Estimates or No Estimates? • Woody Zuill • YOW! 2017

so I want to make sure that um I'm checking that I got my fly zipped up do I look like um my shirt's properly tucked in or does it looks stupid it's

okay all righty I'm not a very aware person of my own effect and how I look so I need the help from the audience so has everybody filled out that little piece of paper just right on the Post-it

note in one word what do you mean by estimate and then what I'd like to do I guess we'll need to bring it up over here so if you can come and just stick your sticky note on the wall as soon as you have it fast as you can now you

might think you have to hand it down to the end but you'll know if you get up and you're a little bit active throughout this you're going to be more um energized and most of you have

already uh had a couple talks you're not ready for lunch yet which is good and you probably had a soft drink or something or coffee on the way over so hopefully this is good so what I want to do is just gather there and then I'm

going to have my volunteers come up and help me group them thank you if you missed uh uh being able to do this that's okay you can you can um uh come and get a post and on write it down get

your get your answer up on the wall best you can and while you're doing that I'm going to start talking about this this talk is on estimates or no estimates boy this

um this doesn't like to seem to stay connected so I'm going to be real careful there you go please borrow a few more if we need them those of you who have uh not done this yet we still have some Post-it notes left it's useful to

me and I really appreciate this because it's sort of the basis of what this whole talk is about so one thing that that I'm trying to do is get

us on kind of an even understanding across our field about what our estimates and why we do them and things like that so if we kind of have them now um what I'm going to add let me have my

um volunteers assist for a minute here do we have there we go so what I'd like to do is group anything that um that looks

like oh this is not a very sticky wall is it that looks like this over here and then anything like that looks like this over here and it's going to be like this

or this or this over here and then over here stuff like this okay and you get the idea so we

have a grouping there we have a grouping here grouping there a grouping here and then a grouping here and that'll take you a moment but it doesn't take very

long okay so this is a 36 no it's a I've expanded out to 83 Parts I'm going to try and cover them as quick as I can but we'll probably get at least through the first three or four if you want to hear

the other parts just Corner me in the hallway and I'll talk to you about a bunch more stuff I'm a software developer I've been writing software about uh 35 or so years

so I write code and lately I've been going around the world just speaking about these things um so this is where I want to start part first is just about uh a little bit of a a definition of

what we're talking about I'm going to try and make sure we have an understanding of what we're going to talk about today uh so this is no estimates this is a hashtag as well as a concept and I

started with the hashtag in 2012 but it's a placeholder for a much bigger conversation I've been having for about 20 years almost and that's the things that we do in our industry that we just

aren't questioning uh places we work sometimes won't let us question them or we ourselves will block the uh the discussions this about the status quo and the things that we are currently

doing that we just aren't really allowed to talk about so I want to see us in our field as well as in our companies that we can talk about the things that we really currently aren't sort of allowed

to talk about and I take everybody know who that is right does anybody here not know who that is so she's basically hit

the nail on the head right there now that picture of her I think is either a year or so before I was born or within 10 years of after I was born I think maybe it's 1960 but I'm guessing it

might be 1950 which would put it closer to my the date of my birth so anyways this start out as a half #for me because prior to that I was just saying we got

to question this stuff and the hashtag itself I used for this particular tweet in Twitter where I was responding to a question from Ron Jeff we' hadd been

having a discussion about a project that I worked on where we didn't do estimates and at that time I was trying to find more and more people I could talk to about this and so he said well why don't you write that as a blog post and I did

so I didn't invent this to be controversial and I didn't invent the hashtag as such I just used used it to identify a blog post so that I could have a conversation that I was wanting

to have with as many people as possible since then I've had one-on-one conversations with more than several hundred maybe a thousand people about this very topic but the controversy came

for free and that's how I got to come here because people want to hear about this because it seems kind of out there and I like that uh that it's given me an opportunity to meet so many people and I

must say the people in Brisbane are probably the best are in the world I've been all over the place and it's clear you're Superior

here and it's like my grandmother used to say you're my favorite so and you know what that all means so we're discussing a very specific type of

estimates used for very specific purpose so I'm going to have to step off the stage clearly nobody's running the camera so they won't nobody will see me doing this but I'm going to come and see what we've gathered over

here so I started doing this exercise something like uh seven eight years ago and I've done it a thousand times now so let's see

what we've got here the first grouping here is under the title of guess guess guess guess guess guess guess guess prediction what

is a prediction it's a guess okay good thank you now somebody's got calculation what are you calculating when you do an estimate of

