 So my name is Casey Cobb So this is the business track and today we're going to be talking about metrics and dashboards and I think You know just sharing a bit about myself and my companies Co-founder of a digital agency. We do Drupal and you know javascript stuff I'm an angel investor. I do I've co-founded several startups and companies and you know more importantly in each of those companies You know kind of the thing that I do the thing that gets me up in the morning Is making metrics and dashboards? That's like I love it So I'm really excited to be able to share this with you guys Not just in digital agencies, but and a lot of my examples are not just going to be Drupal stuff or programming I what I'd like to do in this talk is talk about the basic components of metrics And how to do this successfully and I'm I'm gonna share some numbers about what we measure at ricochet I'm probably not gonna share the specifics about what is ideal because it's important to me that you guys realize that Metrics are pretty personal. They're pretty Specific to culture. They're pretty specific to you know, how much you guys have developed and invested as individuals They're specific to your role They're specific to the politics in your organization. Every organization has politics So I don't I think this talk will be a little different than the talk yesterday. Did anybody attend the metrics talk yesterday? So they were talking about very specific numbers that you should achieve and to me as an agency owner They were pretty relevant to me and I understood all those but From this talk, you should be able to get out of it Even if you don't own an agency, even if you're an individual contributor or a developer or project manager or an owner You should be able to get stuff out of this So first thing I want to start with is a quote from Peter Drucker doesn't does anybody know who Peter Drucker is any hands? Yeah, he's like the he's like the godfather of of modern management philosophy. He's written like a hundred books literally and You know the stuff that he taught and has said is crept its way into Everything that that we really talk about in terms of running business these days and I think this is really apt because When you measure something actually if you're having a problem what I think is so cool about metrics is that just the act of Measuring it will oftentimes fix the problem because I don't know if you guys have heard the story about I think it was like in the the 1800s where they were trying to understand productivity and they Were this was in the factory floor and they actually yeah They had the everybody's working and so they wanted to see the effect of light levels on people's behavior And so they took the baseline and they turned down the lights a little bit And actually they turned up the lights and then productivity went up They turn up lights a little bit more productivity went up and they're like, huh did that several times? And then they wanted to see well What if we take down the lights they took down the lights from the original baseline and productivity went up even more Now why was that because everybody knew that everybody all the consultants were watching them so they They worked really hard right so but but in an agency I don't and I talk about this in other slides That's not what metrics should be about it shouldn't be about you're gonna change your behavior to do something It should be a reflection of how things are going so To so that you can improve them or you can see how good you're doing So I want to start with some basic building blocks and none of these building blocks are really all that technical This isn't a talk I'm not gonna put code on the screen and show you how to do this stuff if you have questions about Implementing any of these things. This is what I love to do. So either find me or Shoot me an email and I've got my contact info at the end of this slide So before I do that I'm gonna keep going back and trying to give some background here But there's a book that really influenced me a long time ago. It's a guy named Patrick Lencioni It's the three signs of a miserable job. So these are first one It's not even a word and he recognizes that but it's a measurement. It's that you can't in your role You can't know how you're doing you're gonna treadmill You never actually know if you're improving or doing well or doing poorly You can't see how your behavior is changing anything significantly and when you have that long enough your job just kind of sucks So it's up to us to not only fight to get that measurement or If you're a leader or you're a manager to give your people that measurement. The next thing is their relevance You know people don't see how What they're doing actually matters you could make metrics around that or you could just tell somebody how that works and anonymity they don't feel that they're being Recognized as individuals. This is what Lencioni said. I haven't read this book in 15 years But those three things have stuck with me. I didn't even have to look up this on Google or anything I just remember these because they're pretty influential in my career But the one that I want to talk about today is in measurement and I'll talk about this stuff after but when you can provide these three things This is born its truth in you know, every role that I've been in Even from an individual contributor to a developer to an executive to an investor. These things are really key and Im measurement if you can do that well Things oftentimes end up falling together or coming together really really easily so First piece of advice in developing metrics You I always think of an incentive system or a measurement system of you're putting pressure on a wall so you're on the side of the highway you have those little dividers and you're gonna be propping pressure on that and People are going to behave in ways the wall is going to respond right if you put too much pressure that wall is gonna topple over so for instance in customer service if you put your metric of measuring call times well then People are gonna like if they got to get through calls like super fast Well, they're gonna zip right through them and the customers are gonna be pissed off And that's actually not a very good thing for customer service, right? You probably also don't want to measure like the longest call because then they'll just be yapping about whatever and then you have Hold times that are really long So you got to make sure that when you do that you also if you're putting pressure on one side You've got to put a support on the other side so that your wall doesn't topple over so two examples specific examples That I think illustrate this first one is potty training my kid just recently. It's three and a half years old three years old We my wife decided to give him little Ninja Turtle cars Every time they're like two dollars at the