Lessons

Lesson Info

Improving Systems

We're in the homestretch. This is the last section we have learned a lot about systems today what they are, how they work, the moving parts, what to measure, how to measure, how to make sure you're getting good information amount about are out of the measurements that you're doing awesome. So now it's let's roll up our sleeves and actually do something with all of this stuff, and so improving systems is really important in the context of business because building systems allows you to do more work allows you to do better work to do it faster, to do it more efficiently to spend less time doing the routine parts of business and more time doing the things that really provide a lot of value allow you to be creative, allow you to build new things and expand and all of those wonderful things that we want to do in business what allows you the time and energy to actually do that is building systems, so understanding how to construct a system that actually does saves you time saves you money sa...

ves you energy is super important like lies when you build a system just like there's there's no such thing as a business in stasis, right? Things are changing all the time the environment is changing all the time, you can't just build a system want an expected to stay the same forever more every system that you build, you need to know how to change it over time, toe update it, to adapt it, to improve it, to make sure it's still delivering the results that you wanted to learn. So there's a tricky part, though, and that comes from the complexity that we have been talking about in the understanding and measurement systems there's, a lot of moving parts, thing's flying everywhere, people do in different things, sometimes things that you can't anticipate, and so it's important to understand that any time you make a change to a complex system that you think is going to be an improvement, there is a non zero risk that that change is actually going to make things worse, and not better. And in the worst case, blow up something important that you rely on that cask aged the rest of the system, so when we're improving systems, we can generate some awesome, awesome results from doing that, but we also need to do that in an intelligent way, in a way that's likely to generate this great improvement for us and minimize the possibility that something unexpected is going. So the first thing that we're going to talk about, and this ties in with our our understanding of human psychology, all the things that we talked about back half of of yesterday and early today is there's there's a bit of a cognitive bias here that we need to remind ourselves before we roll up our sleeves and dive into changing things and the spices called intervention by us as a rule it is uncomfortable when we perceive something wrong there's something that needs to be improved it's uncomfortable for us to do nothing it's always more comfortable for us to immediately jump in and fix the problem even if the activity of jumping in and fixing the problem makes things worse instead of better so this is actually one of the things that I've been learning over the past couple years with my daughter she is now two and a half and she's getting to the age where she's experimenting with lots of things she's collecting a lot of information about the world around her and sometimes the experiments that she runs don't go particularly well she gets in trouble, she gets hurt she gets scared and so my impulse as a parent is oh my gosh my baby is in pain I need to jump in and do something when really the very best thing that I can do as a parent if I want my daughter to be happy and healthy for for a very long time to come is as long as the mistake is not going to seriously injure or kill her I need to let her experiment and manage that hard right it really is but by intervening by always swooping in and trying to fix things whenever I perceive something's wrong, I'm actually doing her a disservice because intervening does not make that whole process better the same thing happens same type of situation when you're a leader or you're a manager and one of the people you're responsible for does something and they screw up and you can see the screw up happening we're about to happen there's this incredible tendency of wanting to swoop in and prevent sometimes often that's not the most constructive thing that you can do it's okay little fairy failures little experiments perfectly fine perfectly fine so there's a little check step before we jump in to fix something that's sub optimal which is really just asking the question is this critical? Is this something that we definitely need to do something or not and if not restraining yourself this is where the inhibition thing that we talked about earlier comes in right? You can notice something and maybe intervention is not the best thing if it's not you want to make sure that an interventions absolutely necessary before you jump in does that make sense now when we think about improving a system a word that often comes to mind is this word optimization when optimize the system we want to make it do better we want to make it do faster and it's important to understand what optimization really means and so optimization come in one of two things you're either maximizing the output of a system in terms of what it's creating, what it's producing or you're minimizing, the input that it takes to run the system, and so for example, it is a for this business system. I want to maximize the amount of profit, the amount of cash that this business throws off that's an optimization problem, or you can say I want to minimize and input of the business let's say the amount of money that you spend in advertising every month, I want to minimize the input that the system requires to run right there, both optimization problem. Now the tricky thing is, ideally, we want to make everything as awesome as we possibly can all at the same time, and that feels like optimization, right? It's just generalized, making