Everything you need to know about KPIs - What, Why, and How
Key Performance Indicators, KPIs, numbers, metrics, measures, measurables, scorecards, scoreboards … What exactly are they? Why should we use them? When should we use them? How do we create them? nd how to we make sure they get the right results … while keeping us clear of the Taskademic that plagues the modern workplace? In this episode, I talk with a KPI Jedi: Bernie Smith: Bernie thinks about KPIs deeply; he teaches and consults on KPIs and has a terrific, down to earth command of how to create and use them. I came across the book "KPI Checklists", and was impressed enough that I immediately reached out to connect with Bernie for an episode. Bernie coaches businesses to develop meaningful KPIs and present their management information in the clearest possible way to support good decision making. His clients have ranged from small companies to big ones likeUBS, Lloyd’s Register, Scottish Widows, Tesco Bank, and many others. By the end of this episode, you’ll be able to replace fears angst and confusion about KPIs with simple clarity.
Bernie Smith - KPIs: What, Why, and How transcript
Scott Levy:
Really excited to chat with you about KPIs because very few days go by where in some way, shape, or form, someone isn't asking me how they can improve their KPIs. Are they measuring the right things? How do you make sure that you're measuring what matters? How do we come up with them? All these questions that you, after reading the book and looking at all your materials, have already wrestled with and can help everybody sort through. So I'd love to just talk a little bit about KPIs. In your view, what are KPIs?
Bernie Smith:
First of all, I find it really reassuring, Scott, when you say that these questions keep on coming up. It was a bit of a bet when I set up Made to Measure KPIs a few years ago, but fortunately, it seems to be a topic of interest. So, what are KPIs? Now, you hear all kinds of very strong opinions about metrics and measures and KPIs and differences. I have a really straightforward view. You measure things that you care about. And if you do that and you act on those measures and you do the right things, then things get better. It's as simple as that.
For me, KPI is the term I use a lot because that is what people look for. It's actually pretty straightforward. Key means important. And what's interesting is, important depends on who you are. And I have quite interesting debates with my clients very often about, well, that's not a key performance indicator. That's a performance indicator. Well, for you, maybe. If someone's running a production line and they're really close to the process, then something like line 17 downtime is a KPI. If you're the chief financial officer, it probably isn't - profit might be your KPI. Put very simply, KPIs are measures of things that have a significant impact on what you're trying to do. And that's it, really.
We can have a chat later about how you can make that more objective. But as a principle, that's normally where I leave it.
Scott Levy:
So one thing that you mentioned there that got my attention was the person on the front lines may have one KPI. The CFO may have another. I find people overthink those relationships - sometimes they're worried, “does it align just right?” And it sounds like you don't need to think so much about that. Is it just important to you? And does it help you get whatever outcomes you're after? Was that a fair way to say that?
Bernie Smith:
Yeah. 100%. And as you're talking though, it kinda triggered this observation I've made of why people get tripped up by it. Because what they try and do, what a lot of people try and do is find one metric that applies to every level of the organization. So this idea of this golden thread running all the way through. Now, there are some things like profit, which impacts on everyone. But, if you're fixing a machine or answering a phone, you don't have direct major control over profit. You have a tiny part in it, but there may be a hundred thousand other people involved. So where people get tied in knots is trying to find something that works on all levels. And you're absolutely right. You don't have to align them. What you do have to do is align towards the objective you can influence. Now if I'm fixing a machine, then actually, the objective I can influence is machine up-time or, maximum stable speed on that process.
If I'm answering a phone call, then customer satisfaction, first touch resolution, average handling time, for me, that's the thing I can influence. Now, there is a thread that goes all the way up to profit, to growth, and so on, but it's not a direct link. So what we need to do is we need to build intermediate kinds of outcomes that we link those individual KPIs to, and then it all ties together. And it's that visual representation, that visual map. So strategy maps or KPI trees, as I call them, binds it all together and makes it much easier to relax and go, actually, I've got different KPIs from you, but they all work together. We're all fundamentally aligned with these top 6 strategic objectives, say, for the organization: making a profit, growing, managing risk, being legal and compliant, meeting our ESG commitments, and so on. But, yeah, you don't have to do it all through one measure that does everything. It can be tailored to your agency. Because individuals only have certain amounts of agency.
Scott Levy:
Makes total sense, and that's something we see a lot. Obviously, our products, ResultMaps, and our most popular view is that visual where you can trace your strategy, your KPIs all the way through the organization. If I said there's one thing that holds people back when it comes to KPIs, is it overthinking them? Is it like, how would you answer that question? So
Bernie Smith:
The thing that tends to hold people back is fragmentation and uncertainty. So, if you take one individual, he or she may have really strong opinions about what to measure or they may not. Then, you have 3 or 4 people, and the chances of them aligning and agreeing easily and readily rapidly diminishes, and then you imagine you've got a hundred thousand people in the organization, the chance of accidentally stumbling across a package of measures that give a balance here of the whole organization that everyone agrees on, if you don't have a method, it's pretty close to 0.
