instead and

we could start changing improving in less than one month from people being hired I mean you don't have this confidence in many organization and they

were not proposing stupid things they were proposing sensible improvement with the confidence of we are not breaking stuff because you told me there's

nothing connected to this so this change is safe and if you have a young company and they want to grow this is actually a very good

enabler po thank you okay I will take um Sebastian uh question and then move on to uh the next

format uh thank you Alberto uh it's rare that we have the opportunity to talk directly to someone that come up with an idea that's so practical to our day-to-day um work and uh this is a

great opportunity I have been able to experience a few sessions of um event storming either myself leading it or being part of uh the sessions and uh I

also take uh the um the point of a very large space can be a fantastic space to explore um I remember one session with three walls were full of ideas and it

was a fantastic Discovery so many people were able to learn the question I have for you is uh I think that this is a fantastic tool to

explore uh existing processes existing business where I have personally found difficult is when I try to use this tool

as an exploration space when uh we have uh already a working system and we are ideating to bring something new to to to the platform um that

becomes quite an ambiguous Tool uh do will you agree or do you have any advice on how can we adapt this when the space of exploration is still quite

open well it it depends uh it depends on the existing system so the I have two directions one well if you are use it to explore an existing uh Legacy

application you just said system but that that's that's one scenario well then it depends on the application like some software is workflow based and and

in this case I really like the contract of somebody telling a business story from the point of view of the user I need to do this I need to open a booking

I need to uh U add add some information then I need to finalize it and and seeing the contast the contrast with the

with the narrative of the person which is entrapped in the operation provided by the Legacy I mean this can be painful for the brain and I like it because I'm

I'm a bastard probably H but but but also because this this pain for the brain is telling that the design of the legacy is not meeting the expectation of

the user and I would like this to identify the moments where the user is asking a very simple thing and to perform this so you need to do three

four operation which means like you're not capturing the business int on this side so when it when it comes to I I know a few people have been using uh

even storing also to explore the behavior of Legacy system for refactoring I would say it depends on the system if the system is workflow

based then it work really well even even if it feels painful if the system is is just like a risk engine calculator uh maybe end up finding a lot

of vertical events happening in no particular order then it won't exactly work on the other end of this possible answer to your to your question um I've

been using similar techniques not exactly even storming also to model in the large scale stuff which is just

gigantic mess in this case I don't even call it even storming anymore I call it model storming the principles are the same we start with something we ask to

the audience what is the next thing that you would like to see what is the most important detail which is not yet represented and then we pick a color or

anotation trick that has not been used we repeat we go in a given Direction and and then we improve the model layer

after layer I've been doing experiments like this in the the last KO camp in in Austria and um and the goal in this

place in space was let's try to model something like um global warming so very intricated very large scale but still we

could start in a it was an open space session so we clearly could not finish that but there were emerging pattern that we could leverage for a deeper discussion I mean the whole idea of

there's not enough space to discuss this no uh we just need a large space And Then There are ways and pattern to represent it and it takes time but there

is a great advantage in having uh physical visible tangible model to discuss and uh just a set of slightly

different patterns in this in this case and um I will take uh Chris I'm sorry I have a few more on the on the presentation and then uh I will go to

the more practical formats I don't have a question Alberto thank you but I just wanted to throw to the room and see if anyone in Melbourne um was uh had a question if they wanted to come up to the mic and ask it because

we had a couple from people on the remote call but if everyone in the room is good then happy for you to move on so I'll move move on to the to the other part

maybe we have a little bit more uh extra to uh to see now can you see my screen yes again yes okay

so all of these tricks were about modeling without deciding like the the key idea was we have one day we need to visualize all the different point of

view restructure them if we have low hanging FL let's all them but uh uh but that that is not designed I mean I can make it work with so many people because

we're not deciding stuff deciding is the hard part uh even more as a facilitation tool when I see that people are trying to decide something especially before we

