Abonniere den Podcast

LINKS

Episode 11 - How Metrics Are Abused in Agile

Transcript

(disclaimer: may contain unintentionally confusing, inaccurate and/or amusing transcription errors)

foreign

 

sessions podcast here in Dallas I am q and that’s your clue man go ahead

 

thanks for making it so subtle hi everyone my name is carry on and I’m coming to you from Berlin and Q what is it you want to talk about today we’re just complaining about uh your favorite metric yeah I want to talk about metrics today and we can start a discussion with my pet peeve and you know I have over 500 of them which makes good podcast topics we already talked about several of them and all the the coaches and all that stuff but I start with my biggest pet peeve and then that’s uh let’s go deeper into metrics overall which is velocity which is misused misunderstood misunderstood and weaponized all the time so let’s start with philosophy all right okay so what’s your problem with it because actually I like velocity in quite a bit well my problem with that is not that you like it you know my problem is everybody misuses it right so uh I see leadership setting targets for velocity I see people say well you know what um we did 10 things last last Sprint we have to do at least 10 on this and maybe you should be doing 12 right and all those things are actually wishful thinking right because there are things that impact velocity there’s people going on vacation holidays and Etc and there is this uh very dangerous assumption that velocity should be always be going up which is unrealistic right because it goes up and down like stocks as a matter of tracking the trends so I my pet peeve is the misusage or velocity okay so you promise not actually with velocity per se but how people misuse it um I can get behind that but like should we be focusing more on stopping people from misusing it then yes we should but the problem is velocity is just one of the metrics that’s misused metrics are overall are misused you see the way I see metrics metrics exist to start a conversation hey did you see the numbers what do they tell you right and not to set targets every time you set a target for a metric a big it instantly becomes at bad measure right so we go back a few years oh we we have too many bugs so let’s incentivize people to find bugs the more bugs they find the better right what is that going to create

 

[ __ ] oh it’s just oh it’s just totally the mind when you said that because I used to work at a customer and I actually checked how many times the phone rang before you picked it up and you can imagine what kind of afghanist that created because everybody suddenly was focused on picking up the telephone as quickly as possible instead of doing my job so I get totally lost the way out I know that’s exactly my point is the misusage of metrics right yeah so we just came from the scrum sessions and we’re talking about Monte Carlo simulation right which is one of my favorite things and maybe you will do a podcast on that but uh if you’re interested look Monte Carlo simulation on the web it’s a great way to forecast stuff it’s a very good thing but the problem another problem with metrics overall is are we looking at one metric only

 

because I had a client once I kid you not I wish I could tell you who they are but of course I can’t they had 79 7 9 79 metrics on their dashboard

 

yeah so what we created with that well I saw some VP once told me and said hey um if it’s not green I pay attention if it’s green I don’t care I said well everything is green so what that tells you everything is great I said no it’s rock I mean just because it’s green doesn’t mean that’s great right the same client client they have this thing called commitment reliability metric that’s already sounds ominous it already sounds bad no you have no idea so this is what the genius is created right Geniuses air quotes tenant if a team said we’re gonna do 20 things to spread it has to be at the end of the Sprint has to be between -10 percent or plus 10 percent what else you get in trouble you’re gonna whip your back right so it was a punitive matter so at the time they had over a hundred teams on the campus you know and it was there for about a month and I I heard the stuff I had the same reaction that you did oh my God right and then I start noticing that not a single team on the campus ever ever got out of the minus or plus 10 percent well let’s face it there is some magic involved in there and right either these guys are geniuses on forecasting stuff in that case they should be managing my stock portfolio right on all my investments or there is something funky going on and it took a while to figure out until one of the sky Masters once told me Q we figured out that they run the batch job that calculates this stuff three days after the end of the Sprint so we go into our tomb right and we just fudge the numbers uh

 