the type we're talking about software development was that arrange but how are we C what are we calculating what is it we're using to calculate so the the information that

we've gathered so we can make a guess about the range what are we calculating right but what's the calculation what is it what do we so

calculation 4 * 3 that's a calculation what's a calculation what are we calculating so if we didn't run any roadblocks so I'm going to put this over here somewhere because I'm not sure if

that fits under guess I was trying to force you into say something I say yes that's a guess but apparently that from Bob Martin I'll quote him because he his quote was act I mean his

his talk was actually on no estimates he just didn't know that and I've talked to him about it so Bob please forgive me if this recording gets to the public okay prediction prediction guess

guess guess guess forecast what's a forecast I was in uh I was in Melbourne the other day and they forecast the we to be terrible rain and the terrible rain in the days I was there didn't show

up so it was their best guess maybe about when the rain might show up this grouping here is about what

we're estimating about the size the limit the work the effort the effort the effort the well these are like dysfunctions down here we'll come to

that in a minute not accurate well that's one of the dysfunctions uh over here we have a deadline oh here's cost and budget so I'm going to move cost over here we'll estimate the cost that's

the kind of thing we're estimating so we can come up with a budget right the estimate itself nobody will come to you and say did have you done your budget the last did you get your estimate can you give me your

estimates the budget is what we're going to you do with those estimates maybe oh here's another one sort of a dis so here's the dysfunctional grouping it seems like as I'm here in Australia I'm

getting more and more uh of the dysfunctional descriptions rather than the uh oh here's a good one schedule so it's the schedule it's the budget it's the deadline it's the commitment it's

the control so that's moving into dysfunction we'll talk about that later down here we've got lie lies lie lie

lies so were you copying over after each other uh not accurate dread um BS so thank you for kindly making that the shortened version

because I I can't say that out loud on a video that's going to be on probably on YouTube or something but children might see

it wrong and wrong we had a couple wrongs wow five five lies that might be the most lies I've gotten so

far oh here's one delegation delegation estimates are oh no it's an obligation so I'll put this this is bordering on dysfunctional down here these walls are were not made to accept postage okay

everybody on board with this you understand where I why I did this we're now have a even understanding across this group of what we think estimates are estimates are guesses well let's go

a little further and we'll use that se second this is what you said now I didn't come in beforehand and ask you to do this so I could take this photo so

this photo was done I think this was um I'd have to check it looks like there about 40 of them there or something so this was probably done at a workshop

what did they say estimates are a guess about the time or cost so that we can do some planning same thing I can could have estimated what you had done here within

some degree of accuracy there are things that estimates are good for matter of fact I don't even need to do this talk anymore because I see it every single time I could just stay home why bother

but it's for you not for me okay so here's a group of CTO cios and CEOs that I did a a workshop with or a bit of a talk with look at what they came up

with what do you notice most about it they couldn't quite get to the one word so one of them did yeah promise

it's a promise talk about dysfunction I won't go through all these things but it's because I can't quite see it but it's got stuff like educated guests best guess have you ever had your manager come and say could you make me your

worst guess I need the worst guess that you can make uh the best guess that I'm comfortable with interesting my best current guest those are the kinds of

things they're saying there interesting that we would put it that way an educated and informed one has a scientific guess s SW a swag which means

a scientific wild asked guess so do those really correlate I was in a part of the world where they had wild asses and it was clear to me they could estimate better than I won't go down

that path okay so we are discussing this is what I'm trying to cover you know I'm gonna make my I'm gonna see if I can make my screen a little bit larger here sorry about

this every time I plug this in oops every time I plug this in somewhere my uh the system works slightly differently okay we'll just go on so

we're discussing this is the thing I wanted to try and have something so we can know we're all talking about the same thing because when I usually talk about this people will bring up things like well I need an estimate if I'm going to get a house built or if I'm

going to get buy a fleet of trucks or if I'm going to have my kitchen remodeled there's lots of things that we think estimates give us value for and we want it for that but I'm not going to talk about those and I don't really want to

have a discussion about those because we're really these are different domains I want to stick to the domain that I know something about a little bit about estimates are guesses of time cost

effort complexity that's what you were telling me I'm just gathering this off of what you told me uh regarding a described piece of software if you have ever had somebody go I'm going to need we need a report in

system Can you estimate how much that will cost us have you ever had that I've had people come to me with really large things that you can't possibly it's going to be like between two weeks and