with visualizing then I jump in I said like no write it down we'll decide if it's a good idea or not later on so that's a uh that's a pretty good enable

when we go to process modeling uh well that's a different thing it's a little bit more like collaborative design of solution we can still use the process

modeling as a precise investigation tool but we will start using it also as a way to uh discover and design things together uh a few assumption we are

working on on an important problem maybe the one has been voted on the on the workshop in the previous day uh the problem has a solution in in terms of a

process that might be supported by a software but also not necessarily I mean it's very hard to have a process which is not supported by software nowadays

but still I know that some people have been us it even storming just modeling uh processes like uh I know one country in Europe modeled the process of

registering immigrants using even storm and they were happy so y uh the other assumption was the key expert are available business people from the

Departments involved technical expert and uh but also service designer ux designer we would like to uh see different perspective and more or less

we have a scope a little bit better defined so there might be a one probably more way to start the process but we

normally start from one way and the process will end in a stable State represented by one or more domain event like um the payment has been completed

the item has been delivered home and the payment is also visible in in some uh account in in some way just I mean stable State nothing more needs to needs

to happen that's more or less the shape that we are looking at and uh the Dynamics of facilitation

are actually really really different because consensus is is very very expensive to reach especially if the personalities uh in in the team are very

different so the trick for me was like uh how can I make Specialists with different backgrounds cooperate towards a common goal without a danger I mean

there was a story behind it uh from a a tad video from Emy Edmonson about having leaders collaborate together I don't

want to have an emergency I'm happy if there is a sense of corporate urgency about we need to solve this problem uh but I don't want to risk anybody's life

in the process uh so the idea was to well use collabora Cooperative games and that's not an original idea Alister cobor was putting this idea in the this

old but very very influential book A J soft development The Cooperative game and uh and I realized this is just uh not only the way I facilitate

educational Workshop this is also the way I work I mean I'm joining people in in organization we explore the problem

together and then W it's a puzzle I want to cck it I want to win it is something like an escape room uh that I play with

people that I yeah many times just met and which are the rules of this game uh very simple one every PA should

be completed the color grammar should be respected every stakeholder should be reasonably happy and every hotspot that during the discussion should be

addressed let me be more explicit when we say every pass should be completed uh we would like to end uh customer

satisfied happy customer coupon sent and in a stable State uh that is don't please don't look at the colors here

this team was doing a very good job in the modeling the flow uh but the colors might not be the best uh indication for the for the grammar but the grammar will

be be represented here anyway we end in we have many different paths like ending with no further action is

necessary uh and yes that's more or less where we would like to be second thing color grammar is respected the original grammar for the format is actually

coming from this picture which is very old one uh I call it the picture that explains everything and and it and it

just says like there is a cycle that that is repeated in in every in every Workshop more or less and human beings

the little yellow one are making decision based by Real World Experience maybe like I'm I'm hungry and uh and information they could see in a read

model which might be well the um flavors of ice cream available their price and and so on and this way I decide to issue

a command on a syst Sy this command produces an event this event might be associated with value generated or destroyed but this event will also uh

can also be translated in another read model like maybe reducing the availability of of something or increasing the the the balance if I paid something but also an event can activate

a policy triggering another command so this is more or less the the grammar in action still events Hotpot are still there policies with reactive

logic policies are the special guest in this in this grammar systems are very similar to what we just done in the big picture format same for for people now

the green part is the read models the data needed to take a given decision what are you looking at when deciding this and command is the resulting action

put in another way all that's that's more more or less the summary for this information so I'm forcing people to stick to the grammar because that's one

of the rules of the game and this is forcing people to think in a given way and the most annoying most common question is do we always need the policies something that people would

like to skipe and I would say yes my game my rules why are policies important is uh is basically because um

it's a glue between something that happens and the consequence it might be automation it might be uh it might be a

habit in the head of of a person but uh yeah it's just it's just something Lil like between the orange and the blue but is also the piece of your business