see see where my pet peeves with metrics are going with this one they were they were they were punitive which is a No-No second is they are looking only at one thing and then they had those 79 things that they only look if it was Green so there’s a lot of mistakes in there first of all is the punitive part there’s no question about this yeah the second one is you never go after an index in isolation then if you want to talk velocity right so you go there and say Hey listen everybody is going up on velocity great we keep growing Veloster we’re really good at this okay then you look for instance value delivered well honestly our value deliver is not that great you know we don’t deliver a lot of value we so what tells you well people picking up the low hanging fruits right stuff that’s easy to do so there are cruel points for velocity and then you get a team that’s actually going up on a velocity and going up on the value and then you look well people are quitting like crazy man they want to get out of their team and so those are the things that drive me bonkers about metrics is the misusage is not having the conversations it’s setting targets that are actually detractive from everything else and overall not having a clue that you have to look at several indexes to have a correct picture of that yeah and actually I think you make up points there and this is um why I’m not the biggest fan of metrics right I think they’re important and for example I really do like philosophy but I totally agree it’s just a starting point for a conversation and also and you have to make sure that you see the whole system with whatever metrics you choose right and you can choose way too many you can choose way too little and you can set the the boundaries or for example where something turns from green to yellow in a really arbitrary or even a political um way so that it doesn’t make any sense anymore or you’re achieving your political goals but this the metrics aren’t telling you anything anymore but I’m also noticing something I’m not even sure if I can put it into words it’s like there’s a certain kind of manager that like really over emphasizes kpis right and immediately if I hear just the term kpis immediately the hair on my neck stands up because um well what does it even mean right so yes I understand that you want to measure some metrics cool and everybody wants that that’s the slightest this mid analytic um and there’s always of course like the the agile metrics right such as velocity or such as wait time right stuff that is very specific to adjunct but then there are of course operational kpis operational metrics um that are specific to the business and what’s really throwing me on off often is that people like either want to make a connection between them or even before they start an agile transformation they want to have a commitment to okay so if we do Agile then we’re expecting our operational metrics to go up by these percentage points right and you’re you’re there and you’re like well actually I don’t know you guys I don’t know how well you’re gonna do the transformation you’re asking me to commit to a an outcome ahead of time without me even knowing anything about how it’s going to work how much resistance there’s going to be and I think like I mean If we’re honest aren’t metrics almost always abused and misused yeah oh because yeah they are but it’s important to have them right sure you cannot afford not to track things you know like we talked in the sessions just a few minutes ago what process efficiency right but that is an indicator that AR process efficiencies I don’t know it’s four percent really low right a we have to have a conversation what is wrong that word process efficiency is four percent what we need to change about it not going there and say we have to move from four percent to eighty percent and set a goal right set a Target no we have to understand this why it’s so low and then we can start to work on that so another thing that I find really badly matrixes I’ll give an example you have a fridge in your house right I’m quite sure you have so when you go open your fridge and you’re gonna get food out of your fridge you look at the expiration date don’t you I sometimes yeah well I hope so because if you don’t you’re gonna end up sick really quick right because stuff spoils no metrics need an espira expiration date too right for instance you have a problem let’s say it takes too long to answer the phone right so you can for instance say a for the next month we have to figure out what you’re gonna do to make the answering time for the phone BET right and you measure that over a month and then you say hey it’s it got a lot better okay you know what forget this metric let’s go look at something else that needs Improvement so that’s what I call shelf life now if you keep tracking the time that takes people to answer the phone between now and five years from now like you mentioned nobody pays attention to the rest of the stuff then right so like food in your fridge right what’s the expiration date for the metric because the idea of having metrics is to influence behavior

 

yeah once you achieve a certain Behavior move on and go influence another Behavior

 