things better. But when we try to optimize mohr than one thing at a time, what you find is you're no longer optimizing your making trade offs because those things might affect each other. So you may get to choose between the amount of money you spend on advertising every month and the amount of profitability of that the business contains. You may be able to choose which one is more important, but you may not be able to do both at the same time. And so any time you try to control more than one variable in this process, inevitably you start making tradeoffs between them so when you're in the process of optimizing something, you have to be really, really clear what is the single most important thing for you right now to optimize if your cash flow is a concern, then yes, limiting the amount of money you spend every month on advertising may be the most important thing to ensure the survival of the business right now perfectly acceptable goal but by cutting your advertising you may be getting less business and your profitability goes down that month that may be ok or vice versa. If you want to maximize the amount of profit that you make, you may have to quadruple your advertising budget and that's ok, because it's giving you the result that you want it doesn't make sense one and only one thing at a time can you optimize now? Sometimes when we think about of optimizing a system changing things to make things better, we get this mental picture of rearranging things to produce a particular result, we're doing new stuff it's important to recognize that one way a very important play to optimize the system is rearranging a whole bunch of stuff and the output of that system is exactly the same when you're done as it was the beginning so remember we were talking about the car factory and in the constraints section which rearranged the entire process to make it more efficient to put a windshield in the car the output of that system was still a car at the end but the output didn't change but the process that that the car went through in order to get that end result may have changed dramatically this is an idea called re factoring changing a system to improve efficiency without improving output sometimes everything that you're doing in terms of the output is great needs to say the same but if you tweak how you're doing it it becomes way way more effective uh in order to rot requires less time less people less energy less resource is to accomplish the same end result that's an awesome thing right? You get the same thing for less those savings really really add up over time so a lot of times it's valuable to take a look at the system and you say how can we make this more efficient how could we make it less costly how can we make it run better and faster and so I find myself going back and forth and what we'll talk about this here in a minute but sometimes you're working on the system to produce more output and the building that you do in order to get the more output part to get the better result more customers higher profitability whatever the building that you do there you're doing a lot of experimentation, some of it works some but doesn't, but the things that don't work kind of leave after effects, right? You've changed the system there's kind of parts there that really aren't doing anything any more there's kind of a hand out re factoring is the process of rearranging the system to pull all of the things that are unnecessary get all that unnecessary crap out of the system. So you're not wasting time and energy continuing to do it right that's the value getting rid of all the waste to make your system run faster and more efficiently, and we have a good, good example of something that they've done to re factor either your your daily life, your work, the business that you run in examples one of the things that I did, it was for a long time, just like many folks, I want to have a website on the internet. I wanted a system to handle that, and I used a system called wordpress. It was great, I got my side up very, very quickly, it was easy to use all of that stuff, but it wasn't very wasn't a very efficient system for me to run because I would have to spend time updating it, it would get slow sometimes, and it wouldn't do what I wanted it to do, and so after a certain point like I'm spending so much time maintaining this system to keep my website online, I want to do something else something that doesn't require whole lot aa lot of maintenance and so I decided that I was going to moved to a different system and use a program called jekyll it's on my computer now so my whole sites on my computer I make a change to my site I run this program and it creates a website that looks exactly like the site usedto look on wordpress it's just a static file and then I uploaded to a web server and I'm done a completely different way of publishing online that produces literally identical output between the two systems, but one system allows me to do the same thing with a lot less work. It was an example of re factoring in a way that probably saved me hundreds of hours other examples in fact, if example like so in my clinic, we do all appointments by phone or we used to and we still do offer that, but a lot of our patients were like, we don't want to have to call you can't we just go online and make an appointment? So the end result is that we get an appointment but now they just it takes out that person factor the fund ask she's still there and doing other things but they just go online and book appointments. I love it. Yeah, that's great other examples like personal life before, like when I woke up in the morning. Yeah, I like check my phone to say what I want for breakfast, just like I think a lot about all these things, and now I've created a system where I know exactly what I'll do, like steps, but not make up. So I'm not thinking about any of that. There you go and re fact, every once in a while you can, you can see. Ok, what is the series of things that I'm doing in the morning? And is there another way of getting me a