process that has the highest probability of being controversial put in another way this is where everybody lies you might get into workshops where one

business Expert is telling you one story then you capture this inside the policy and you talk with another expert telling me a slightly different story and the

reason is policies are the things that evolves when an organization grows the steps are the same the policies will change because of the change of size

change of experience new regulation and so on but this is great we know that the are where we should focus on as detectives is exactly the policy the

other the other reason is ambiguity does not compile we are technical people we need the binary precise information to write

code and policies might be yeah ambigous and controversial so let's make uh our experts be binary when they're telling a

story the other part that I really like about the the policies is uh if captures a spectrum of evolution so uh you might

have habits uh uh policies without an agreement explicit team agreement or implicit uh you might have uh rules uh

regulations uh you might have regulations which are violated in practice um and at a given level you will have Automation and uh that will

might be implemented with listener sagas process manager or whatever can fulfill the the job the the there is a

drift in moving from manual operation towards automation but there is also an opposite drift moving from automated to

human decision which was actually uh a very great card to play in some modeling session not to mention that the policies at all over Place were now ai is playing

a role so you might uh want to pair the human decision maker with an AI and see uh the way it's working what I wanted to highlight is

just uh I'm not g designing a tool which is given a onetoone relationship between the the sticky notes and the code I'm

trying to shape the conversation leading us to an artifa which is robust and coherent and unambiguous and then you have a little bit more freedom for the

implementation it's not about the code is about the conversation mostly um a few extra information about the rules like uh stakeholder should be

happy and U I mean it um I mean we use this little notation on the in the online version we go emojis and these are part of the

official annotation for the uh for the event storming on on mirror type of format uh but the thing that I like is just like uh looking And discussing with

the people in the team well every step can actually create or destroy value but when I say value most of the people

think money and money is um actually kind of boring there are so many other values involved in a in um in a business

process it might be time it might be safety it might be uh reputation happiness engagement and and a few other things like this question is about value

that we create is very open and uh and might actually lead to very interesting discoveries like um or for my company was interesting one day to discover that

uh our partners were not doing it for the money they were doing it for the engagement so we started optimizing for the engagement instead of money then

well actually money improved as a side effect but we were doing the right thing um also um reading suggestion there's a fantastic

book from Nicholas modic the book book is called this is lean and he said something simple and Powerful at the

same time like an organization can only optimize for one thing and it might be revenues it might be happiness it might

be safety it might be uh Effectiveness or whatever but uh mathematically you can only optimize for one C coordinate

so pick yours carefully and don't be ambiguous in the process uh other thing about dealing with value like money is the obvious one but it's a zero some currency meaning

like I'm giving you money I'm losing and you're getting of course I will be getting something like like good other currencies which are more interesting uh Joy reputation satisfaction

engagement uh the other thing that I like these are not Theo SU game anymore like if I am happier maybe I'm making you

happier so we are both happier I'm not giving and taking we are sharing same Dynamic with ideas engagement and creativity so why

not um why is this working well I mean this is a place that was not a blue ocean uh uml activity diagrams BPM and

were already existing notation in this area but uh to be honest uh they were not designed to support a conversation

they were designed to be good precise documentation there was the lightweight version of Let's go to a whiteboard but

in a whiteboard you can only draw many uh rectangles for for a complex process uh but actually never saw so many

conversation happening on and inl uh even more uml was not designed to be interdisciplinary it was the moment

you went uml or bpmn well engineering is taking it the lead um not so interesting uh one thing that we uh discovered quite

recently we've been working a little bit more in the product exploration State um there's a very interesting area

and called continuous Discovery and and this is a place where process modeling G and storming when modeling discussing simul ating behavior of of the user in a

specific area works very very well pretty good cocktail another thing that we can do with this with this format is

extracting user stories and the moment we model the whole flow and we highlight the places where we are generating uh uh