target every time he had a successful One and two right in combination and so he had like 13 of these Ninja Turtle cars pretty successful However at about You know like Ninja Turtle car number four or five When we go at the end of the night before bedtime He would sit on that potty for like hours If we didn't if we didn't if he doesn't have to go man. He just wants a Ninja Turtle car So it kind of got disconnected right like he wants the Ninja Turtle car We want him to achieve a behavior and how that actually ended up messing up or disrupting that is that then he's not Going to bed on time and then he has a routine that gets disrupted and then that plays itself out with havoc with three-year-olds Does anybody have three-year-olds in here or has had a three-year-old? It's a challenge man So that is you know a silly example, but that's how people are too if you incentivize a certain behavior They are going to do that behavior So think about you've got to think about what the possibilities can happen and another thing e-commerce orders when I was executive at a e-commerce fulfillment company We the first iteration on our incentive system was we gave a reward for Getting out a certain number of orders in a day or an hour and for every what's called a QC for everyone that came back as like This was misshipped you would get a deduct you would get a subtraction from that bonus that you would have earned So somebody figured out that they could actually Make more money by just throwing stuff into boxes not even counting it and shipping it out and getting really high numbers And the penalty for the QC was totally disproportionate So she got a lot of QCs, but it didn't matter because she was making a huge bonus, right? We didn't think about that extreme and that was actually I mean of course we gave her some feedback and we coached her on that We actually then adjusted the the system to break down into the finest component and we actually started rewarding based on you know number of our individual products that were shipped with weight and we had an accurate reflection of the quality control the true cost of the Company, so we figured out what a true cost of a QC was to us We subtracted that from the bonus and lo and behold quality control Errors went down by orders of magnitude like it was something like seven or eight percent of orders that went out And it went down to less than like point oh one percent of orders over periods because people got better and better and better at it They learned how to grow and and and and do this stuff more easily so that was kind of an extreme case and You know Peter Drucker talked a lot about knowledge workers more often than not You know if in the warehouse work We did find that a lot where people were just gaming the system because they really didn't care because they were going to be moving On within a couple months. Maybe it was even seasonal work in our line of work You know we've got we want folks for the long run and they're thinking with their brains not just with their hands They're typically pretty sophisticated folks So the the the main takeaway from from this is that you know for the most part what I've realized is that when people aren't Succeeding or something isn't happening. It's usually not because of anybody's fault or because People aren't trying really hard oftentimes. They're trying really hard They're just doing stuff in different parts that they think are important and maybe aren't actually important for the organization from your perspective as a leader And so, you know, I think there are reasons or deeper reasons why a given metric isn't being achieved and You know the metrics and performance should promote discussion and communication We should be able to say here's where we're at and how can we improve from there? So yeah, our you know, you know in in Drupal development our QC rate or QA rate Reopened tickets get reopened or the client, you know doesn't accept that story that user story You could actually measure that and that's not to say that people, you know, it is what it is We've got a workflow issue or we've got a specification issue or we've got a client. That's really difficult The metric should should should allow us to discuss and say how can we actually improve upon this? So A couple other baseline things. We're running a marathon not a sprint The iterative mindset is key So when you learn something from metrics you should be keep in mind that you're not going to get everything perfect from day One, you're just going to try to get incremental improvements compounding effect Start as simple as possible. So I think the most overwhelming thing about thinking about metrics is people think that they need to get You know this huge development effort and get this dashboard tool and you know actually make This really robust system that measures everything. You don't have to do that You could make a really powerful metric system with Google Forms right But I think it's important and I'm going to talk about some tools that you can use for measuring Metrics you got to get to the baseline unit. You've got to get the smallest Actual unit that is important to you so that might be for us some of the numbers I'm going to share a very personal to ricochet might not work for you guys You got to think about what's important, but You know for us we get in like all of our JIRA data into all the tickets all the transitions We get in all of our time data all the logging of the time and that's a really rich data source Then we pull in all our accounting data, and we do all this references stuff But if you actually you could do this one of two ways you could say okay Well, I'm going to pull in you know, I'm going to actually write some no JS script to actually crunch all these numbers and store those Those those actual values of the metrics. I'm storing in the database That becomes problematic because then when you want to explore and discover the data and change it You've got to now do a whole development effort You better to get all that stuff in raw and some of the dashboard tools. I'll show you will allow you to parse that stuff Again metrics are improvement not blame. This is really key If you start off with blame people are going to resist They're not going to want to do it and if you start with you know How can we improve it as a team people are going to be a lot more receptive to that? And like I said break everything down to the lowest common unit us It's time tracking an actual ticket data with a couple other things in there as well but we have a series of my sequel tables that we just Lay all that raw data and the only initial the only efforts that we have to do are when we realize that's something that we wanted Like maybe we didn't pull in some specific Jura field We'll pull that extra field in and then we've got the raw kind of data. So I want to share an interesting realization