So that's the biggest single stumbling block. And that's really where I've set out my stall and said, okay, there is a way you can do this, but it's not gonna happen by accident. And there's a talk I do a lot where I share a picture of a street that I walked down in the Canary Islands, a nice holiday island off the coast of Africa for anyone who's not been there. And I walked down the street, they had this wavy sidewalk. And if you look closely, there were these stumps of about 20 or 30 mature trees. And what has happened was, the local authorities had relaid the road surface and the sidewalk, and they've done it and waved the sidewalk around the tree trees that were there. And then, almost immediately after they did that, they went down and chopped down all the trees on the street. And the reason I start with that as an example is I see this with KPIs.
Individuals making the best decisions that they can, thinking hard, you're doing the right thing. But it's not a joined up enterprise. So you end up with all these weird disconnects and arguments and alternative measures, [and] measurement regimes. And the bigger the organization is, the worse it gets. So I think this fragmentation, and also the uncertainty that comes with not having a method, paralyzes people, paralyzes teams, paralyzes individuals, paralyzes organizations. Because deep down, they're not really sure they're measuring the right thing, and they sure as heck aren't aligned as well. So there's lots of uncertainty and flip flopping between different measures. Maybe we shouldn't be measuring downtime. Maybe we should be measuring efficiency, or maybe we should be measuring availability, or, what about breakdowns? It all just becomes really confusing.
And as soon as you lose that focus and clarity, all the effort moves from fixing the issues to going round and round in circles. And it was actually those conversations that led me to set up my business, Made to Measure KPIs, 15 years ago. Because I was actually a process improvement guy before I set up my company. And I saw this happening so many times, I thought there's got to be a logical, repeatable, structured way of doing this that doesn't require you to be a super genius on Cristobalite mold-milling, or heart valve manufacturer, or polyethylene cracking. There's gotta be a process for doing this. And that's what interested me. Could we systematize it? Could we turn it into a method?
Scott Levy:
What comes to my mind as we're talking about it, is it’s absolutely chaotic when you get an enterprise organization, which we've both worked in and consulted to, I find there's sometimes this idea that we're a smaller team. We don't need KPIs. We're gonna leave those. And, unfortunately, it feels like sometimes teams use that as an excuse instead of doing the work to come up with a good one and experiencing the acceleration, the focus, the help that having those KPIs brings them. Do you ever come across that where people are like, oh, we're a smaller org, we don't need KPIs?
Bernie Smith:
All the time. All the time. But what's interesting is they almost always have KPIs that they use and they care about. Just don't think of them as KPIs.. I've yet to meet … I've yet to meet a small business that doesn't focus on profit, sales, and cash flow. That's almost like a given, but, it's in there. And, a friend of mine, he was so close to the manufacturing plant, he could tell before he got out of bed whether the plant was running or not. That was his KPI. Am I feeling vibrations through the floor? Is the plant running? So, right, everyone cares, everyone counts, everyone shows an interest. But it may not be as formal, and that's fine. A little bit of active consideration, a little bit of thought about what we're measuring, and why, can go a long way. And I'm absolutely a fan of baby steps and minimalism. ONe of the ways that people get themselves into a pickle, as we say over here, is it's a bit like New Year’s. You join a gym. You go for 2 hours each time for 3 days, and you wake up on the fourth day and you can't move. And you just give up. And I see the same thing happening a lot with KPIs.
I am a big fan of tiny baby steps. So if you're a small business and there's a bit of pushback in KPIs, just pick one thing that's gonna make a difference. Don't over engineer it, just get it into everyday use and try and make sure it's one that's insightful, ideally leading, and gives you a chance to fix something and get the feel of it, start to use it in meetings, start to review it. I'm a child of the seventies. I've probably got a bias here, but analog is good. Software is great, And I've got 5 screens in front of me, but there is something strangely engaging about a whiteboard or magnetic letters or whatever. I've done a lot of hands-on process improvement, offices, call centers, shop floor. There is something almost mystical about people actually writing their own numbers up. And you start to see people come into the office and they'll look at the board. Soon as they come in, they'll see how they're doing, they see how the last shift did, how the last team did and you start to get that competitiveness. And healthy competitiveness is a really good sign of engagement. You don't wanna overdo it, but anyone who's played Monopoly knows what it's like when people start getting infused.