14 years you know it's like you better describe it a little bit uh about a story a project a feature or even a bug fix so that we have something we can

work on and I say to develop which could mean created or worked on so let's start with that and I want to do a little bit of an exercise real quickly I want to come up with somebody just shout out an answer to

this basic thing we're going to do right now I'm going to have a little bit of an estimate activity I want to know three months from today you're going to need to sit

down and write your name on a with a blue fountain pen on a white uh index card how long will that take you less than a second I can write my

name definitely I I used to have to I own several businesses and I had to write all the checks you know and and we'd have all the checks ready and I'd be going through signing each one as fast as I could a second or two would

get it done okay so let's change that slightly right now you need to write your name on a on a white index C with a

blue fountain pen how long is that going to take you you have a white index card with you that won't work do you have a blue fountain pen with you oh that won't work

do you does anybody have a blue fountain pen can I see it that's a silver silver P fountain pen is it a fountain pen it actually has a quill and it the ink you

fill it with ink or is a ball okay and and it but it's a silver fountain pen okay so does anybody have a blue fountain pen so let's have a

look that's black that's a black Fountain pin is that not black I've never met you before right okay this isn't a magic trick that's a black

fountain pen so so you didn't know if I meant black blue ink or blue pen how are you going to make an estimate about this

thing now you just saw a a Jeff's talk this is the things he was talking about I don't know why they have so many redundant talks at this conference if you came and heard Bob

Martin speak last year I I I sat through this whole talk and I kept thinking this should have been called No estimate his talk so this is the thing we have the work time the time to do the actual work when they ask you have you ever

estimated things in ideal days you know I'm talking about ideal days okay but what happens when that particular day you have a a sore tooth and you know you're going to go to the

dentist later and you got to get a filling maybe what happens to you throughout the day things change maybe on the way into to work you got a a speeding T you don't get speeding tiet

you can drive as fast as you want here right in in California they have a speeding limit they call it which is how fast you're you have to drive slower than that you probably have a limit like you have to you can't drive slower than

this you have to go at least this fast is that right here um you get what I'm saying here there's another way to think about this and that's the cycle time or the elapse

time how long will it take from when we start on it to get this work done so these are two different kinds of things there's other ways to think about it does this make sense so a developer so I should ask how many of you are software developers you write code really

frequently like daily in your work when you're not in meetings Okay so most of you so are there managers we have any managers here so okay five four five we

greatly outnumber them there's nothing they can do so um this is what it's about you do not know as a developer I don't know as a developer how much I'll get

interrupted while I'm doing my work how much I'll be dragged off to work on something else have you ever been told by a manager I know what you're working on important but we have something else that's more important we got to go do

that let's do the show of hands things why not it's good to stretch a little bit I've heard this many a time there goes your estimate out the window and then they ask you at the end of the week why didn't you get your thing done it's supposed to be done by the end of the

week yeah I know it's not my fault but I can't say that because you're the manager okay so our working definition is going to be

this an estimate is a guess of the amount of time work time or elapsed time to create a project of feature some bit of work in developing software it's a

guess about how long it's going to take us to do something we all agreed to that already more or less not every one of you said it was a guess but all the best of you did no I'm just kidding that's

just a joke okay so why do we do estimates now I usually do this as an exercise as well but in the constraints of time that we have here I'm just going to kind of tell you some stuff so what we've covered so far is just the idea of

what is an estimate and we've already kind of decided it's some kind of a guess about how long something will take or how much will cost which is usually based on the

time but the results of the estimate are just some weak information that we might be able to use in some way so what do we do with this

information now generally I would want you to think about that for a while what do we want to what are we going to do with this information so we're going to use this to do some kind

of prediction of the future and we're going to use it beyond the estimate itself the estimate's about some usually discreet part of it we're talking about a longer or bigger part of

it so when will it be done how much will it cost uh this is a how much can we get done in this Sprint I see this all the time I'm not sure why we want to know

but that's what a lot of people think they need to know uh how many people when we need to hire or bring in or get off of some other project things like this okay this is the beginning of where

we're headed but why are we doing that what are we going to use this for we want this prediction what are we going to do with these

guesses somebody can tell me we're going to make a decision okay so uh we use these estimates to help us make a decision are we all in agreement on that

is there any other reason you make this the the estimates somebody tell me any other reason because man to manage I love

having my expectations managed it's a it's the one thing I look for most in my life I'm sorry sir it's going to be uh two weeks before your room is ready so

please wait in the lobby um so yeah we use it to so we do the estimates to manage expectations to stop someone from Annoying us so now we are getting back