now I like where you’re going with this and I think so far away are in complete agreement so we think that metrics are important we think that you shouldn’t just take one we should we think we you shouldn’t take too many right if you have 500 so they become meaningless we think they should be a starting point for a conversation but I really do wonder and this is what in the end has turned me kind of off of metrics I you know I have a business background so of course I like metrics I like to look at the numbers however I find that so many managers just completely abuse them like first of all they don’t understand them even though their metrics then they abuse them so much that it has completely turned me off because it it seems to be more of like a coaching problem right the problem is not the metrics the problem is how people deal with them especially in non-agile organizations in traditional organizations because they’re just a weapon to be um used against political enemies they’re just something that you justify your results with right it seems like they’re almost completely political when they could be so much more well in all this reminds me the megapixel Wars of the cell phones right because this one has 200 megapixel cameras the big thing that came up this year okay does that make the picture much better or not right or you’re just doing what we call a vanity metric right another problem that is often said in Matrix is it’s a vanity metric right you’re just boasting those for instance if you look at classical product management we have what we call The Innovation accounting AKA pirate metrics right so you have acquisition awareness and activation right those are the first three in the pirate metrics now if you look at that let’s say you just came up with a new product and you say I had three million sign ups for our product you know it’s a so people wanna hear about it great right after an even things now six months later how many people who actually sign up to know about your product actually activated how many people actually bought it right and this is the same thing with anything in life so let’s say you were in the I.T department and you say hey you know what my uh lead time to value my cycle time whatever I want to call it it’s now five days it used to be a month great uh did you notice who many bugs come out every time you release something so you are boasting about something but there is an extract structural problem elsewhere that you’re not looking at yeah and uh one of the things I always tell people especially in it environment is like you guys have to start thinking as product managers right because in product management we have this thing called the North Star the primary metric now many times I seen the North Star being misused because during the product cycle the north star changes right product is brand new how many people are talking about it after a while how many people are buying it right and after after that I mean a couple years like how many people are upgrading to the new version right and then how many people are student using it because if there isn’t a lot we have to kill it right and move on and it’s the same thing so if I’m boasting that my cycle time got so much shorter and my quality sucks right and again connect the dots you cannot you look at metrics in isolation so you can have a primary metric on North Star which has a a it’s fundamental to know hey where we’re going with something but you cannot go into vanity mode yeah hey I have this big numbers right yeah and so maybe in another episode we can get into more specific metrics on the product on the side of the delivery site but I think we’re getting close to something and I apologize for my negativity right because you can sense my frustration with metrics but I think you’re making a very good point that like whatever metrics you’re checking I have to be balanced in a way right you can’t just look at one part of the equation that’ll leave everything else out it has to be some balance they have to be a couple of different lenses that make sense when taking in in total right you can’t just say oh we’ll focus on this but leave everything out there has to be some um some some balance in it yeah exactly and I I always repeat this do not miss the point is to start a conversation the reason that we measure things is to start a conversation which goes back to my point and I said well if it’s everything is green I don’t need to do anything about it well just because things are good it doesn’t mean that I’m gonna stay that way quite the opposite yeah right it happens things are going to start failing or you are not improving them so I often wear the things that’s another one of my 500 plus pet peeves and those people say well if it’s not broke don’t fix it right if ain’t broke don’t fix it well that’s a really I’m gonna be very blunt that’s a really dumb way of thinking because a when we were leaving caves why are you gonna move elsewhere right the cave is nice it’s a bit humid sometimes and sometimes a bear comes in and kills one of those but for the most part is a great thing why I’m gonna build an artificial cave elsewhere right we’re gonna why are you gonna get a bunch of people together and call it a city or Village or whatever right so we will never evolve if we had this kind of mentality yeah and you know part of my frustration comes also from the fact that if you manage to use metrics well they can be such a superpower right if you get the transparency if you get the analytics so for example it’s um one of my customers a team has built this really nice interactive dashboard where they can see okay so which customers are using the product how right they can really analyze everything and you know they they can say Okay somebody has bought this that they have they are not really using it so dear salesperson can you reach out to them and see what’s hindering them right and so you immediately then get the feedback about some problems or you can get a follow-up with a customer to see is everything working all right and so this really becomes a superpower right because you create more feedback loops which then feedback into your scrum processing which makes you accelerate even further and it’s just so frustrating to see what’s actually possible if you respect those kind of those those several points that we’ve been talking about because then you create a superpower you can just accelerate your system by so much because you have the right information at the right moment and can make good decisions based on that while so many people just waste this opportunity by playing political games by misusing them by not knowing how to use them and it’s such a waste in the end yeah and uh you see I agree with you being negative sometimes and frustrated because this means usage right and the other one is sometimes we actually don’t necessarily misuse them but we don’t Implement them correctly right then uh a few months ago I was talking to a friend of mine and he said Q I don’t like cool KRS objectives and key results right I said why not well because it only works if people implemented them properly and I said well that is true of everything including scrum including including combat including of anything anything that’s not implemented properly it’s gonna have a negative result right so I see people sometimes complain about scrum some people complain about this some people complain about metrics but the point of the matter is you have to understand if something is a fit for purpose you have to be careful not to set targets because then everybody ignores everything else and you have to think about the shelf life of that how long you’re gonna measure that right because again is about having the conversation what can we do to improve something and that changes from time to time and not pretend that hey everything is great you know I don’t need to change anything right then there is no innovation I I think one of the things that is very dangerous with metrics is we go into process Improvement mode a Q you’re saying you don’t like process Improvement I said no the the thing I saw in many Industries and clients I had in the past is they get so into improving improving improving improving improving what they have and they forget to inject Innovation into the process as a so we have to be very careful that again we don’t get into a vanity matter can say hey uh our cycle time now is a day and used to be three months five years ago yeah but are you innovating that process no no we’re with one day is the maximum we can do yes with this process right now what about if you disrupt that you try something completely different something more radical right so Innovation it’s key to improvement improvement is not about to make the number smaller or bigger whatever the situation may be it’s about using Innovation to shake things down in order to get things a bit out of the status quo and then use the numbers to prove that to be true or not yeah so this really sounds like Target fixation this concept I think it was in in the second world war where fighter pilots would just Ram into buildings that they were trying to shoot at from the sky because they were so fixated on the Target that they uh they forgot to disconnect and just crashed into the target if therefore trying to hit yeah very stupid idea and I can really see how you can get into that yeah and uh it’s very easy to get into that yeah awesome so I think that’s it for metrics you know we should come with another topic next month you know maybe we’ll talk more about the product product management side of metrics but uh you know since you always can play you close the podcast today my friend

 

let’s watch again sling uh we hope um this has been valuable to you to talk a little bit about the metrics about the frustration that can come if you misuse them you gave us a couple of really good pointers of how to do it properly like balance them out and so on and yeah maybe in one of the following podcasts we can dive down a little bit deep on specific metrics that we think are helpful in an agile context and it can help you um create better awareness and information in your team in your organization so this is me Carry On from Belen signing off from Dallas signing off see you guys soon bye-bye bye bye