value well it sometimes extracting user stories because 90% of the conversation already happened is just rewriting a sequence of sticky notes from a command

or a triggering event to a stable state with somebody happy uh okay well that's a huge story I just need a label for that the other thing that we do a lot is

um extracting acception test uh while we have a good flow is making yeah looks like we covered all the all the

situation uh not enough Let's uh let's put names let's put numbers let's that into well the good old bdd format given

the past event or the information we can see in the in the read models when we are uh activating this command then we

are expecting this events to happen and this data to be observable in this in this read model uh it's not about the test automation is about challenging

your expert with numbers names and Corner cases absolutely powerful now you can see exactly where this is going to

fit on a visible model uh the final thing is um we using a lot this process modeling format um by taking ideas from

the this this book from Jim Benson the collaboration equation he's actually talking a lot about the Oba concept which is uh the decision room in in Lan

or a large space where we can model everything the way we do this and for this scenario we move to the digital format is by hybrid dating there's

something like little processes this is something uh is is not visible enough uh uh intentionally because uh that that's uh so that you you might not get what

what this is about but uh there's something which is the navigation funnel with screenshot on the company website there are um processes up front like a

me media and and marketing uh deciding for example uh which uh pages to be uh boosted by advertising campaign and

there's a complex funnel for for purchase and then there is the the people manage the area of direct sale post sales and so on H but that's a

place where we where we could have a more sophisticated um conversation and the decision making looking at data looking

at the actual step and building a contradiction uh not not a contradiction collaboration between uh different departments now can we do it remotely

well the answer is clearly yes because this was uh uh happening on uh on mirror but in uh in general the scenario is we

have reduced loss compared to the big picture it's something that happened in a different format it's still slower it's just like we're never going to get

to the same uh speed of modeling it together um it has a heavier Duty on the facilitator you need somebody who is

good as a facilitator and also good with miror and also quick in doing this so maybe will'll need two facilitators splitting the job uh the obvious

Advantage is the artifact could be made available to lar an audience so you can also play say we are running this Workshop maybe not everybody could join but the results are there so a

synchronous contribution theoretically possible in practice not always perfect but uh yeah human touch is kind of still still

missing and uh I would like to get quickly on the on the software part to get to a uh a little more of of Q&A this is uh this is going to be a little bit

uh faster so when talking about the software side now we are looking for information about

how to design our system and with did it in mind how can we make it work we never mentioned the aggregate we're part of the original blog post the solution for

me if we stick to the um uh collaborative games approach well it's just an expansion set we we add

stuff to the uh to the original game and what we add is not so much I mean but many expansion are really really a way

to get more money uh from The Gamers but we add one more extra rule like aggregate should be coherent whatever that

means we had the new color the aggregate is yellow and the detail grammar now looks something like this uh so the

system that we're talking about now is explicitly external that's that's something that we might also called Black Box we are not changing it while the yellow one our component is

something that we can change they are both taking commands and the meeting events sometimes also the policies are something that we might be designing but

we still keep the uh the same uh the same color also we start visualizing the boundaries between the different areas of the software because they might get

relevant and this is one uh one picture from yeah uh an inperson

workshop and uh yes more or less same grammar events hotspot policies external system people read models Commander action then the Aggregates which are

well we might see them as state machine or independent components and um so the problem with modeling

Aggregates is they are still a tricky Beast like most of the people they start uh by naming this like well we are

selling a ticket or we are selling and so this is a ticket and maybe the first thing that you're dealing with is not yet a ticket I'm I'm actually looking a

lot for State machine logic so something that that that is uh uh um yeah pending and then suspended or confirmed or

cancelled um the focus in an even storing session is most on Behavior not data data will fool you in in so many ways and one of the trick is do not fall

in love with the name so you can play Maybe with name lens Aggregates for a while or I put a lot of iph like uh this is the first name that PO in my mind but