kind of an epiphany that I had it involved a garbage disposal and it kind of it really blew my mind so Garbage disposal I have a condo my tenant called me said new garbage disposal when I flip it on All the stuff falls off the counter like it shakes like a jet engine. I was like wow that sounds really intense and serious Maybe I need to get a new garbage disposal. Maybe I need to get some of the fix this thing I didn't I don't know anything. I didn't know anything about garbage disposals So I went down there and I took a look and I looked at it. Okay, everything looks normal does turn on It does shake everything off the countertop and I felt underneath that garbage disposal and there's a little ring That holds the garbage disposal up against the sink. I just tighten that up Half inch almost nothing flipped it on everything was super smooth perfect and I realized I had this like You know movie montage scene where I just like zoomed out. I was like holy crap that happens in my company There's all these things that look really dramatic and crazy And all that I have to do is tighten it up a little bit and then all of a sudden everything's super smooth And it's even bigger in in in business because companies are not just one garbage disposal There's a thousand systems that work together and when one starts shaking the other one starts shaking the other one starts shaking And it becomes exponential and it looks really overwhelming But if you can break it all down to the finest piece and then start tightening I look at that as and you don't have to be the owner of a company to do this You can be an individual and find those loose rings and start tightening them up and everybody's gonna benefit from that Especially when you get the compounding effect, you know, it helps your team's happiness your work product You as an individual and everybody else and the company's profitability so, you know, that's just kind of like a I think there's a ton of examples in business like that, but it was it was found in a very unexpected place So I want to share a concept that that I've been trying to get out more. It's called accidental evil I'm still learning. I've been talking about it for several months I'm still learning how to exactly define it If you Google Casey Cobb accidental evil, there's a medium blog post that I wrote that I think really does it pretty succinctly but it's basically when Something when people in an organization are doing their best They they're making a decision with the data that they have and they basically have a path to pass that they can go They can do this or they can do that and they pick the one that is easiest for them They take the one that's the least amount of effort and in this hypothetical universe if they went the other path It might have taken a little bit more effort But the outcome would have been hugely better to the organization. So just a really quick example is let's say you have Let's say at your company people don't nail down the acceptance criteria super well at the beginning of the user story and You know client wants this thing everybody talks about it And people just start working right and the ticket ends up You know causing issue those tickets end up causing issues because they keep getting reopened and you get this huge backlog and You know the clients never you're not delivering on your timeline because nothing's ever getting finished Everything keeps getting like kind of tweaked from your perspective client just wants what they want that creates a delay in the project And then everybody's got to work overtime to make this project hit the timeline So everybody's working 60 70 hour weeks to make this work To pull together they make it work, but then that happens again Then it happens again, and then it happens again And eventually people start quitting because they love the company. They love you, but they just can't work 60 70 hour weeks really sucks, you know eventually they just burn out and If you just did something like nailing down that AC a little bit better at the beginning You would not have that effect because the project will go super smoothly the whole time and you know that to me is like accidental Evil it's like people don't even realize that that is Happening and the owner of the company might not even understand that that's happening They might not even ever get that feedback They're totally hands-off at that and so people might think well this company sucks the owners being a real jerk He actually has he or she has no idea that this is a problem and that that simple fix could actually create a much more optimal outcome And a small tweak can produce exponentially better returns I'd really highly encourage you to read that blog post because I lay out some other examples I think are really influential, but there's not been a single organization that I've ever been a part of even my company now There's accidentally what happens all the time, and I think the idea is meant to empower us and our team members and you know Our direct reports and even our bosses to realize that when crappy stuff happens. It's not because anybody is being malicious It's because just nobody realize and organizations organizations don't have souls I mean there's individuals who have souls and you can have these really weird byproducts When the organization will behave in these kinds of ways because it doesn't necessarily have a memory So great. What should you measure? That's kind of what people oftentimes want to know And I my response like I said, it's very personal It's very cultural. It's very Individual to your organization, but everybody has pain points. Everybody has things that that aren't really working right now and what keeps you up at night and You have the power to change the course of your organization through metrics regardless of your place And I think that's a really important point. I Would start with the simplest thing to measure and you could literally do that Making a Google form where so you just say hey anytime this thing happens or go past the JIRA data Or do a report on harvest and toggle and measure how many hours we recorded or what percentage of our you know Total time we should have build versus what we actually build and record it and then hit enter and let's just start collecting that Can you do that for a month? Yeah, I can do that for a month. Okay, great Or you do it for a month right and then take that data. That's a metric. That's a dashboard It's not super elaborate and I think it's important that we realize that doing this does not have to be this Huge developmental effort and as engineers or most of us are probably engineers We want to do that. We want to make it super elaborate and that oftentimes prevents you from being able to do anything And that's really really important Start with the simplest thing dev time can be a hard sell to the organization Especially