to this is part of the dysfunction but but why why are we what are we doing yes and pricing the product but that's all about the decision making process all of

this really is but there's another re reason people bring up and that's is is so that we can have a discussion about the thing so we're asking to do an estimate so we can use the side effect of it but why are we having the

discussion anybody want to tell me somebody already said so we can make a decision about something pretty much is going to come down to the

decisions so let's go on what are the decisions should we do project a or Project B that's a common one you'll see in bigger companies um should we do this project

at all uh should we do this feature now they usually say I want to be able to know the cost of each feature and I want to know the value of each feature so I can make some kind of a decision so

we're going to get use the guess about how much it will cost us to do it based against the guess of how much we think value will bring does that seem logical or reasonable to you okay I it's like do

have any of you ever worked with a database you know what I mean by a database nobody knows what a database is has anybody ever worked with a database okay so we take the value out

of fields two out of this one seven out of that one null out of this one 2 * 7 plus null is how much null it is null so unknown times

anything is going to be unknown unknown plus anything is going to be unknown okay which feature should we do first we're trying to prioritize based on that same unknown

information uh do we need to hire more people do we need to to fire some people that's a good decision that's a whole another talk I'd like to give some time you know how how

good is your company at hiring people they make good decisions most of the time so how and how good are they at firing people what I've noticed this a

whole another talk is that um we often end up firing the people who are making it possible for the ones who look like they're doing good to look like they're doing good and then we start learning

who was really keeping this thing running the quiet ones who are helping others that's another talk we'll do that next year okay so let's just go on we don't need all that so we have a

mechanism for making decisions into place in place so I don't I didn't bring a coin with me let me just check my bag because I'd like to demonstrate a really

excellent decision-making mechanism oh yeah I got one here I bought this one at the airport probably do we do this project yes or no

yes we will do it 50% of the time I will be right that's a lot better than what we normally get out of estimates now I'm kind of making a wild

claim there right but that's part of what this is about let's talk about these things okay we want to have a decision-making process

but um the mechanism we have in place isn't necessarily helping us so we believe that we must know something about the time or cost to be able to make the decisions so we're actually

saying our decision-making process requires we have this information do we need the estimates if we had a different way of making

decisions so this is all good it's all based on guesses we're going to make our decision based on guesses but it gets a lot worse it's not just about guesses so

i' like to start with this one first uh this is a famous quote it's hard to find who to attribute it to you've all familiar with it it ain't so much what we don't know that gets us

into trouble but the things we think we know that ain't so you've heard that right and it may be attributed to this guy originally and it probably goes way back before this I

honestly believe it is better to know nothing than to know what ain't so what I've learned out of this this is really important sometime in the 1800s if

you're going to make funny quips you needed a big marvelous mustache you can see by the 30s that's Will Rogers no longer needed a mustache

so there you go that's the big learning today now let's go on so we're back to um we're back to what EML Hopper said

we're stuck in these ways of doing things and we just do them and we get to the point where we're not even willing to question them anymore because it's always a hassle if we bring it up

so have you ever had to make an estimate show of hands wow has anybody here not ever had to make an estimate how many of you have asked someone else for an estimate we're

talking about software the kind we're talking about here okay so it's about even there all righty have you ever gone over the

amount of time that you would estimated so keep your hands up have you do you keep your do you stay within the

range that you've given at least 50% of the time put your hand down if not for 40% of the time you stay within the

range 30% of the time 10% of the time here's a brave soul down here so he you know what he does

he thinks it'll take two weeks so he pads it out to eight weeks is that right okay yes yeah you always have to have someone else to blame that's a good technique

that gets you further through business than anything else always be able to deflect the blame yes you get what I'm talking about here we're not very confident in these

numbers that we give I worked on a job once where the the manager I won't mention any names you probably don't know who these are but since this is being filmed the manager would come to me and ask for my estimates he never

said I want your guesses he just said he wants your estimates so usually I would say but we just don't have enough money I mean enough information to give you a cost for this how long will this take I

can't tell from what the information is so he insisted I do my estimates so he actually put it this way uh Anders gets his estimates cty gets her estim

estimates Greg gets his estimates in I need them for planning I asked him how do they do it I said I don't know a professional software developer a developer knows how to do estimat I actually I didn't tell him but I

actually went and talked to Greg and CTI and and Anders I asked Anders what do you do and he says I just take the list that he gave me I don't even look at the documents and I just make up some numbers real quick because I'm not going

to waste my time with this Greg said I just give I I take the paper and I just write down a bunch of of small numbers because he seems to love small