I I'm not loading it so I might feel free to uh uh rework on on the name uh please yeah strong recommendation don't fall in

love with your intuition the other thing which is interesting in terms of modeling is um if we stick to the timeline which is helping the business participants of

this of this Workshop we are missing something in terms of uh uh role and responsibility type of design so we might look at signals which are more

like parenthesis like uh there's there's an opening and closing symmetry money flow so this big

uh um little yellow at the beginning and at the end probably is the same thing or needs to be the same thing because maybe

here we are starting an enrollment transaction which is pending for the whole flow uh and needs to be marked as

completed when when we are done same can go for maybe for this policy this external system can be maybe an external payment system and we might need the

payment coordinator uh uh for the for the actual um activity so um that's uh we are giving up something to uh uh make

the model more readable for non- business for non-technical person but we need to discover new patterns in the way uh the other thing is is um people is

talking about stuff but people can only see so much stuff so the business can tell us something very important in terms of the actions they want to

perform on a system in terms of the data they need to see on screen and also in terms of user needs and and so on but uh

our territory is different like how to call how to name the aggregate that's my problem not yours and uh so I'm taking

the business information with um yeah selecting this is a great idea no this is my territory great idea yet still but I will be the one deciding I'm not

letting the business naming my aggregate um but I think that that that is the most important summary like uh if we want to have a format which is inter

disciplinary if you want to talk with business ux service design and Technology every every notation needs to drop down something we are giving up a

little bit a little bit as individuals in order to succeed as a team um also a single tool won't be enough to solve

everything including even storming if you feel like uh okay I'm seeing something but I need another representation well make sure you have a flip chart available nearby make sure

you can go to uml make sure you can also that's graph and data if what you're missing is is business data or a conversion tool for a for a sales funnel

whatever is uh can be added to the learning is great and um y um final uh recommendation for for

modeling is uh this this Hotpot like uh people are are doing a couple of recurring mistakes one they are trying to go for a good solution too early and

it's not going to happen and then they will be open in many many branches this Hotpot is going to be your friend meaning that well when when you're trying to model a system and somebody's

telling you oh wait a minute you also have the returning customer you also have to handle cancellation yes great not now let me finish the Baseline we we

get back at it later this is uh necessary Just Like Oxygen I I mentioned the fact that modeling both in process modeling and in soft

design requires an in interdisciplinary team but it also requires a complementary attitude um so I would

like I would like to have people from the business from uh uh from software design user experience and design but in terms of attitude I would like to have

exactly this type of Behavior so um I have these three behavioral archetypes that I can describe and I guess you have them too

um archetype one is the driller that's a person usually coming up with Alternatives what if scenario Corner cases what if the person buys uh two

tickets and forgets he already bought one uh what what if the the the there's a there's a currency fluctuation uh just during the purchase transaction yeah

all of the possible things like it's great to have one person in the team like that it's very dangerous to have everybody in the team like that without being balanced by other personalities um

I know many of you are going to be Drillers I am one of them as well but uh we need to be balanced we need to have a pragmatic um maybe somebody coming from

the business maybe even better somebody with data somebody telling that's why the numbers the percentages while all of these cor cases are great but 80% of the

customers are single customer buying one they're doing one single purchase so if we can go live with this simple scenario we already covering 80% of the market then next week we go for the other

Alternatives that's fantastic we need to balance these two roles I have no idea which one is right I would like to just the discussion to

happen uh the third role to balance is is the empathic uh the person caring about emotions feeling the person saying like oh yes but if you give them just a

bad exception in the face that's not a good experience we probably need to tell stories in a slightly different way provide a better me message make sure that even a rejection is not going to

turn into losing a customer and so on so one of the little thing is just a big picture needs an explicit facilitation

because 30 people needs to be managed while uh process modeling and software design U they are usually happening in smaller groups we can also make them H

happen in larger groups if needed in this case facilitation is probably needed but if you're happening in smaller groups you just need fluency not