really great results doing the same thing? Right? It's great. Okay, that's, re factoring, you can change how the system works, the exact same output just a more efficient to do it that way very often, when when you analyze the system, you'll notice that some parts of the system create a lot of value, and some parts of the system don't create very much value at all. What, you're still doing them anyway, but this is an idea called the critical few, which also has lots of other names thie eighty twenty rule the prey does law all sorts of things that, you know, eighty percent of whatever it is that you're doing, twenty percent of what you're doing provides eighty percent of the value those types of ideas when you really look at what you're doing, you find that all of the things that you're spending time and attention on a very small number of things or providing most of the value most of the end result so one of the easiest ways that you can improve any system is to five do the measurement first, right? We've already done the analysis of the system and then you say, ok, these few things that are providing most of the result, I'm going to take those I'm gonna focus on it, I'm going to do that most of the time, that's where my energy goes and the eighty percent of things that don't really do anything for me at all I'm just going to get rid of it stopped doing it so telling the story of timber all earlier in terms of segmenting is customer base and saying eighty percent of the results that I want to get in terms of revenue, profit enjoyment, working with my customers comes from these types of clients, so I'm going to focus on these types of clients and all other clients can it very respectfully and very kindly find another web designer to work with right, and by doing that he was able to maximize the amount of time he spent working with his ideal customers it's a great example of using the critical few to improve his entire business system. So you find this type of discontinuity pretty much everywhere, so if you open your closet and look at your clothes, you will find that you wear about twenty percent of your clothes about eighty percent of the time. Awesome if you need to save space in your closet. It's really, really easy to figure out the things you wear all the time and the things you haven't worn in years and just get rid of the stuff you know you no longer wear likewise with your, uh, contacts list in your cell phone right there, a couple of people you probably call all the time, and there are probably hundreds of people that you haven't closed in years, so if you're cleaning up it's really clear which ones you're probably eliminate, that focus on the things that give you most results as you are in the process of analyzing a system with an eye of improving it and looking for things to optimize optimization is really great, so you can say, ok, this is what I'm working on, I want to let's say, maximize the amount of business that a new customer brings in in terms of dollars, I'm gonna optimize that that's what we're looking for you can look at which types of customers spend a lot with you, which types of customers don't spend a lot with you, which types of customers suck up all of your time and attention and don't spend very much and so they're wildly unprofitable, and you can use the critical few to get rid of the bad customers and maximize the good customers and spend all of your time serving your ideal clients. Yea, that only works up into a certain point. You can't do that forever, and the point at which doing more of that no longer works is what's, called the point of diminishing returns. After a certain point, additional optimization of a system does not necessarily make sense. So here's an example from the advertising world. One of the things that I did in my time at p and g was we would have television commercials, we would show those commercials on a network, and there was a calculation for every dollar spent showing this commercial. We can, with pretty good levels of certainty, say that it provided a certain amount of return, and so you can optimize on that, right. You show the commercial, you calculate the return and you try to optimize, make sure your dollars are working, showing the commercials that provided most of the return it's great. Now, what would happen is that the effectiveness of those commercials would wear out over time people would kind of get tired they start to ignore the commercial the effective of effectiveness of it would go down and so there was a choice ok what can we do to optimize this commercial let's try to make it work better let's try to make it work better but really the best strategy was not necessarily to continue to optimize a certain piece of marketing it was at a certain point to say ok, we've reached the point of diminishing returns on showing this particular piece of advertising a dollar invested in this thing is no longer providing the same level of return so we're going to stop showing this one and we're going to use that dollar to show something else that does provide that return thanks so up to a certain point optimization just doesn't work and so it's in your best interest too reallocate your time energy money attention from the things that are no longer working so well to the things that are working much better and the way you do that is by calculating as much as possible this point of diminishing returns when does it stop working for you and then go do something else and that's totally cool that makes sense you look deep in thought yeah well you quit that yeah because this supplies I was wondering what else we can apply this during different scientist because because when you reality mention commotion there's a point where you wearing a film that you were specific saying over and over hundreds of times that you begin and you you have to go back to sea way was but you begin to make more damage more than anything else and and we never we don't have it done in the film for that, but now I have one. I am so glad you brought that up because