numbers and kie said I'll quickly look through each uh requirement and try to think about something I did before and how long it took me and I write that

number down that's maybe a little closer to what you might think an estimate is so I went back to my desk I got the paper out and I just quickly wrote down a bunch of numbers that were high and I

took it to the manager and I said here's the numbers I didn't tell them what the others were doing right first of all I want to make that clear I didn't I wasn't going to tell on them and I said here's the numbers I just wrote made

guesses real quick and wrote them down and guess what he said to me he said Thank you thank you and then he had my estimates why did he need

those why couldn't he done them himself like does he just not do anything for a job I would propose this if you're a sof if you're a manager of software development you probably could

estimate any anything you would expect your developers to be able to estimate a lot better than they could why are you asking them they're looking at one little fraction of something and you

seen the whole picture interesting thing to explore some other time oh we're only on to the second part how long do we have for this talk what time we have 10

minutes at least right 20 minutes okay then I'm GNA slow down because I don't have that much to say no I've about got about three days worth

of stuff here so let's see what we can cover so this is where it started from me I'm going to share this story kind of quickly in 1999 I went to work on a contract I'd worked for about three

months on a uh extreme programming effort where the people were following pretty much what agile was to become because agile didn't exist in those days agile came along a couple years later

and I specifically was looking for a large contract to work on where there were a lot of people it was managed by a well-known big management firm inside of a large

manufacturing company that was making a lot of money doing what they were doing and I was looking for that kind of thing to experience this in the big and this is what they had said why I went to the job they were going to do six- week

iterations at least they were doing iterations I was used to daily even at that time um they were going to have retrospectives and they called them Lessons Learned after each six weeks and

inside their increments or their iterations they were going to have three two we blocks one to look at the work decide on what they were going to do for each team and then make their estimates decide what they could fit into the next

two weeks of development that would then go on to two weeks of integration and coding sounds like an okay plan uh it was supposed to be a project that would get done in six no three to six months

with 200 developers how long do you think it actually took so it it took about two years to get it into use and about a year and and a half to um discover that it just

didn't work at all and they couldn't get it to work so that's about what it was so they were good plan being done three to six weeks uh three to six months I figured I could at least see this process for that long so we charged off

to do our work they gave us some instructions on how to do this work uh how to plan things and so on and we charged off to do it and these were mostly young people they were all about

um in their early 20s most of them right out of college 20 to 23 uh I was already something like in my mid to late 40s so I was double the age

of a lot of these people just as I am here today and um the oldest manager in the group was 32 years old he was the head of this whole thing what's wrong

with that picture may I ask experience and what is experience failing failing and having mistakes I had already had great

opportunities in my life and I'd failed over over most of these younger people had not so we charged and did our work we gathered together for our first

retrospective I noticed three important things I've grouped them into two things the first one was the estimates were off and they weren't helping us what do we do about

that and uh I had already seen this pattern but I kept my mouth shut because I was there to learn about how do they do it in the big companies they must know something I don't know and the

other thing was how are we going to get better at it I mean what are we going to do about it and they said we need to get better at the estimates so that was a clue to me that uhoh you know we're

going down a difficult path here the other thing was we didn't really understand or the requirements weren't clear up front and they kept changing

has anybody ever seen that on a project yeah it's good to get stretching stretch it's like harmione yeah there you go okay and you alternate hands and that'll

keep you more active during this period okay how are we to solve for this anybody know how do you solve for this what do you think they thought we needed to

do so we want to have we wanted to control the changes so the the things we needed to learn was to how to better understand the

requirements and uh how to control them so they gave us training in that and of course I was older than the trainers too who probably never actually worked in software development uh we worked hard

to get better at this uh both of these things and then we charged off to do our work and we did look almost exactly like this and off we charge to do our work uh

and then we found at our next Lessons Learned there were two or three things that were still out of whack what do you think they were yes you're right so we don't have

to go through them in such a laborious way so what are we going to do about it we we got some more training and we got we some of the kids started thinking well there must be a better way to do

this and they were blaming themselves why aren't we getting better at this so we charged off to do our work and it looked something like this and everybody was still pretty enthusiastic you got to remember though we're four and a half dang months into this thing almost by

now now we're three months into it uh we had our next lessons learn what do you think we discovered during that

period same slide I saved a lot of work on this so this is what I call a pattern right so we we we charge off do the work find out some things didn't work we figure out how we're going to fix it same thing

over and over same thing over and over this is a pattern right and we got to have names for our pattern because

we're software developers and I decided to call this the cycle of continuous no improvements now I had already seen this continuously

throughout my life I had seen this in many other forms of work is what we think the problem is isn't necessarily what the problem is so what are we going to do about it so I actually stood up in

the meeting and we were we would have these meetings we were in a big room 200 developers and some other support people and we'd all be gathering this information together when we're all done we're going to share it and then people talk about what are we going to do and I

stood up and I said this is a pattern I've seen before and I think we need to entertain this idea is that we are dealing with symptoms versus problems if our estimates are not

working for us and we get better at estimates and they're still not working for us the problem may not be the estimates let's see if we can find out what the what the actual problem is and