facilitation um and the only thing that I do as a facilitator except obvious controlling the grammar um I'm just balancing the dominating attitude

if everybody's a driller then I try to be the pragmatic or the empathic I'm reading the room and playing the counterbalancing RO uh

role I'm actually gonna uh go quick on this uh strategies and complex moves three possible openings you can start from the beginning you can start from

the end and go back um or you can make a little exploration mess with event but that will need strict boxing um every strategy has advantages and

disadvantages um but uh yeah yeah the OB the weird story is I use all of them and I use all

of them in the same Workshop so you start in one way and then probably we need to see from the end we need to find to use different styles for modeling

inside the same flow so don't it's not going to be a prescription collaborative games their Essence is the rules are clear but then finding the best strategy

is up to the team let's put this to the to a different stage team agreements we are not really good in modeling together maybe we are not even

really good in modeling alone but modeling together makes uh things really really hard that's a that's a picture from an architectural discussion in the

company I used to work uh in in in the past and Architectural discussion could be a really really cruel game like you need to destroy your opponent making uh

making them look stupid so that your ideas is better that's not the way we want to run in with with even storming and uh and one of the most

interesting Tool uh we uh we discovered was uh uh playing with temporary leadership when I say temporary I basically give a set a timer mightbe

four minutes more or less sometimes we get down to three but for the next four minutes one person is in charge and they

can Define what is the working agreement for the this short time slot so something like can you please be quiet I have an idea but I don't want to be

interrupted or I have an idea I don't know how to flash it out can somebody join me as a scribe and uh or yeah we might all try to go and try to solve it

uh or take some ideas from more programming or Ensemble programming uh can you please uh uh tell me what what to do or can you please stop telling me

and just give me only silent feedback with hotspot and and so on or sometimes if finding agreement is really hard uh we can go the split and merge and and

having more teams coming back with with the solution because a Green in 10 might be harder than a green in five and then merging into into two the thing is I

don't know what is the best one but there are this time box idea works very very well can you do it

remotely yes uh we are losing something compared to the big picture you can play uh poker uh at the table and you can play pocker

online uh it's not exactly coll collaborative games a bad example but it's different if you play it at the table there's a lot of extra nuances

which are making the game more uh more interesting um yeah I I would say like is risky on the on the Human Side here

because you might be losing so many nuances and uh I know many many thinkers which are focusing on the solution and completely missing the emotional

feedback from the room and uh if you do it in person there's a there's a joyful side that could be used to recover situation while it might be missing if

you do it online uh but the common tradeoff is that uh uh we're running many small session maybe improving the

original artifact so wrapping it up um I'm I'm I'm using a lot the pizza metaphor because the standard recipe especially for uh for the big picture is

more or less the same tomato mozzarella okay that's a basic then you can add topics um yeah no database tables no

pineapple debatable uh but the thing which is interesting is uh these three formats are very compatible they used to be one thing so we could have a very

business oriented discussion and talking about the big picture then zooming in with process modeling and Zoom even more in getting into Aggregates and even zoom

out the conversation is coherent consider that we can also provide coherence in terms of Team organization value generation and so on so the having

three artifacts uh providing a coherent View and a coherent discussion not bad I I mean I like it second thing with this

three format is about the choice between in person or or remote this depends on your organization if your organization is collocated I would go in person for

big picture and process modeling for soft design I will start in person in person usually it works pretty well if you are focusing on on a set of features

modeling uh based on paper and then when you move it to the other stuff we turn it digital and for for the references if

your organization is natively remote okay gaing for big pictures and then well the other session will be remote and your people will be used to do

remote session and uh yeah and I would say this is the final thing that I wanted to highlight this is a tradeoff thing it's

not as precise as bpmn orl but it's meant to be engaging and to establish a crossover uh collaboration so yeah that's that's what wanted to uh

to say plus references and now I can get back and see you if you're still there now I will pass the ball to um to

crease in terms of timing and thanks Al that's been fantastic I I thought we might start by asking the room in Melbourne if they have any questions first to give them a chance to

go first and then see if anyone on the call has some questions I'm happy to take to spend a bit more time on questions I think that would be really valuable for people so what do you think I don't know SEL is there anyone in