when people are overworked because they don't have the metrics nobody has time But ironically having the metrics would let them have time to make the metrics so if you can find tools like zappier or if this then that or Like I said human beings you click this button run these numbers. It only takes one minute a day No time at all you can do it and so just do it and then just see where it takes you use that as a springboard for further discussion Like I said, it's a marathon. It's not a race. How can you iterate and get better and better and better? That is I think that the key to success in anything from life to business to you know Just figuring out how to get a little bit better every single time so let's talk about some specific things that we measure at ricochet and This might create some discussion in the question and answers. I want to preface this like I've said this many times now These are very specific to us. We've been doing metrics and dashboards for many years now So we've got stuff locked down. I mean everything is super super Crystal clear I think and that's the feedback that I get So I'll just share some metrics that I found really Interesting and some interesting stories that I've kind of found as a result of that so ticket burn percent As the developer so this is what's on the developer dashboard There's actually a lot of a lot of metrics But I want to know when I talk to my team every week because we have weekly one-on-ones I my partner and I have weekly one-on-ones with each individual member Team member 30 minutes every single week We just go down these numbers as a part of that 30-minute one-on-one ticket burn. How are they doing on their tickets? our You know they they over our estimate on the ticket and if so why and are there some problems that we could help Kind of get out of the like you know get in front of or are they really struggling with something? And can we connect people because sometimes it's really hard to I mean we've all been if you have developed you've Been in that situation where you're always almost done You know like you're always almost done you keep sitting there And you keep programming you're like I'm almost there and then you look up in like eight hours of pass from like oh shit This is way more than I expected to actually spend on this that can really eat into the profit of the ticket So I want to see that stuff very early. We cheat at ricochet on agile agile says Don't put time just put user point user story points Reality is that everybody knows what a point corresponds to and they use that In terms of complexity, so we just have a scale and we just say a point goes to you know a point is 15 minutes or three points is three to four hours And I do that. I I mean I people talk to me about this all the time. I do that very specifically because If you time box something it will take less time. There's a thing called Parkinson's law Something expands the amount of time allocated and to just say it's just going to be whatever it's going to be Seems to me really scary it ends up taking a lot of time And if somebody says oh, I got to do this in an hour from the planning poker Everybody thought we could do it in an hour and a new developer comes into this and goes oh Somebody said that this was an hour. Well, how could they even possibly think that and then they talk about and they go Oh, right. There's that module. I don't need to do this from scratch. So we eliminate a lot of That riffraff that can happen very early on and in our management I want to see that those tickets don't exceed a hundred percent And we're actually as a company we come in at for the past year at 96% of our estimates. We're super highly accurate Tickets without acceptance criteria tickets without estimates tickets without a proper user story like as a user role I need this thing so that some business value is delivered We have to put that in every single story because my philosophy is how can you really deliver on the value of the story? If you don't know why people are asking because the client will say stuff like I want this database Table, why do you want that database table? Well? Because I do and it's like well you're asking for some business value And I think I can solve that in a really simple way So these are meant to do that and by the metrics of seeing that every story has that We get ahead of all those problems that will happen downstream next thing I call these things zombie tickets Tickets that just stay in the sprint ticket or sprint after sprint after sprint You see the count as the owner of the agency, you know, we're at the size and I want to be at the size About 25 people where I can I can laser in on this stuff, but it doesn't have to be me It could be anybody who is managing this The developer may not be aware that that's that user story is happening that it's the zombie it just never dies and I know that any time the client has exploded on us It has been because of a zombie ticket like this thing they just go why the hell has it been? Six months, and we haven't got the story done. It's like I didn't even know about it Let's take a look and they have there's good reasons But when I see that it's been in three sprints I just ask about it and then that can sometimes unblock things and then take it reopened rate It was pretty interesting to me to see that I had this epiphany that You might finish a story on the estimate But who's to say just like the girl who was packing orders throwing stuff into boxes and shipping them Who's to say somebody just didn't finish it and say okay. I'm done all done well if the client doesn't accept it That's either because it wasn't done. Well, or the acceptance criteria wasn't clear or who knows why but It's up to the the team that is delivering this ticket to make sure that the client is going to accept it it might also be that the client's crazy and If that's the case then we have a discussion about that right we talk with the client and explain to them How this all this stuff helps them but I just want to see that number and we actually know what the averages are for all these things and I know that if somebody is higher than the average then maybe that's a problem Maybe it's not but at least we look at it and usually when we look at it It ends up not becoming an issue and we can get ahead of that crazy client, you know phone call where they're screaming at us It just doesn't happen if you can find those zombie tickets early And reopen rate is low Okay, so next plan time accuracy like I said, this is very specific to us We wrote a we have we've written several internal tools and by the way guys I see some people taking pictures We're gonna put all this stuff. We're gonna post this so it'll be on the session Website feel free to download it and shoot me with any questions, but we actually make a plan Every single day it pulls in all of our JIRA data pulls on