Scott Levy:
100%. We've seen a corollary to that even though it's a digital product, but we've got two ways you can have your KPIs updated. And for folks that are students of other methodologies that like other terms - scorecard, measurable, whatever you wanna call your numbers - the important numbers is what we're talking about with KPIs. What we find is you can have these things auto-update from some kind of API call, but there's also a form you can fill out with the ones that you own. And we find a lot of people just like the rhythm of, let me pause, let me check my numbers, I'm gonna go ahead and enter them, because it forces me to stay in touch with them, and it dials me into this process.
I find that really interesting. Because we think it's great to have everything digitally connected, and we can totally do that. But there's something about that rhythm that you just mentioned and it must have something to do you know, something similar even though we're not at all
Bernie Smith:
I think you're onto something. So there's two things going on there. The first thing that springs to mind is the Japanese rail system. They have this interesting thing where the drivers, when they're driving the train, they say out loud what they're doing. And what they found is, they'll say something like, I'm accelerating. I'm breaking now. Even if they're on their own, and what they found is that the act of stating what they're doing increases focus, concentration, and engagement. And it’s almost a form of mindfulness. You become aware of what's going on.
And when you take a step back from measurement and think about why you're doing it, we're doing it for a reason. That reason is normally improvement. So if you have a human in the loop, people start to look at what's going on. And I found this particularly when I worked in manufacturing. If you've got big complex processes, yes, we want to know what the downtime is. Yes, we want to know why the machine broke. But, actually, you can't beat the person on the process thinking about what's going on, making a note of it, and maybe adding a comment as well. Because ultimately, someone's gonna have to go back to that data and say, okay, what's going on here?
And the insight and the sort of enriched notation that a human can add is fantastic. Now, from a data management point of view, it is a bit of a nightmare because quite often the common data is unstructured. But the combination of observing, recording, and annotating massively increases buy-in when you come and say, okay, we've seen the biggest problem is this. They'll nod because they've seen it and they've noted it. They've engaged with it. So the latter stages of what I call the improvement cycle, that's when we decide to do something with our data, with our analysis. It becomes much easier because they're already engaged. So I think you're really onto something there. You can over automate it, and it then becomes a separate process from the people and the actual business process if you're not careful.
Scott Levy:
100%. That's what's been interesting, because for years, we've done 1-on-1 onboardings. And we find in sales calls, everybody wants to know about the API and the connectivity. We've got it. We've got it. But when we get into it, people will ask us, what can I do to set myself up for success with these processes? It's that. It's that pause and reflect. Get that rhythm of pausing and reflecting, becoming mindful of these, and it's fascinating.
There was someone … I was coaching them on KPIs because they had all these numbers, but [they didn’t know their]precious few. They were doing a support queue. So you can trick that out: every request in however much time, but they're getting so many requests. That wasn't gonna work, and not all the requests they were getting should have been actioned. But they ended up with … of the requests that were deemed urgent and important (and there were criteria for that), okay, we want a 24-hour turnaround maximum.
Watching people go through the process of thinking through what KPI was gonna be most meaningful, that just started to create a whole other set of interesting discussions. So that leads me to this question I'm dying to ask. [I’d] love for you to talk a little bit about your process for coming up with KPIs, because you've got a really rich process that reminds me of exercises I've led people through for innovation and product and everything else and you actually take KPIs through a very similar process. Could you tell us a little about that process?
Bernie Smith:
Yes. The method I built for choosing KPIs has a very un-glorious history. As I mentioned, I was a process improvement guy, and I was watching and seeing multiple ways in which KPIs go wrong or KPI selecting goes wrong. So I came up with a process called the ROKS method and the ROKS method stands for Results Orientated KPI System. Put very simply, it's all about working backwards from the outcome that we want. So, my view is KPIs are like tools. You don't go to a DIY store, buy a pile of tools and then decide what job you're gonna do - well, you shouldn't. What you do is you decide what you're gonna do. You're gonna build a kitchen or put up a garden shed or whatever, And you go to the store and you buy the tools that you need. And my view on KPI selection is that it's just the same.
You find and identify the KPIs that will help you reach your objective. So the logical first step of the method is to decide, on a high level, what you're trying to achieve. Now, it turns out that can be quite a tricky conversation because there's so many different takes on strategy, varying from the quite mundane practical through to the sky high almost theological. So I've boiled strategy down into a fairly consistent set of starting points that many commercial organizations aspire to, even if they don't consciously aspire to it. So if you think about most commercial organizations, they want to make a profit and stay solvent, they want growth, they want to innovate because innovation is a good lever on profit and growth. They want to manage risk and that's a huge family of things, but the headline is risk. It might be operational risk. It might be legal risk. It might be conduct risk. It might be environmental risk.