Melbourne that wants to get up and ask a question I think we have a couple on the call no looks like everyone's happy so maybe uh was it I think Daniel did you

have your hand up Daniel m are we do have one from the room okay so someone here have a question yeah yeah you want to just ask the yeah

no thanks um the question is around the granularity of policies and events one of the things I often find is that technical people like myself and others

can drill down too far or it's hard to some know how far to go like with a command it makes sense because you start with well with a command from a user it makes sense because you can start with a real world thing but inside say the Pol

the aggregate or something or whatever that system is it's easy to maybe go through a nested set of more policies more commands more events and sometimes

it's hard to know how deep you go do you have any advice for how to do that in a room with people because sometimes it's hard to control um well most of the time is uh

it's it's not a rule if you if you expecting a rule from my side no you might be disappoint like I use a lot peer pressure like at the big picture

level uh you will clearly have people with different uh uh precision and granularity might I mean you you might have a board members and developers in

the same room and uh and then peer pressure plays a role like um board members they might get with something which is very abstract but not

actionable um while some technical people might say something like a a button click query

executed uh information fetched on Json do we need to care about it well most of the time this doesn't even happen because people are looking to each other

and they they are self-correcting to make sure we are matching the implicit expectation of the audience uh so I

might get there and refine this if uh if what I'm getting is a process with no errors no friction no mistakes then why

did you call me I mean it might be you're putting money into getting an external consultant I need to see problem so maybe we need to go at at

deeper level to make sure that uh the contradiction are visible but um what I wanted to say is uh it's not necessary a

facilitator responsibility it uh that the team will uh will probably kind of self balance because they are trying to find the perfect

perfect balance so I never established this up front because there are so many extra signals leading to this like if you are in big picture you have 12 or 15

meters of modeling space are you just going to give me five abstract event five stickies I mean sometimes it happened and then every body looked at

the at the guy like why is that I mean I gave you 15 Metter for a reason and okay oh so we should we expand well probably

yes and uh other times if we don't have enough space or not enough time it it's kind of self balance and you might need to uh drill in into something the same

self balancing also happens in the in the validation phase like uh somebody knows that there is something wrong at the even level and

the moment you tell the story at the level of granularity which is not making that part visible somebody's gonna

correct it so maybe there's a the the the other little tool here helping you is the implicit sense of urgency of this

workship is happening now we're not going to repeat it tomorrow if there's something that you see that you don't like now now is the time to to stand up

and this is also one of the reason why the imperson version works better because uh when facilitating you can capture also the little signals like uh

the boss is telling a very simplistic description of the flow and then somebody in the audience just had this little way of

looking to the ceiling in this approval I mean you're not stopping the worship and say like oh looks like somebody back there is this disagreeing with the CEO you don't do this you don't expose them

but you would like to have a conversation with with because uh looks like the story was not exactly the same or you might ask explicitly to the

audience like is this the only way to tell a story do we have exception to this scenario is this the only scenario happening and you you you might provide

a safer Place uh for a dissonant voice voice to be heard and also visualized sometimes you can do this in the workshop sometimes you need to be around

and have this extra conversation uh later on I probably told board and you were asking but no that's awesome thanks I'm not sure how I'll convince a board

member to come to the session just so I can get Engineers to shut up but um I'll work on how to phrase that uh there's there's a couple of

extra thing here like um so uh why you can let aage food you might not call a session just like it's there's a free event with sandwiches make sure that

smelling right and you might not be in a room you might be in a place which is accessible that's one more reason to make it happen in in presence but also remotely you might say like I don't know