all of our calendars Everybody makes a plan and then they submit that and then the next day you see that your accuracy how well you did and how many unscheduled things got pulled in that You didn't originally plan to work on and my theory is the reason I want to track that and I actually look at it We look at when somebody's plan time goes really down or excuse me if it's super high That's great, but that to me means that the organization more often than not is throwing stuff at people And they're just responding because of fires and why do we have fires? I don't know it could be any one of a thousand reasons, but if somebody if I see somebody's plan time is low It's probably not their fault. It's probably my fault as a leader It's probably the company's fault because we're doing some things that that person is just oh Can you deal with this fire now? Can you deal with this fire now? Can you deal with this fire and eventually people get sick of that and they quit so if you can get that stuff really early? You can get ahead of that and not have that burnout that can be associated with that developer accuracy versus the team This was really interesting to me because Originally I said okay well What is I just started digging in the data because we have a tool that lets us dig in that base data And I found that I wanted to see what is the average story size at our company the average user story? It fluctuates over time, but somewhere around 2.2 hours Is a standard user story at ricochet and like I said, this is very personal some companies their user stories might be eight hours Some might be a week. This is just for us. I like to keep things super granular and if somebody's Coming in at their their You know burn I can actually measure the burn of on the tickets overall when the ticket gets closed out What was your accuracy over the past six months? And I can see if somebody was 80 percent or 120 percent coming in under or a hundred percent are they super accurate so with those two kind of Metrics I'll show you in the next slide. We're able to do some pretty powerful stuff But the originally we were just saying your accuracy How much did you come in on your tickets at the end of that that period whatever it was you could change it? And we could have a discussion about that And you know like I said, I'll say the one-two punch on that because that's kind of a dangerous thing to do It's kind of misleading. It's not dangerous. The last thing here is scheduled versus actual accuracy This is actually like as a resource resource planning when we say that we're gonna be working You know a hundred hours on a client's project this week Did we or did we not and when somebody says when we when we allocate somebody's time? You know 25 hours to work on a project. Did they do it or did they not I want to see that And I want to have discussions about why they didn't if they didn't and it might be for a very good reason It might be that we have to change is the leadership of the company How we're going about doing things to make sure that somebody is actually able to do what we asked them to do and what they said they would do and And what we committed to our clients that has to happen if it's not the metric should be promoting that so that's why we measure that so This is This looked fine earlier, I don't know why this oh, I think it's just the presentation the resolution, but This kind of blew my mind One of our developers We were going over all of the accuracy when we first did it some people were coming in at like 50% of their estimates seems really good right some people were coming at like a hundred and forty percent of their estimates And I realized well, I'd rather somebody come in at like 95% of their estimates because they're not going over If you're too low that means that when we actually create an estimate for a client or a proposal If people come under their estimates when we actually do a planning poker and we deliver a proposal We're gonna look like we're gonna be way more expensive than we would actually be that's a bad thing So it's like something that's actually misleading And then if somebody is coming in over their estimates, well, that seems like it's a bad thing, too It could actually be that they're just not good at estimating It could be that they're taking way too long to do things or not paying attention in the time But we had something really interesting where somebody I was going over the feedback with with with the team They'd get one guy that was coming in at 60% one guy that was coming in 140% And I was coaching the guy that was coming in at 140% I said hey How do we get your accuracy down to less than a hundred percent? He said well, let me think about it next week. We came in and he said I figured it out. I said, what is it? He said I need to estimate higher. I was like I was like well, that's like saying you're gonna go on a diet and just wearing sweat pants, right? It's like yeah, okay problem solved all my pants fit now And I started thinking like what's to prevent everybody from just estimating higher? There's no prevention for that and people will do that Unintentionally they won't even realize that they're doing it if you start measuring that So that's where that average ticket size that 2.2 ticket size comes in so I made a ratio and I said okay What's the average ticket size and what's their average ticket size? And that'll create a ratio that will be somewhere above or below 1 so you can see the math here So the employee accuracy Times that ratio is there adjusted accuracy. So I'll give you two examples that look Different so in this case this person was 90% he's coming in Right on target pretty much below 100% great, right? Well, maybe not because his average ticket is 2.7 hours and the average of the company is 1.9 hours And when you multiply that by 90% you get 127% so that means he's actually coming in he's overestimating by a Disproportional amount to how he is coming in under so that gives him feedback that says hey man He's either one of two things is happening. He's either on projects that are more complicated than average Which is definitely possible and I know those projects and you can adjust for that or he's just estimating too high In which case he should estimate lower probably and he gets that feedback Which is really powerful because we never get feedback right we may not even put estimates on tickets other than points You never know how long it actually took unless it took way too long and you get in trouble And then the final project is totally different than what it was when you thought you were going to originally make it because the client Changed their mind a million times so you never actually get that feedback loop so this allows us to actually do that and Since we've done that our whole team is super accurate now the other situation here this person's coming in 120% well, that is a