this is one of the things if you're a creative person where you're trying to optimize the awesomeness of whatever it is that you're creating there's there's this tendency of ok, I need to make a better edit more I'm going to invest more, I'm going to invest more, and then you're spending an enormous amount of time and in the early parts of the process, you're kind of picking that low hanging fruit right there dramatic improvements that you can make with with an hour or so's worth worth of effort. But the more you do it, the less actual improvement that you're making per unit of time spent. And if it has to be absolutely perfect before you ever show it to anyone, you'll never release it there's always something different that you could do, so I always I actually find this in the in the process of writing a book most of the work aside from that early burst of energy, getting the first draft in place is ending reading the entire book again and changing in tweaking and rearranging and all of that stuff and really I could do that until the end of my days and never releasing of another book ever again and so there comes a point where you just kind of have to tell yourself it doesn't have to be perfect, but it does have to be finished and you choose it is you know, books are not so much completed as abandoned just like a movie isn't just like anything else creative at a certain point you say you know what we're good can you ship it out the door eyes in that mark twain they're doing leaves of grass over sixty years and if you never feel like you reach that point of if you don't understand that diminishing returns absolutely exists, it'll never supply today, right? So this is just a good check step for yourself you're spending a lot of time and you're not providing a whole time improvement it's probably better to drop it and start optimizing something else wow moments but they're always like five sec on a five second delay where record way how it affects me and in different areas of life and I think it's comparable tio an athlete you know and it transfers over to your creative life and you're business life also I was a gymnast when I was young and there you push and push and push for tricks and then you go take a weekend off and you dressed and then you come back and not really push for it and you get it that might not seem like it's comparable but it's really it really is when it comes to business or creative don't you just can't push for it anymore we let it go it comes teo I think it'll close to even like romantic life where you analyze your relationship to the point where it's like you know talking and communicating is very important there's got to be a point where you're just living at least you know what it's like so you know there is I think it applies to most areas of life where that's getting it approximately right is better than getting it perfectly right because you won't get right exactly exactly so just like the critical few shows up in pretty much all areas of life diminishing returns also shows up in all areas of life at a certain point you just kind of have to call it good and start focusing on something else and when you do what's nice is when you shift the object of your attention to something new that's not super optimized yet you can go back to getting those really quick, enormous winds that feels so good because there's a lot of opportunity there it's not an over optimized area of life so switching from one thing to the next when those improvements slow down it's really great strategy one thing to really pay attention to on a huge opportunity to improve the function of any system is this is a concert up that comes from physics fortune our good old friend friction and friction is any process that removes energy from a system over time and so you know just very basic thought experiment uh actually just taken from physics take a hockey puck and drop it in a field full of really really tall grass and hit the puck with a hockey stick how far is it going to go very far why friction so the puck when you hit it comes in contact toe a whole lot of grass and every time the puck it's a blade of grass it takes energy away from the puck and it doesn't go very far okay so if you want to optimize the distance that the puck travels one thing that you could do just take out a big old mower and mow down the grass so let's say you know short grass maybe a couple inches you hit the puck what happens now is better it goes further it's not coming into it as much contact with the grass and so it goes further now let's cut down all of the grass take out a hose water to the field until it's, about two inches thick in the freezing a plane, a vice you hit the puck. What happens now goes forever because the amount of friction that the puck experiences is so low that you can apply the exact same amount of energy at the beginning and it just goes further because less energy is exiting the system over time. That's friction and the thing about friction is we all experienced this every single day and every single thing we do for our business and in our life. So in general, if you want to do something more often faster, more consistently with better results, decrease the amount of effort that it takes to get it started, and if something slows down that process, remove whatever it is that slowing it down. So actually, raj and I are talking you're working on this really awesome software startup called woody parrot, which is helping people respond to emails in a much faster way. It's a tool that's designed to eliminate friction can you talk about this? Because I'm actually really excited about it. Yeah, so when we designed the system with a parrot or the object was simple great companies or individuals, they want to send lord of females because there is a ton of information coming to them, they also want to be consistent so the problem wass if the pick consistency there, low speed, if they lose, if the pigs speed there, lose consistency. So we said, why don't we create use the metaphor for lego blocks? We have