exactly who to invite but this was the result of our first round of exploration and uh and you can make it public like uh you're going going to be

visibly wrong so a good disclaimer is actually your favorite friend here but saying something like this is the result of the exploration session that we had

on that day if you know that something is missing or wrong or if you have different information feel free to comment or get in touch with us or join

us us on this slack channel so the first invitation is never going to be perfect but can be used as a tool to discover

who the real experts to be involved should be and being visibly wrong is uh challenging on the psychological side

but very interesting as a as a tool to attract people explaining you what you should have done instead so take a long breath you add the notes but then your

model gets better awesome thanks Alberto uh is there anyone online who had any more questions for Alberto yeah

s hey Alberto thanks for the talk um this is just a quick question maybe I'm I am exhibiting the Digger personality here but uh what do we do like because

this is mainly a Time bound event right so I'm wondering how do you estimate how much time time is going to be required for a particular

exploration or an event sing session because you need to at some point of time also estimate like I'm going to need this much time and within this much time

I know I'm going to come up with something which we can take forward and design uh and Implement so I'm just talking about the time no no that that

that's a good thing like um so for big picture um there's a political time boxing which is one day and because I'm I'm asking accessibility to the important people

and it's very hard to ask one day and a half and I mean this this is going to make it harder so the result is since I I have no idea what I'm exploring I I'm I'm not staring the business flow before

the exploration I just know that organizations uh they normally have uh I

mean I'm estimating the size like 10 12 or maybe 15 Metter and then the Team Dynamics to get to a decent uh final state within within one day the thing

that happens is the morning is kind of naturally time bound um and also people getting tired

or hungry plays a role so we stop anyway even if something is not yet happening well the afternoon when we do the validation this is something where where

the team is pushing on the accelerator or on the brakes because they all see that we have 12 meters to cover and we only have two or three hours if

somebody's getting too detailed the other are starting to look like do we really need all of these details and the answer is sometimes we do sometimes we

don't I mean people know what is the most intricated area and you might still use the the uh sticky uh the Hotpot as placeholders like this area has been

skipped for for time reason but it's probably not the most important area for the discussion we're having today so I would say big picture is self-regulated

in terms of Engagement as a consultant we offer some usually something like uh two days plus one so I'm organizing in the travel for three days uh and uh

while depending on what we discover at the end of day one we decide if I should engaged for one more day or two more

days and uh well worst case I'm going to go in a uh in a coffee shop and uh uh write something or code something that

uh that might be the case as a consultant when it when it comes to um design session be them process

modeling or uh or software design well in this case really depends on the complexity of the of the flow uh there's a lot that you can do in terms of

facilitation like uh U half a day is a reasonable amount of time but it might not be the good way to to deal with the

Dynamics like uh in DDD Europe I was uh I was running a workshop format that was called uh rush to the goal and basically

I had less than two hours available and I wanted to model a flow without knowing the domain so in this scenario I I met

in the morning the expert was from a payment company okay we model the payment system in less than two hours I had a room full of 30 35 people and uh

and I facilitated really aggressively like uh basically uh using the paper roll as a as a stick and saying like can you please shut up otherwise I'm not

going to finish but because everybody was a driller they wanted to give me Alternatives and what about about this what about this yes but I want to finish we managed to Baseline the the default

case and the multicurrency purchase and operation which was a good result was actually eight nine meters at

the uh aggregate level but I was facilitating very aggressively like U um normally I would suggest take two half day for the first

exploration like something the afternoon of the first day and the morning of the following one so that you people will have the time to think about the problem

um with a clear mind and also you need to balance extroverts and introvert people and people which are quiet

thinker people who are visual thinker um many times the hard part is dealing with the different personalities so if the problem is tricky I cannot guarantee we

can find a perfect solution in one say usually a decent solution yes but I would leave space for um Collective

thinking solo thinking and then another Collective convergence later on with the possibility of sleeping on the problem

having a shower and and so on