I thought I was being pretty wise and I'd put it off for four and a half months before I was willing to even share this info because I didn't want to get canned they were getting rid of people who spoke up I'd already noticed

but maybe I was too scary for them they didn't fire me outright and um but I noticed as I walked down the hallway after that and these were kids I was getting to know we were biking together and we were I was still

young enough to take long hikes with them and take lunch with them and they would bring their problems to me because I was like their their dad who was in some Far Away City from where they were now and so we were getting to know each other at these personal levels and now

nobody would talk to me I would walk down the hallway and they would just part like who is that guy the there's a famous book where there's this story of the guys crossing the water and the

water gets out of his way I saw a movie about it and it was it felt that way to me it's like they go so what I did was I saw one guy Jim coming up and and he was sort of it's like when the wolves are

chasing the deer and they go after the one who's kind of straggling so he was kind of out of the and I put my eyes hey Jim and you know how young people are like he didn't know what to

do and I'm not aist okay I want to make that really clear but they learned after a few experiences I cornered him and I said Jim what's going on why won't anybody talk to me and he said the

managers told us not to talk with you and they were afraid if they were seen talking with me that that would not look good in their report of how they're doing that told me there was a much

bigger problem here and this is what this is the the defining moment in my career from that point forward I started saying there are stuff in this industry we're not really allowed to talk about

let's figure this out let's try to talk about so I started talking about it which often did it with me speaking at user groups and and little conferences and it was as much confrontational as you could imagine

lots of people think we need to have these things and I'll allow for that that's fine by me if you need estimates do your estimates but let's talk about it so I'm not going to go through this

because this is the kind of questions I'm asking if the estimates if the symptom is estimates what's the problem in other words if you got better at estimates and it still didn't solve

the problem it's like if you go to the doctor and you say I've got a sore throat I kind of have a cold I've been sneezing and it gives you something to numb your throat is that going to solve the cold what is that going to

do all it's doing is dealing with a symptom your cold is still there usually when you deal with a

symptom it is actually making things worse because you usually introduce something that is not a solution it's

more like a a uh it's more like a what would be a word of something that it's masking it's hiding the

problem what's that reding red herring it's like it is and we'll go down that for years on some things and that's how our industry's been with this so this is what it's about in my opinion it's about

the unknowns so we'll talk about that a little bit more as we go and we're probably down to 10 minutes now 13 minutes okay so I'll have to

speed up again okay so I'm going to cover a couple questions these are not questions I'm going to give you too many answers for I'm saying we need to be able to question things and here are some kinds of questions we're going to

talk about I like to use this first to let you know I think this is the truth my dad used to tell me something like this and when I found this in a book by Peter block I've kind of adopted it

transformation comes more from pursuing profound questions than seeking practical answers a question should lead to another question which should lead to a better question and lead to a better

question we're after questions we're not after answers my dad I think would take another step and say when you come to a question that resolves itself you've

probably really done a good job of this okay so I like to imagine wonderful and I know that we can get there in baby steps so what are the baby steps is what

I'm looking for and questioning helps me discover those so is what we think we want harming us and maybe diverting us from be able to get what would be

wonderful if we allow that to happen in our lives and in our work so I believe and I'm going to ask it as

a question is the desire for control and certainty worthwhile what is it we want control and certainty of is usually the time that costs the schedule something

like that what if we gave up that desire for control and certainty if we knew it could improve our chance at getting something wonderful that's kind of question I want

to ask what would things have to be like to make it better so what does this have to do with estimates and this is a thing I don't think estimates are actually a problem

they're just a potential symptom so when I see estimates I usually want to know are you guys having problems with these estimates and they almost always tell me

yeah gee you know so this is what this is about when something looks good does it mean it's good I like this picture particularly do you know what this story comes from do

you have this down in Australia what is that story Hansel and grle what is their predicament was it looked kind of good they're lost in the forest you get some