all the repeatable content, we call them bits, and then we have them available on the cloud out on there that stopped. And instead of composing a message, they just assemble with the right side of lego blocks, which is content blocks. And I am I used the system myself and probably spend one tenth of the time I used to spend on e mails. Yeah, so if you're sending the same types of messages over and over in your business or in your life, you have a choice. When you get a message that needs to be replied to, you can either open it up and if the reply button and spend lots of time trying to think about whatever it is that you're going to say and typing that message up manually, or you can use the system to say, I need to say this, I need to say this, I need to say this, it puts it into the email and you click the send button and it's done in a couple seconds that's, a tool that decreases friction. It takes, you know, if if it reduces the amount of time you spend answering an email from a couple minutes to a couple seconds that's a huge improvement, you can get more done in a day that's awesome. So a lot of the types of things that you can do to improve your productivity or to improve the productivity of your business, or how the entire system is working really looks like when you're looking at how the system is operating, look for areas of friction, what takes a long time? What takes a lot of thought? What takes a lot of effort over and over and over again, you remove those things that rob energy from everybody around you and the entire system runs way better. So make sense. God, in the student's conversation you talked about before the model it is, do it believe it delegated are different. The biggest thing, the friction that I see in the delegate part is because they're not set up a system where they trust that once they delegated, it was actually gonna happen. Yeah, so that's one of the reasons when we designed the system, which is very easy to delegate it sure, I almost feel like you're responding to it when somebody else's risk point, because they're actually responding, what you have written that's that's perfect yeah, so anything that you can do to reduce the amount of energy it takes to complete something the better the system is the whole run's okay? And who knew who knew things that from chemistry and physics would be sold so useful for things like business but it's the same concept applies in the same way now sometimes you're able to remove so much friction from a system or from a process you don't even have to do it any more. You can build some system let's say in a computer or some physical system that actually does the work for you awesome robot army doing your bidding like this is great it's called automation automation is quite simply a system or process that works without human intervention. And so when you're building software when you're building a manufacturing line that uses a whole bunch of robots to assemble stuff, you're building a system that is designed to work as much as possible without humans actively doing things. And in general, if a human doesn't have to decide to do something, think about doing something considered the best way of doing something compared different approaches all of the things that make us really good at making decisions if that's not required that particular process is probably going to be way more efficient then there if you build a system to do it without humans involved here's an example if every time someone wanted to do a search for information about the personal mbia on the internet and google sent them to my door if I had to to be sitting on the other end of the computer waiting for a request reading the request going into my computer and finding a file and then one by one sending that back to the person who had that request I would go insane right and there's no way that I could serve tens or hundreds of thousands of people with the same information if I had to do it all myself if I had to do it manually that's what web server software is for it does exactly the same thing someone requests to document and finds the document in memory and it sends back to them that's how it works but because it's automated because I don't have to make a decision because I don't have to focus on that process happening when somebody makes the request they get a reply in milliseconds that super awesome my robot arming is doing its work so automation if you can simplify a process if you can take the friction out of it to the point where could be fully automated that's really really grand that's the best case scenario in a lot of situations with a couple of caveats and the caveats look like two things the first is an idea called the paradox of automation the paradox of automation is the more efficient the automation is, the more critical human beings become in the operation of that system. So here's an example. Most of and let's continue with the computer metaphor. So most of the microprocessors, the chips inside our computers that make them run, they are designed and developed by human beings. But the process that manufacturers them is fully automated. So robots to do all the actual assembly of silicon in a way that creates computer processor. So let's, say, the system that that is creating these microprocessors let's, say there's, a little error. And so when a processor is created, something in the machine becomes misaligned. Let's, say a drill, and it drills a hole right through the middle of the microprocessor, ruining it. Nonfunctional. The faster that particular machine operates, the higher the number of processors that will be ruined every single minute that that machine is continued to allowed to operate right. The mohr efficient the system is the more it's, automating things, the faster it's going one mistake, one unexpected thing happening in the system. And that mistake it's duplicated over and over and over and over and over again. And the machine usually is not able to self identify that there's a problem, so we'll just keep doing something