They wanna sell the right side of the law because organizations that don't, typically, don't go on for very long. They also typically want to meet their environment, social, and governance commitments as well and that's becoming an increasingly large thing. So the first step of the method is to take all our existing strategy and try and boil it down into around about half a dozen simple statements of outcomes that we're looking for. And what I've just mentioned, I call them the big six. They're a good starting point. We'll fine tune it for the organization. Now, some organizations will also have mission objectives, typically charities, medical research organizations, things like that. Cure tuberculosis, educate the world's children, whatever. So we'll also have perhaps one, maybe two customer objectives. So we create that and we agree on what we're trying to do. Now the next step almost could be the first step, but we have to do it this way around because unless we know what we're trying to do, we don't know who we need in the room.
But, as soon as we straighten out our strategy, we need to engage the right people. We need to get them in the room, we need to get them on board for two very good reasons. Firstly, we need the right brains on this and my method doesn't bring subject matter expertise - the subject matter expertise always lives in the organization. Secondly, and probably more importantly, if you don't engage people, they won't buy into whatever comes out of the process. So it doesn't matter if you come up with the right KPIs, if people are not bought into them, if they don't have a high degree of ownership, you might as well go home.
So the second step of the ROKS method is engagement, and that's a mixture of stakeholder analysis, communications management, putting workshops together, getting the right people in the room, and crucially, making sure you don't leave people out, because the quickest way to break a KPI design process is to annoy people, to exclude people, and to get them off board, on the wrong side. So we have to think about that carefully. As I said, I've made loads of mistakes over the years. I remember having a call in a big UK, well, global organization, where I built this dashboard, and I was so proud of it, and I identified the stakeholders. And we're having an online conference … I think, 20 stakeholders or something. And I'm watching the people appear on the call, 15, 17, 19, 25, 30, 35. By the time we got to about 24, I knew I was dead in the water because you can imagine the first comment I got was, “This is the first I've heard of it. No one's involved me.” They're like … dead in the water.
So I learned the hard way that you have to engage people. That's the point when we get to the KPI tree, the strategic map where we break things down. So we take the original strategic objectives, we're breaking them to lower level outcomes. So for example, we might say for sales, that we need to maximize sales revenue. We need to maximize margin. We need to maximize repeat custom from existing customers, and we need to win qualified leads and new customers. So it might be next level objectives. And of one of those, you break it down further. You say, okay, how do we retain existing customers? We need to be price competitive. We need to contact them in line with our sort of stakeholder management or our customer management plan. We need to address problems promptly and well. And there's some things we can do, and then we can break those down into KPIs. So we build this map.
Now, I've actually separated the KPI selection into two steps because we have two types of stakeholders. From my experience, you've got the dreamers, the people who want to measure everything and anything. And they'll create list after list of KPIs. They're quite often senior because they don't have to actually implement them and mean, it's a particular type of individual who just wants to measure everything. Then you have what I call the pragmatists, and these are the people who talk, but they go quiet and start to shake their heads when new KPIs get listed out. I like that.
Scott Levy:
I'm gonna have to use that.
Bernie Smith:
Yeah. And the problem is that you can't keep both parties happy in one step of the process. So what we do is … we have long listing, which is the KPI tree, and I warn the pragmatist. I said, look, you're not gonna like this. We're gonna have far too much stuff on here, but don't worry. We're gonna bring it well under control. We're gonna reduce that list dramatically in the next step. So be patient, humor us. We're gonna come to you next. So we do the KPI tree. It's not uncommon to have 6, 700 KPIs, potential KPIs at that stage. It's scary.
Scott Levy:
Enterprises, right? These are large companies. This isn't gonna be for your hundred person company.
Bernie Smith:
No. But you'd be surprised, actually. Even a hundred person organization has a lot of stuff going on in it. And what we do is we'll set the threshold much higher. So, because they haven't got an entire BI team, we're gonna need to be more brutal in the shortlisting. But even something like a hotel, there's a lot going on if you think about it - we're dealing with fresh food, we're dealing with staff, we're dealing with bookings, we're dealing with maintenance, we're dealing with customer satisfaction, loads of things going on. So we create this long list, we base the film, and say, okay, it's gonna be scary, we're not even gonna attempt to measure 90 percent of these things, not even that. So I build a tree, we've got really good engagement, we get agreement that everything we could want to measure is on there, but now we need to bring it under control.
So what we do next is to create a simple matrix, a bit like an Eisenhower matrix. Using this we say, “Okay, on the X-axis, let's plot ease of measurement and importance.” A KPI like “cash in bank” is a 10 for importance, and a 9 or a 10 for ease of measurement, so it's up in the top right hand corner. That part is easy. And typically, in the bottom left hand corner, we have “not important and impossible to measure.” If it's empty, we don't have to worry about it.
So we've got two other zones to worry about. So next, we look at not important but easy to measure. And people always start looking a bit puzzled and go, what's going on there? Typically, that's populated by external systems implementers who come in, they’ve brought in an ERP system or a CRM system or something, and they've turned all the reports on. So just to demonstrate value, they go, let's just switch every report on.