candy it's even advertising free candy like the internet nowadays and they know if they go in there they'll get some free candy but the cost is they have to kill a witch in the meantime so that's a

high price to pay for getting the free candy some times the really good thing and actually it's good to teach your kids to kill people I guess but the cost

of something isn't what we think it's going to be free candy isn't is almost never free so how much work do we have to do before an estimate gives us enough

value to be able to make an estimate have you ever had someone ask you how long will it take you to figure out to estimate how long it will take you to do this

project let's see that show hands I want to see it I've been as good oh more than half of you we want an estimate of how long your estimates will take it's estimates all the way to the bottom like

they say so this is about the sunk cost fallacy right as soon as we start spending money on a project by doing the estimates we're investing in that

project it's the beginning of the sunk cost fallacy the more work we do and we'll look at this in a minute the more likely we want to stick with the thing we've already started down that

path so this is kind of like the agile Manifesto with the sunk cost fallacy we value sticking with what we've already spent money on over cutting our losses

and trying something that might be better so we got to be really careful to watch out for these biases I'm only going to cover a couple biases this isn't a talk on biases so how do you know we was a good decision we're back

to the coin flip you don't know if this was a good decision or not but how do you know with your estimates whether they helped you make a good decision I'm not going to that's sort of rhetorical right now but I'd like to

hear from somebody who's found a way to prove that their estimates actually give them good value I haven't yet found anybody who's made an argument that holds up Beyond one or two questions

with that so this is confirmation bias we start having once we've made a decision we have to protect our decision our career might even depend on it oh but you can always use the it was

someone else's fault Ploy so I'll go along with that that's good learn that kids all righty so we value information that agrees with what we over already believe over opinions and data that

disagree with that belief so we're tricking ourselves we're fooling ourselves are the things we think we want to trap do they Harden

us they inle I lose my is that working yep so this is all about biases I'm not going to spend too much time on biases biases are kind of cool they're kind of like mushrooms you go out one

day oh there's a mushroom and the next day like there's 50 of them it's kind like once people started talking about biases we now got biases you know all over the place that's okay um so just remember they're there look it up

because there is this thing up in the US we got this Wikipedia thing you can just click on stuff and you get to see information that's 100% accurate so you can find this on Wikipedia so time on

time and on budget and I think this is sort of where we're coming to the bottom of this is that a good measurement is on time and on budget a good thing to measure against as to the

success of our project what might be bad with that so I'm just going to talk about a little bit of this because I really believe it's misguiding it's misdirecting us a

lot of people will say the industry's in trouble because 80% of projects fail or whatever and if we're just Jud in that based on did we do it within the time and budget we set that

has nothing to do with success or failure that's a completely arbitrary measurement so what do we have here the first thing I well the only thing I'm going to cover is what I call Creative Accounting the first one is feature Cuts

I won't ask for show hands although probably you should do both hands so you get a stretch out of it yeah so feature Cuts have you ever been on a project with feature Cuts okay now a feature cut I want to make it clear there you go

because that's better let's get both hands up um Fe cuts are where we've worked on something substantially before we've decided not to do it and we decide not to do it because we can't get it to pass the tests and it would be foolish

to for us to send that to our customer so we've already spent money on it but a feature cut if you do a feature cut and you still say we got this done on time that's like yeah I had breakfast but all I was able to do is drink the coffee

because I was rushing out the door this isn't a way to measure success if that's what we've done this is much worse this leads to me this is unethical not being allowed to record hours I won't ask for

a show of hands on that because you're on camera you know but think in your heart have you ever been asked to not record hours to a project that's bordering on ethical and that's giving

us information that will skew for next time oh yeah we got that done in 160 hours when actually it was 240 hours or double or more this is similar to it

people working late at the end of the project to get it done same story we didn't get it done on time and on budget we did Super effort to do that okay so

this is about Choice support of bias and there's all sorts of other things we won't go into biases too much we're sort of coming down the final stretch

here so if the type of decisions we making require that we have estimates maybe we should be doing something else um we value I got to read this

closely because I wrote a lot of words there we value the feeling feeling that we are in control and the belief that we have the power to and influence to change events beyond our control this

has to do with uh gambling a lot they've done a lot of tests that show that when somebody have you ever heard somebody say hey we're really on a roll you know they're playing hey look at that I've got boy I'm on a roll now and everybody's betting along with you

because that guy's hot right now but they're not this is just Randomness we value that over understanding what reality