that's, how companies lose a lot of my there's a mistake that's happening in some part of the process it's not caught for a very long time and it's multiplied over the entire output of that system you remember thie concern with toyota automobiles about an error in braking systems that was potentially causing crashes that resulted and I believe something like an eight a billion dollar recall this is not a hypothetical this happens every day ok? It was a potential air actually there was an investigation there wasn't actually an issue is it mostly a concern? And but so it was a big enough thing multiplied across the big enough scope that the cost of the air was enormous and so the more efficient the automated process the more the process really relies on skilled, informed human beings to watch the process to make sure nothing like that actually happens. So the more efficient the automation the more important human beings actually are in the system because if something happens you need to stop it really really quickly now here's the irony the more efficient the automation the mohr boring it is for humans actually watched to try to detect these errors in the first place if this this process is whizzing along doing its thing really reliably remember humans like novelty to earn need novelty to sustain intention the more efficient the automation the more thing nothing changes right and so it's really, really difficult for humans to just watch an automated process happen over and over and over and over again before they get bored. And so one of the things that need to be really careful about with automation is it could be an enormously powerful tool it can help you do a lot of things that otherwise you wouldn't be people of doing but there's a risk and the risk is something unexpected happens and it's not caught and so you need to have things like so for example, for efficient automated manufacturing types of things the people who are responsible for watching the system have lots of responsibilities designed to measure and improve and they're active with the tool in ways that aren't just watching the thing go that make sense those are the types of things that are needed to sustained attention. But that's the rule that mohr efficient the system is the higher the potential risk of something bad going happen happening that is multiplied across the entire system and the mohr human beings who don't get bored are in that process. Does that make sense? Okay, so what are some of the things that in your particular business you might be able to use toe automate something that's not automated yet? So what are some of the things in your business or in your system that you just have to do over and over and over and over again and it's really bored and if if you had your way you would never, ever touch it over again. What are some of those things? The affiliate tracking affiliate tracking? How are you doing it now is that we're giving people a cord and they're there to enter the court and we can block it but if you put in a system, then we don't even have to look at it. There you go. All right. So that's a way of saving yourself in enormous amount of energy and attention doing something you don't actually have. Tio what else? Maintaining and networking and contacts and things like that. We actually had a question from guba on the exact same thing. Yeah, I just hate having to figure out where to put it, how to organize it, how to find that person later and that's my thing. Yeah, totally. What else? I mean, uh, for each workshop, we create a set of documentation that we used to communicate each other and with the producers and all that. So yet creating documentation for every single workshop it would be nice to make it a little bit more automated ago. My big thing is scheduling hate scheduling ex, sending e mails back and forth trying to find a time where books both works for you to get on the phone or have a meeting or something it's just wasteful and inefficient and not fun for anybody, right? So so one of the things and what's making software in particular such a large segment of our economy now is the recognition that in business and in common life there are a bunch of different things that have to happen and the important things that need to happen over and over and over again and software it ends up being a very efficient, very quick, relatively inexpensive thing to produce that can automate a lot of those really standard processes that happened over and over and over again. So the reason that technology is becoming such a such an important part of daily life and business life is because it's automating things that wasn't previously uh possible to automate place not in the same way oh, um, I was going to say so I might nanny connected with mom website that get good outcome I was like when I get a nineties request to build a profile, so I was like, oh, I have to look at it myself and make sure that it's inappropriate picture and the content is appropriate for her profile and then I'll approve it and then get it posted and then I was like, this is going to be hard because that I can't possibly look at every single one and and so I just put this flagging system and so don't they see something that's inappropriate? It can be flagged by a user and just, well, yeah, absolutely, and that's an automated way of accomplishing the same thing. And you may actually be able to build the system where a certain number of flags removes it from the system. Without any human being, you're looking at it and go, probably don't want to do it once, just in case somebody clicks the button accidentally. But two or three flags from different people was probably a good signal. That's. It should at least be reviewed. It's, great eighty twenty where, you know, yeah, some pictures are going to be posted there, going to be inappropriate, and they'll be there for a little bit until someone flags them, but and that's okay, gets it close enough, that's. All right, ok, does that make sense? So the benefits of automation, as well as the risks?