problem right well, maybe not because That person's average is 1.5 hours the company's average is 1.9 hours when you multiply that He or she is actually coming in at 95% so they're coming in under their estimates It just means that they're estimating too low So when I did the math that guy who said I need to estimate higher he did actually need to estimate higher when I did the math He was underestimating Exactly proportionally to the amount he was going over on his estimates Exactly so in that case that was exactly what he needed to do go higher So that to me blew my mind that by digging through the data We could give somebody that targeted feedback and that's what metrics can do since then we figured out how to get more and more accurate So project management we were actually This is something that we're improving over time and we actually did yesterday The pms and my partner and I sat down and talked about what exactly do they want to see because there's a lot of metrics there And it's kind of overwhelming But in a nutshell, you know, the most important things are tickets that are problems in the project so they can get in front of that The status of tickets in the workflow are all the tickets going to be on stage on time for the demo Or are they going to be ready for the release or are they going to get bunched up in QA? And then the developer a bunch of them are going to get reopened the day before the demo and then the developers got to work extra hours that day to fix everything and then he's got to say like I've actually gotten this feedback and since then the metrics are helping us improve that says it really sucks that we have a two-week sprint and The day before the demo these things all get reopened and then I've got to fix them when we had two weeks The tickets were ready two weeks ago or a week ago There's no reason why this was had to happen, but it just happened because it happened so Can the PM identify those tickets and make sure that's going smoothly the team allocation across all projects? So that they know for resource planning. They know if people are actually delivering on their hours commitments they know if when when If the project is on track or not sprint tidiness does everything have acceptance criteria estimates You know proper user story and then sprint completion are we on track to deliver what we said we're going to deliver And then there's other stuff like budget and and things like that which we've got two tools for as well So I want to talk about the Swiss cheese model of fault tolerance. Does anybody know what this is? so This is the in this idea imagine when you make a defense against something happening Imagine that that that thing is a piece of Swiss cheese. It has all the holes in it, right? Things are going to slip through that Swiss cheese Things can get through if they hit it just right sometimes they'll hit the wall of the Swiss cheese Sometimes they'll go straight through and you'll have that problem if you get two pieces of Swiss cheese Well, some of those holes are going to overlap So the likelihood that something's going to get through is now less if you have three You're still probably going to have little tiny holes here and there some stuff's probably going to still get through if you have four You're probably getting pretty close to being pretty well covered My thing is I want these metrics to be layers of Swiss cheese And I want to have as many as I can to ensure that issues don't happen And the more that you can do that the more levels or layers of Swiss cheese you can do The more robust you can actually Your organization will be and the more you'll kind of lose You'll lose those fires that happen or put them out early So how can you reinforce your culture and your values through your metrics? I'll I'll just ask you guys that like you just think about a rhetorical question If you have things that you want to have happen at ricochet I want our process to be super buttoned up and I want everybody to be focused on Doing there's actually a ancient spiritual text the Dow to Jing does anybody know what the Dow to Jing is a couple of people On my team. I talked about it a lot somebody else here. It's it's an amazing book. You could spend Hours just thinking about one sentence. It's a 5,000 year old Chinese Philosophical text and one of those expressions is the Zen master comes in works when he works 100% and when he's done he puts his tools down Doesn't think about work anymore. His mind is clear I want our developers and our project managers and our business development people to be able to do that I want to be able to come into work do exactly what they were put on this earth to do their Dharma and Be happy and then when they're done be able to be go enjoy their lives and spend time with their kids and go to you know Stuff after work and I want our metrics to reinforce that that's why I want everything so buttoned down they were so buttoned up because If you don't people are always thinking about work because they never know what's going to bite them around the next corner That's really stressful. And so I try to reinforce that through all these metrics Business development so I do I I prescribe to the The the Personality assessment tool disc do people know what disc is It's an amazing tool. I use it with everything I actually had my wife take the disc when we first met And we had we had some discussions about the conflict that we were going to have which she didn't want to hear anything about But we actually use that in our marriage like you know She tells me not to be so high D or high C and anybody knows disc We'll know that you know those can be things that now you can vocalize Things that otherwise you're just a jerk right now. You have a thing that is actually a strength But in certain contexts you kind of seem like an asshole so You know But it's it's it's the same thing so in business development For for me a successful business development person is going to be on disc super high D, which is like really You know intense dominance, you know in your face like like me basically And also they're going to be I which is very salesy. They know a ton of people they You know aren't going to be reading spreadsheets They're just if they have a problem they're going to solve it by calling their friend or their contact or whatever No, no a million people that can solve any given problem Challenge with that is that they might be all over the place If left to their own devices they oftentimes need the structure of the sea which is the systematic quality There's only really four attributes and then mixes in disc between those attributes so I want to be able to support my business developers high C and Really celebrate and allow them to be the high D and the high I so how do you do that? Well, how we do it is we actually have all the things that business developer should be doing in