So to give you an example, I work with a big telecoms organization in the UK. They had a call center with teams of 7 or 8 people, and each morning, the call team manager and their 6 or 7 direct reports were supposed to sit down and review 128 metrics. Guys, you can barely look at 10. 128 is insane. But the reason is that these call handling systems generate huge amounts of analysis, they can generate huge amounts of reports, they've all been turned on. And no-one had the courage to turn them off. So anything that scores low for importance but high for ease of measurement, we need to be really careful about because it's a kind of a vine-weed that chokes meetings, chokes individuals, overwhelms them. It's the 500-page PDF that arrives automatically in your inbox every day, or the 50-page Power BI dashboard where people need 3 of the KPIs.
Scott Levy:
It’s like a temptation. It's like the temptation, the 5 temptations of the BI team are like, “Oh, let's get you some metrics quickly. Here's all of them.” I've led teams where, on occasion, that's been done and we had to coach people to pair it down.
Bernie Smith:
Yeah. And, one of the drivers of it is where the BI team or the people who are implementing it, for whatever reason, I'm able to talk with our internal customer. One of the ways to really cut it down is to understand or agree how the information is gonna be used. Because as soon as you know what it's gonna be used for, you can cut all the fat out. But I've often seen this situation where the BI teams are significantly separated, for some reason, from their internal customers. So, with the best intentions, they kitchen sink it. They just chuck everything in case. So that's a kind of real potential area for pruning things back.
The final quadrant and the one that I really focus on with my clients is, things that are important or potentially important, but appear to be difficult or impossible to measure. And these are what I call gold-bar KPIs. So these are things where people would dismiss them because they look at them and they go there's no way we could anticipate our customers' intent to leave. It's impossible. Just disregard it. They are by definition hard to measure, but there are methods we can use to make apparently impossible things measurable. If they are important enough, then it becomes worth deploying those tools. So I built a 3-step method called the Holmes method, where we deploy 3 tactics for making things measurable.
So to give you one example, I worked with a large pension provider, a retirement fund provider in the UK. They've seen a step change of people transferring their pension funds to other organizations. Now if you are a pension provider, that's really bad news. You want people to invest with you and you want them to stay with you until they die. That's the business model. So if people start transferring out, then you've got an issue. And I was in an operations meeting where this was being discussed, and a couple of the middle managers said, “it's not like we can read our customers' minds.”
And then one of the younger process teams [people] popped up. She said, “Actually, no. I think we can.” And what she'd identified was that there are things that people do just before they transfer out, there are behavioral indicators. And in this case, the behavioral indicator was contacting the organization for a fund valuation. Because the thing that you do just before you transfer your pension is you contact the organization to find out How much your pension pot is worth. And this is going back quite a bit. So this was done through telephony, through voice, and they've actually just automated that process, and that was what had led to the increase in transfers because there was no longer a human in the loop. It was done through IVR and an automated service.
So two things happened. Firstly, they identified what the issue was. But secondly, they realized there was this last touch point where you could actually recover the customer and maybe even keep them for life. So if people come away from the ROKS method with nothing else, I encourage them to think about things that you would love to measure, but might discard because they don't look easily measurable. There are ways you can do it, and often, they can be game changing.
If you look at Google or Experian, people like that, they've taken this idea of measuring something that's not particularly obvious how you do it, and they built the entire business around it. With the Google search algorithm, how do you measure the importance and relevance of a page to a search term? It's not an easy thing to do, but once they cracked it, the rest is history. So that's the shortlisting process. So we score based on importance and ease of measurement.
Then we can create a ranking score. So we multiply the two together and you end up with a ranking score between a hundred and zero, sort the list, and then we go through and as a kind of negotiation discussion, we talk through: which ones are we gonna do initially and ones are we gonna park? We'll do those a bit later on, and [indicate] which ones are we just not gonna do for the time being. And the beauty of that approach is you can start really small, but you have a plan. So you decided to pick maybe 1 or 2, but you have a priority list. You can do them in tranches. You can split them between different parts of the organization. But you have a structure, you have a ranking that's been agreed with the group, and you have a road map. So it's a good way of acknowledging what we wanna measure, but without getting overwhelmed.
One tip I would give though is if you do that, make sure you record your reasoning behind the Scores, because one thing I've discovered is that people instantly forget why they gave something a 7 and something else a 5 or a 4 or a 3. But it's a great way of ending up with a short list that the pragmatists love and the dreamers, the people who come up with the long list, they're kinda comfortable with too because they've had their moment. It created this big cloud of KPIs, and then we shrunk the list down or prioritized the list to meet kind of practical constraints. And then the final bit of the kind of selection and definition bit is step five, and then there's a couple more steps where we design dashboards and reports, and then we go live.