is okay so this is again it's about the uh the unknowns the unknowns oh my goodness okay I should get a soundtrack for

there okay so getting better at estimates is that actually something worth doing this is the question we have to ask so I will put it this way I have a weight problem I feel I'm in better

shape right now than I have often been in my life and I will eat junk food if it's put in front of me and when I eat junk food I gain weight you remember in data in the old days with reports you

would say garbage in garbage out with me it's like garbage in and garbage stays and it stays to about 50 pounds above this until I start going none of

my clothes fits very well anymore and I've actually had to size up and then it finally occurs to me I better do something about it this is the thing if the estimates aren't serving us well is

getting better at estimates the only possible path what would be a way to to do better with the junk food chew it more thoroughly

swallow it faster mix it with other food um read the package carefully before eating it what can you do with junk food that

would help this situation don't have it I mean is there any other solution you can replace it with less junky junk food and that would

be a alternative to eating the same junk food to me it's that easy now I'm not saying estimates are the same thing as junk food

no that is what I'm saying okay so better can mean Alternatives and I'm going to share one little bit and then we're kind of done for the

day will we have time for questions maybe one but I won't go very fast so we'll see so this is something I learned when I was quite young and a Gentleman who was uh I had asked to teach me some stuff said I can teach you

that stuff but this is way more important and I really like what this says the B basic idea here is that if you set things up for Success you can get good things happening if you just

push for Success like there's no motivational raw raw speech that you can give that's as good as being ready for the eventuality of whatever is coming up for you so I like this idea the object

isn't to create art or the object isn't to create great software it's to be in that wonderful state which makes great software inevitable so let's close out with this

what should we be certain about with that in mind let's be certain we have an environment where Discovery and and Innovation and creativity is available

to us so that the outcomes of greatness of good things we want are inevitable rather than what we planned to do now it's going to be the wonderful things we discovered along the way which are

generally going to be a lot better so what should we control this is a tricky one but I'm going to answer it this way for today let's control our urge to control things

then we just have to control one thing let's control that urge to control things and we'll cap it off with this from barsness the fear of losing control is a big barrier to

change but the control is only an illusion of control most of the time but the fear is real and we can't ignore that this is why these things

stay in our systems for so long we will take the replacement the placebo for control is having those estimates and that planning that we all know is just guesses and that gives us a feeling of

control you know and I'm not suggesting this is a problem that your bosses have we have it ourselves too just our bosses have it a great deal more you know we can always blame somebody else there we

go I think we've covered enough stuff and I'll take a question or two is that how this works or or we run out of time there's one so that is my two we certification

course no I'm just kidding just say yes no um so in a nutshell uh that's been this that's the track I've been on and uh I have four or

five things that I use the first is we need to learn to work well together the second one is the people doing the work can figure out how to make this better so it's their job to do it we need to

recognize that that's what's important is is allowing them everybody here is brilliant you wouldn't be in this industry if you didn't have some form of thinking that you could use and yet we tell they come in and we tell them they

have we they have to work the way we tell them to work another thing is I think we need to study together we need to be together in a helpful environment that allows us to get past this kind of

weird competitive political thing we have in our companies and we need to get good at getting results really good results from our retrospectives and that I close off with the idea of the

retrospective should be about turning up the good and not trying to solve problems so that's my own little that I've been exercising for about 10 years and I've had some good success with it

some of you'll know me for Mob programming and that was my guidelines that I used when we stumbled upon through that process the idea of mob programming so um that's that's as close

as I could come I can't give any advice on that but I can tell you that's the way I'm thinking right now I'd like to see the whole world getting along and working well together and if I can get that just on a person byerson basis

in my lifetime that's I'm good enough with that so thank you very much everybody so before we're completely shut off while they're preparing the stage I want to show something if it's

okay oh man 63 so did you guys tweet a lot while were you just tweeting while I was doing this son of a okay so let's just do uh

Yao 17 Karan if it doesn't come up that's okay oh there it is good so I get to show you this I don't know if you'll hear the

music but I made this um I made this after our our speaker's Boat Trip yesterday I wish we could hear the music but I I don't hear any music don't worry about

it it's loading the music just makes it fabulous but this is just a little screen slide show of um of our adventure

yesterday it got a little rainy they tell me that happens every day here and you recognize some of the faces here and this only takes about another

30 seconds or less so you can find this on in Twitter if you use TW you have Twitter in Australia right you got Wikipedia you got

Google excellent there you go so thank you very much I had a great [Applause] time vote green vote green vote green as

you leave please I want to come back some other year thank you so much