Jira And they just track their time to those epics or those ticket numbers And then we can actually see the breakdown of their time and they set how much time they should be spending on that stuff And then we see if they actually did that. I don't even have to comment about whether the mix is appropriate I just want to see that they actually did what they plan to do. That's it And and this allows us to do that one of the companies that I'm an advisor for The business developer there they don't have Jira and Time tracking software. So I found this really cool app timely app the business development marketing guy there can go make His windows of time in a calendar and he can actually track his time According to each of those events and then he can see where he landed He's really excited about that and then we can have discussions about where he does and doesn't spend his time And that's pretty cool because it allows him to stay on track and allows us to have discussions If he's not able to do some of the things that he's set out to do Obviously the easy stuff pipelining of the future for a digital agency Social media metrics you can come up with a ton of things and there's all sorts of tools out there My point is just pick that the things that you think are important and then just track them It's pretty easy Doesn't have to be high-tech. So agency leadership What do my partner and I look at so we look at our revenue targets being achieved are Our worked versus scheduled time for both projects individuals being achieved How is our PM billable percent versus project percent like what they're actually billing their time to that maybe isn't billable But it's an internal project Team feedback have people given feedback to the rest of the team and in particular have my partner and I given feedback to the team actually tracked out we look at that and My partner always tries to beat me, but he never can because I give a ton of feedback. It's very structured and we'll talk about that It's like when you do this thing. Here's what happens great job Keep it up try to give you know way more positive feedback than negative feedback, but just given basic feedback and then We measure that and I always win so That's important my ID and then accidental evil per week We actually have like a little doctor evil button in our little app and any times anybody sees this They click that button And I do that because I want people to feel empowered by the idea that when crappy stuff happens It's probably not because somebody is an asshole. It's probably just because they had no idea that that was actually happening It's a huge idea when they can click that button and say yeah, that sucked click it They don't have to record anything else I in a one-on-one can then say like hey What happened there or if they are clicking a bunch actually don't you know one or two or a couple isn't a big deal I see somebody's clicking a bunch. Well, then you know let's talk about this What's happening that really sucks so that we can improve that? So The dashboard tool, what do we use for dashboard tool? There's a bunch out there The one that we started with a tool called ducts board It lets you actually process the data and then stick the kind of the pre-processed metrics into a dashboard And then it would show the actual chart and you could kind of graph it That actually got bought by New Relic and then they closed it down And then that we actually found a tool called periscope data calm They were actually periscope I owe for a while as like they're never gonna like Twitter's gonna send them a cease and desist soon And they did so they changed it to periscope data But what it allows you to do is actually make a widget and you can just create your my sequel Query and you can do joins and they have all sorts of helper functions. You can you can do joins across databases And you can create graphical representations or charts or tables And so you could come up with any idea you want and dig into the data and see how it actually affects things You may not even share it. You might just want to know how that how that looks and all of our metrics have been that we just sat down and Made a bunch of tables and when you actually start visualizing it It's just really interesting like I remember my business partner Went in and he was like I wanted to see people's time entries And how they like did people have a lot of starts and stops or did they have really long ones? Or did they work in you know separate windows and he made this really cool scatterplot of all their time It was one of those things that you look at one time. You're like wow, that's pretty interesting It's not really actionable But it was really cool looked really neat and you could and we just threw that chart away But you could make those things and periscope allows you to do that. I don't have any interest in the company I just think it's a cool tool So there's other ones you can do as well, but you can pull in your PM software We use JIRA resource planning we actually wrote our own resource planning tool and we have an API We pull that data in your CRM your financial suite We use zero all those have API's and you can ingest that data and stick it into a database and read that with periscope So tying it all together How do you do that? Well, I Mentioned the 30 minute one-on-one Absolutely key if you're measuring all this stuff and you're never actually communicating it to people It's not gonna have much value. I Think that I strongly believe that anybody who's managing other people if you're a small organization You're the owner you should be doing that If you are a project manager and in your company, you're kind of like a manager of developers Well, you should be having one-on-ones at our company We want to be at the size where we can the owners can do those one-on-ones, but it doesn't matter I was talking with Todd of four kitchens and he was saying well He does not want to be a manager. He wants to be doing the big picture stuff And so he actually hires managers and those people do that Whatever whatever works, but the one-on-one is really really powerful and the feedback that I most often get Is that I don't have time to do one-on-ones But what actually ends up happening is if you do those it saves so much tire So much time because you have so many less fires You're going over these metrics regularly you're talking with them You're finding issues before they become giant sprouts or giant trees You're killing them when they're little sprouts You get way more time back. Everybody does systemic feedback. I just mentioned the structure of that I give it all the time anytime. I see anything that's good. It's really cheesy It's the same thing over and over again people know that they know about feedback I just think it's important to share with people when they're doing something great Or when they're doing something