So just one last comment then. The definition stage is the kind of bit that everyone agrees vigorously is really important and we should all do it, and most organizations don't actually do it. So what I find is you have to share a few horror stories to motivate people and help them appreciate what the risks are of not defining it.
Scott Levy:
I really like that, and we do that with this concept of actionable core values. I like putting the story behind it so that it sticks with people and why this is important, and they have that context. Because so often, when I talk with people in organizations, in particular as the organization grows, they lose that context. As you mentioned, they forget why it was put there. And having that story, it just brings it to life. I'm asking this question pointedly because I know some of the questions our customers will ask in particular when we're dealing with smaller, scrappier customers, again, there's always the I get why you would do that at a huge company, but we're only 10 people, we're only you can fill in the blank at the size.
Bernie Smith:
Yep. Smaller companies can benefit from the process. At the early stage, you're not necessarily gonna have 6 although maybe with the help of AI people get a little crazy. I'm pragmatic. I actually created the ROKS Express method for smaller companies. So I have a separate method where I build KPI trees for most of the common functions, pick out the most frequently used KPIs and define them or do a partial definition of them. So because I acknowledge, I understand small businesses don't have dedicated teams doing this stuff. So the idea behind the ROKS Express method is, it's a stepping stone to the full KPI tree method, but for people who haven't got the time and resource and energy to build KPI trees.
Oh, and you notice certain things coming up time and time again. Health and safety, it's got certain things that you are definitely gonna want to measure pretty much everywhere. Finance is the same. HR is the same. One of the nice things about smaller businesses is probably 80 percent of what they do and what they want to measure is fairly predictable. I use something called traits to identify what the characteristics are of a business. Do they employ people? Do they have vehicles? Do they deal with fresh produce? Do they have confidential data? There's about 55 characteristics that you can use to very quickly profile an organization. Some of them will be almost universal, like finance. Others will be more niche. Do we have research and development? Do we deal with patents? That kind of thing. So you very quickly pick out the key traits and then associated with each of those traits are certain commonly-used KPIs, and then second tier ones where in certain situations, you want to use them. My third book, Getting Started with KPIs, is a manual to do that. It's trouble is, it's a bit of a big book because it's got so much in it. Now what it is, it's a sort of simple how-to guide talking through a lot of the principles I've talked about. The second half of it is a library of defined KPIs with examples. So for a smaller business, the idea is that they'll read through the guide, understand the principles of prioritization, engagement, and so on. But then, within a day, they can pick out some of the get-you-started KPIs from the back, use the examples, get up and running. But the place where many of my clients get stuck is, where do I start? What do I choose?
Scott Levy:
We see exactly the same thing. We often have someone who's charged with owning the overall scorecard for the organization, the overall KPIs, who thinks that they have to be an expert in every area. And what I like about both your approach and then this idea of, here are these KPIs, is you have to learn how to ask the right questions and guide people through it. You don't have to go become an expert in every single area of the business. That's not necessarily the best path for you.
Bernie Smith:
Just to give a tip, my one magic question for getting started with KPI trees and ResultMaps is: what does a good day look like and what does a bad day look like?
I've never met anyone in the organization who hasn't got an opinion on that. And that will give you a thread to pull on about what outcomes we need to think about. Now, it won't be everything because people tend to focus in on problems and immediate problems and sometimes lose sight of the longer term, low vs. high importance things, but it's a great place to start.
Scott Levy:
I love that question. I've got some friends with a company called Produce8. They will measure your app usage, and they're not measuring it from the perspective a lot of people in the industry take, where it’s, “we're going to monitor our employees.” Instead, they do it from the standpoint of that question, “What does a great day look like?” And if I know, how can I replicate these great days? If I have the intention of having a great day, what does that look like? I love the use of that question, that mentors of mine like Tony Robbins, Stephen Covey … I just love that process, and I love seeing more and more people inject that question. Because if you ask that question, now it's less, “Oh my god. What do I have to do?” And more, “if I'm firing on all cylinders and I'm having a fantastic day and we're performing well, what does that look like?” Much more powerful question. So kudos to you for now.
Bernie Smith:
I've had a lot of long conversations with people about performance. One interesting thing, and I don't know if this is a cultural thing because I know from my work in America, you're generally a lot more optimistic and that's great. Ask Brits though, if you ask them what a good day looks like, he's staring at the shoes, a bit of shuffling, whatever. If you ask them what a bad day looks like, you can't shut them up. It's just, they reel off all these problems. And I discovered two interesting things. Firstly, us Brits anyway, like to talk about problems a lot. But secondly, when it comes to building a KPI tree, they work much better with positive outcomes. So for example, if my positive outcome is to arrive on time at work, okay, that's one outcome. There are probably a thousand different ways or reasons I could be late for work. So putting problems into a strategy map or a KPI tree doesn't work. What you can do is you can tap into that pessimism, you'll be able to list out all the problems and then flip it on its head and say, “Okay, so what positive outcome does that imply?” And then I call that method reverse-brainstorming. I didn't invent it. Someone else invented it. And it's such a useful tool. You just get people flowing, talking about problems. Operations people as well are quite problem focused. You just write it all down. Just get it out.
Scott Levy:
There's great research out there explaining why our brains are that way. It's an evolutionary mechanism. The people that weren't able to quickly assess the risks and the problems were selected to not be in the next round, so to speak.
Bernie Smith:.
So you mean the pessimists were right? But no, I'm an engineer by training. Almost all engineers are pessimists. They're constantly thinking about how stuff goes wrong because that's how you build machines that work. You figure out how they're gonna break, and you try and stop it happening. So I guess I'm one of those guys as well. But yeah, It's quite useful. If you're ever stuck for conversation, just get people talking about problems, and they'll just go on and on. It's fantastic.
Scott Levy:
And this reminds me of one of the things I really enjoyed about your book that I read. You start out just talking about checklists and why checklists are so powerful, and I wasn't expecting that. I was expecting us to start out [like,] “We're gonna talk about KPIs.We're gonna measure. Let's go.” Instead, you started with checklists and the power of checklists. And when you were telling me about the engineers, the drivers of the trains in Japan, it made me think of that. What is it about checklists that is so useful from your perspective? What is it about checklists that you find so empowering and useful?
Bernie Smith:
I'm a bit of a joke in my family - you may or may not be able to see all my labeled drawers behind me. There are little IKEA drawers with labels on. The reason I like the checklist is the reason I like labeling stuff. I hate wasting time or hurting my brain trying to remember stuff. I don't like trying to remember where things are. I don't like worrying about whether I left my laptop charger behind. So my entire life runs on checklists. When I run a course, there's probably 4 or 500 items on the checklist. Do I like running through a checklist? No, I have to make myself do it. Does it help me sleep at night? Absolutely. You must have this when you go traveling. And have you ever woken up at 3 in the morning going, “have I packed such and such?”
Scott Levy:
Oh, I too am a checklist fan. We've this product that does it as well … Literally in front of me on the desk is my morning routine checklist. It’s not so much that I love doing my checklist. It's when I started observing what my great days look like, there were certain things that seemed to keep coming up. So now I'm thinking to myself, let me go ahead and run this every morning no matter what mood I'm in. And no I don’t hit it perfectly, but whether it's there, whether it's with our software code deployments, whether it's with our weekly meetings that we run, whether it's with how we do daily updates and we actually like … that's what the product is about, is enforcing these, but I really gravitated right toward that. And yet I wasn't expecting to find it at the beginning of the KPI book. So I really enjoy the stories at the beginning of the book around that, and the flying fortress, and how the checklist kind of became a thing. It was just fascinating.
Bernie Smith:
Yeah. It's hard to believe it's an invention of 70 years ago. It's one of those things where once you use it, you think it's from the dawn of time. But someone actually formalized the process, And, yeah, it just stops you from doing stupid things. There's nothing more frustrating than overlooking something accidentally, shooting yourself in the foot in the process. So if it's an avoidable error, avoid it.
Scott Levy:
On the checklist topic, one of my sons had knee surgery, and I remember watching the staff run their checklists. Every single person came in and said, “Right knee, correct?” Even though they knew, even though it's written there, everybody was just keyed in, and you can tell that they were running by the checklist, which is great. You don't want that messed up. And I want the pilot of the plane I'm on or the train I'm on or wherever running those checklists. But I think there's a misconception sometimes that those of us that embrace the checklist, we just enjoy that. And I think to your point, it's a discipline. It's saying no. I just really enjoy the outcome of that and don't so much love leaving it to chance, getting that outcome.
Bernie Smith:
100 percent. I'm not a particularly strong completer, finisher, and I know this. So that is like my defensive shield against forgetting stuff and messing up and not finishing. I know if I make myself get through the process, it's gonna work. I'm gonna be happy. If I don't, I've got this nagging thing in the back of my head. Have I done it right? Quite often, I haven't. I can't remember if I mentioned the book. One of my drivers to adopt checklists was how many phone chargers I had to buy. I was constantly forgetting charging packs and whatever when I was traveling. It's like the fifth time you buy a new phone charger or laptop power supply, you go, okay, I need to fix this. I need to do better.