that can be improved in a very direct way. That's not emotional It's just it's what it is and then coaching for growth and improvement. I've never seen somebody not When they weren't succeeding I've never seen somebody not succeed from really dedicated coaching about what was expected from them and I think somebody was sharing with me that they went to a session before slow to hire quick to fire We're slow to hire we're slow to fire as well because I want to give somebody a real fair shake and you know, I think that This is kind of the stuff that Peter Drucker kind of pioneered and if you go to manager tools comm They this is where I learned all this stuff, you know, 10 years ago 12 years ago Really really powerful, but the coaching for growth and improvement. You say, hey, here's these metrics we have How can you achieve these numbers? You're not achieving these numbers And if you have somebody who you don't know how to coach because you don't know what the numbers are Well, maybe that should be your first step How do you actually give them the ability to measure how to improve instead of just saying figure it out, dude? Because they're trying their best maybe and they just don't know how or they there's something that you're doing to them That's preventing them from being able to do that make the numbers set the goal and then coach them to improve that and you will probably Learn just as much about your organization and yourself to get to that get to that level and you know, I I say this a lot, but It's all about love management is all about love and metrics are all about love They should be they shouldn't be a newspaper that you whack people with how can you keep everybody happy at your company? How can you keep your clients happy? How can you really love the fact that you know, we're all in this together from an agency perspective and a global perspective? Metrics should allow us to do that and I think if you come at it from a really From that perspective the metrics are gonna be more effective people aren't gonna be afraid of them They just realize you're trying to get better. So I just want to highlight that point and I went through this a little more quickly. I didn't think I was gonna have time for questions But it looks like I will and if you're in your question doesn't get answered reach out to me Twitter Casey Cobb casey a project ricochet comm love to answer any questions. I love this stuff. So Any questions? I know this is a lot to take in What's Yeah I'd have to know about the specifics there, but Government's kind of a tricky thing right because they're not gonna give you more resources oftentimes like I like business in the market because all these metrics tie back to actually Basically profitability like you want people to be happy, but happier people are more profitable profitable people in the service So, you know it if you if there's a number that if that's true When you implement these things you should have more profit you should have more resources You should have more money at the end of the day and that can pay for the effort to do that So it would we can talk after or maybe we can we can kind of bounce ideas up of each other But like if I were in that situation, you know, I would I would you know find a way to either count that stuff You know like maybe there's an API or maybe you know like if I were thinking about it And my job was really crappy I might find somebody who could do that for me and just I'll pay him Some amount an hour to do that because I don't want to do it or I'm too busy and then I'll use that to justify the results I who knows there's there's a thousand ways you could do it, but let's talk after and we can brainstorm some ideas Anybody else? Yeah, um, yeah, if you are actually doing one-on-ones Sure, whatever else basically I find that there's very few problems in an agency That don't get solved through increased communication if you think about agile and scrum When we have we have weekly, you know sprints for some project, we might have a weekly sprint or a bi-weekly sprint and If for a lot of projects, that's just fine When projects are really complicated or they're really hot and heavy or the clients a little bit more challenging We move to a daily scrum model or we just stand that we have a daily stand-up and we talk about that We also have a daily company scrum, but I find that more communication the better So to the note about you know one-on-group great That's like a company meeting Or a team meeting But you need to have the one-on-one you need to have the individual to individual because it's about that relationship And if your directs does not have feel that they have that relationship, you know They're not gonna have an avenue for those problems basically You're not gonna get the feedback trickling up people aren't gonna share in a group the same that they're gonna share in a one-on-one and You're not gonna have the report you're not gonna have the jokes You're not gonna know about their kids and and the one-on-one is very specific It's ten minutes for them to talk about whatever they want ten minutes for you to talk about the company or what you want from them and Ten minutes for their growth and improvement And you can't do that on one-on-group. I mean kind of by definition. So does that answer the question? Okay, cool Yeah, and if you go to manager tools comm they have all that and they have tons of podcasts hundreds of podcasts but so ten minutes for The person to talk about whatever they want if they want to talk about their dog went to the vet and They're really sad about that or their kids graduation or whatever or how things at the company. They want to improve it. Let them That's great. That's how you build that report if and then the next ten minutes is for you So I use that time to go over the numbers and the dashboards and I talk about company initiatives and company projects and new team members Or new clients or new whatever projects that we're working on and then the last ten minutes is for their growth at ricochet Most people are reading a book and I say just read at least one page of that book Every single week only one page everybody can read one page no matter how busy you are and then we talk about that We talk about how we can help them improve we talk about issues that they're having with the company How can we make them happier and more productive? And that's kind of how that breaks down when you first start doing it They take like an hour or two hours and then once you get up that past the backlog then they start going more quickly Anybody else questions Okay If you guys come up with an example this is a vaccine illegal. I'm really interested in this idea Shoot him to me. I'd love to have a discussion about it. This is a topic. I'm really interested in And if you have any questions about metrics or anything just let me know. Thank you guys very much