Scott Levy:
That's it. And we [at ResultMaps] talk a lot about having some form of weekly meeting or touchpoint where you surface issues for exactly that reason. That's a powerful practice. It’s one I see a lot of teams dismiss just whether it's pausing and reflecting to look at our KPIs, which we talked about as a practice, or pausing and reflecting to say, I forgot my charger again. Log it because, oh, I didn't realize that's been 8 times now or 4 times or whatever it is. Instead of grumbling about it, let's just do something about it. How can we make that up? And teams that do that … I've seen this again and again, the teams that have some way they do that - this is whether you're in high school sports, a high performing team in the professional domain, it's less I find about people being these geniuses, kissed by the divine, although we all are - but it's much more about, they run a sort of a checklist like that: “Okay. What did we notice here? This thing slowed down. What can we do to prevent that?” And being methodical enough about it that they're not being arbitrary, like saying, how was I not perfect this last time? What are the patterns I'm starting to see? And, therefore, what ought we to do about them, which is taking me back to why we have KPIs to begin with.
Bernie Smith:
Tools for improvement. That's what it's all about. It's interesting, you were talking about having basically improvement sessions. You review the data. You go, okay. How can we not leave our charger behind 6 or 8 times. One of the things I observed in my former life as an improvement consultant is, it's actually really uncomfortable the first few times that people have to do that, in the same way that forcing yourself to use a checklist is uncomfortable. But if you get through that initial discomfort with the unfamiliarity, that's when it starts to work. When I was working with teams, I'd reassure them. I say, “look. First few times, you're gonna feel a bit silly. This is gonna feel a little bit clunky, a little bit awkward, a little bit unfamiliar. Push through it. It's really gonna help.” It's freeing, right?
Scott Levy:
Just like the checklist. Because now you're not worrying about … It's the same with business. And where we see CEOs and leaders in particular express relief once they see that they are starting to capture that information now and see it one place. Combine that with a regular cadence of checking in on it - it takes you back to why we have these in the first place. We have KPIs so that we measure the right things and get the right results. You had a nice way in the book you explained it.
Bernie Smith:
KPIs are basically the way that we focus our limited time and resources on the things that count. That's it. Because there's always, you know, I've got literally 400 things on my to do list. I can probably only do 10 of those well. So how do we decide what we're going do? We prioritize them based on the things that we care about. So we figure out what outcome we're looking for, we work out what things we can measure that will deliver those outcomes, and then we focus on the things that will make a difference, put very simply.
Scott Levy:
I love that. And there are so many eloquent gems like that in a book called KPI Checklists, of all things. Really, really enjoyed it and really happy that you're sharing this information. It's heartfelt. Everybody who's taking this in, absolutely, go check out all of Bernie Smith's work.
Bernie Smith:
There's more. There's things about gamifying your KPIs, which is another topic near and dear to my heart. It’s very focused on targets and incentive design as well. One of my frustrations is whenever you Google target design, all you get is “SMART”. But if you look at things like the Volkswagen EOD emission scandal or Wells Fargo, they're all SMART targets, and they still broke the organization. I spent quite a lot of time thinking about how we build targets and incentives that actually work in the real world, that real people will use and deliver the outcomes we want. Turns out it's quite a big question, a really interesting one to discuss.
Scott Levy:
It is one of my favorite topics. I have to give a shout out to another good friend of mine, John Humphrey, a friend, mentor. He's got a system he calls ConnectPoints (https://www.theconnectpoints.com). He's got a way he scores when you connect with people, when you reach out to people, when you have a meeting, when you record stuff but what I found is it's a blast. What I like about gamifying, where I'm going with this is how when we're clever about it, it makes it much less of a grind. You and I probably think of the checklist as this is fantastic because this is saving me. Whereas for someone else, it's oh my gosh, more stuff I have to do. So if you can connect the incentives and everything, I can see where that's just incredibly powerful.
Bernie Smith:
And there's all kinds of interesting research around variable reward and things like that, where you discover that if you, for example, apply verbal reward, you get dramatically more motivation from it than fixed rewards and things like that. Yeah, fascinating topic and some really good material out there. A lot of it is quite software focused, so you need to pick and choose a little bit for KPIs specifically, but an awful lot of it does work and is relevant. If people are hesitating and thinking about KPIs, imperfect beats perfect but not implemented. So my suggestion would be get out there, do something, figure out what's gone wrong, do better next time, don't worry about being right the first time. What's it? Perfection is the enemy of action or something. But basically, go out, give it a go because it's an iterative process. It's not a one-off. As long as you don't do anything crazy it's pretty hard to get wrong, and you will get better and do it better the next time around.
Check out Bernie and Made to Measure KPIs
KPI checklists (Bernie’s very good, easy read on KPIs)
What is a KPI - 60 second video https://madetomeasurekpis.com/what-is-a-kpi/
Structured Approach to Enterprise KPIs https://madetomeasurekpis.com/roks-kpi-method-overview/