Real leaders leave a legacy. They capture the hearts and minds of their teams. Their origin story puts the safety and well-being of their people first. Great companies ubiquitously have safe, yet productive operations. For those companies, safety is an investment, not a cost for the C-suite. It’s a real topic of daily focus. This is the Safety Guru with your host, Eric Michrowski, a globally recognized ops and safety guru, public speaker, and author. Are you ready to leave a safety legacy? Your legacy success story begins now.

Hi, and welcome to the Safety Guru. Today, I’m very excited to have with me, John Richardson. He’s a professional incident investigation consultant working for an organization called Bywater out of London, England. John, welcome to the show. First, I’d love to hear your story about how you got started in safety because it’s actually quite interesting.

Yeah, well, thank you very much for having me, Eric. Yeah. So how I got into safety, I suppose, funnily enough, I always start by saying that I’m not formally a safety professional. I think that as an incident investigator, I get involved with the safety incidents, with the safety functions of organizations, it’s not so much my specialism, but I suppose I’ve got a subspecialist within safety. I always describe myself as a professional nosy parker. I like to figure out what’s happened, why things have happened, obviously for the purposes of improving them. But I started off down that road professionally, I suppose quite a long time ago now, I suppose, back in 2008. I actually started in law enforcement. I joined what at the time was Grampian Police, based in Aberdeen in Scotland, which is actually where I still live at the moment. And I was in law enforcement for, well, from 2008 up to the beginnings of 2015. I did various different things within the world of law enforcement, the uniform, the frontline uniform, the more serious criminal investigation. But again, it was volume incident investigation, volume incident response, 10 investigations a week, 20 statements a week, all of the core skills that people need to do investigations. We’re in quite a unique position here in the northeast of Scotland, where we’ve got the North Sea oil field off the Coast. Which means when something serious happens, the police obviously get involved in investigating. The police in this area of Scotland have a quite a heavy involvement in investigations. I would imagine. This is the company I went to join when I left. The police were started by a former senior detective who had spent his career obviously investigating the really bad criminal stuff, but also all of these industry incidents, the fatalities and the big stuff that happens out there in the energy industry. What he’d observed in his career was that, yeah, if you want the black stuff taken out the ground 100 miles out to sea, you get an oil company out to do that. They do that very, very well. What they don’t do so well is investigate when things go wrong. He set up a consultancy when he left the police, basically to help oil companies do this. As the company grew and it start to expand, he coaxed myself and another young detective out of the police to go and join the business. And that’s how we got into it. So, over the course of the last 10 years, I’ve worked for a couple of different organizations, a couple of the different proprietary root cause analysis brands. That goes very closely with what we do. And I’ve done the consultancy, the actual investigating of incidents, and a lot of training around investigation with obviously oil and gas, but different industries all over the UK, all over the world. I’ve been down into Africa to do this stuff, out to Asia, the Middle East, the United States, and different industries. We’ve done obviously energy, manufacturing, civil construction, rail, waste management, you name it, wherever there’s the potential for things to go wrong. We’ve probably gone and done a bit of I work there. And now here I am, 2025. I’m working for, as you said at the start, a great company based in London called Bywater, who do management system training, among other things. But in the last year, we’ve redeveloped the Root Cause Analysis program, and it’s going great guns. So, I’m spending a fair bit of my time on the road to go across all sorts of different industries, delivering instant investigation training.

Tell me a little bit about what makes a good investigation. And What do organizations that are good at investigating, what do they do differently?

I think the first thing that is really essential, and this really sets apart the good ones and the bad ones, is the desire. The desire to actually get to the root cause and fix it so that it doesn’t happen again. I mean, that might sound really obvious. Well, everyone wants to get to the root cause. But I think there’s an awful lot of organizations who they want to do an investigation to show that they’ve done an investigation, but they’re not that fussed on the outcomes, and they like to keep doing what they’re doing. But I think the ones who do it really, really well and the industries who do it well are the ones who really want to see improvements. And I think probably the ones for which improvement is really a critical part of the business. 

We talked about when we first connected, sometimes there’s a need to even commit to a skill around it. I think you referred to it as an investigation function. Tell me a little bit about how an organization can start building that.

I’ve seen all sorts of different approaches. Again, they’ve got to suit the organization for sure. Again, having the right people with the right skills is really key. Let’s face it, this is a bolt on skill for most professionals. What I usually say to people when I walk into a training room, if you had no incidents in your organization for the next month, are any of you going to be sitting around with nothing to do? And they all laugh and go, No, of course not. And it’s like, Yeah, because you’re busy. You’ve got a full-time role to do. But you’ve got to get people, first of all, to give them good skills Again, another common occurrence in my life is I go in a training room, and I say, who’s been doing investigations throughout their career? And all the hands go up. And I say, who’s had formal training in it? There’s maybe one hand at the most. Sure. So, you’ve got to get people, I think, to the skills, but also practice the skills, also be doing this stuff on a regular basis. I always say it’s not rocket science, but it’s a skill that has to be learned and it has to be practiced regularly.

Otherwise, you get skill paid. So, I think a key thing here is identifying the right people in the organization to actually do investigations. Who’s best placed, who’s going to have the resource? Who’s got the mindset for Because not everybody’s got the mindset for it. And then make sure that those people are doing it regularly. Now, that means, as we said, a function. And I think the best industries are the ones who have a function whose job is to investigate incidents. Those people aren’t often popular, but if the culture is right in the organization, they are popular because everyone knows why they’re there. Or as I I say, if that’s not an option in a business, you’ve got to make sure, like I say, you identify the right people, you give them the right skills, and you give them enough time to actually do this. But that comes from having the desire as an organization. It’s culture.

It’s culture. How do you go about interviewing key stakeholders? I think you revert to them as witnesses. Tell me a little bit about how you go about getting it, some of the risks, the biases that come in through the interviewing process.

Okay, well, there’s something you mentioned there, stakeholders. I think that’s an important thing to identify at the start of any investigation because there’s going to be a lot of stakeholders in the sense. There’s going to be a lot of people with an interest in the outcome of this investigation. Sure. But not all of those stakeholders have anything of value to offer. Some of them are, and quite often these are people in maybe your management positions, senior management positions, didn’t have anything to do with the actual incident Itself. But tell you what, they want to come down and tell you what the thing happened, and who they think was involved and how they think you should be doing it. So, I think the first thing before you even go near the doing interviews and such like, is identifying who the people are who can really help you. Sure. And that comes through the investigation. But in terms of good approaches to interviews, I think, again, I’ll keep coming back to it. You’ll be sick of hearing me say this, but it starts with culture. It’s really challenging if you go somewhere and it’s all… You get the usual management plan to you.

It’s like, oh, we want to make sure this never happens again. And everyone’s sitting there thinking, all right, that means someone’s going to get sacked. So, you’ve got to have the right culture. That’s going to make life much easier for an investigator. But I think, again, skill and practice, learning the skills, there’s a structured approach that we use in law, or we did use in law enforcement. It’s actually used to this day in law enforcement investigative agencies all over the world called the Peace Model or the Price Model. It’s a way of breaking down these fact-finding conversations. It starts off with preparing well for your interview. What is it you want to ask people about? What are the key topics that you need to get into? What are the gaps in your investigation? You want to figure that out beforehand. Now, if it’s a really serious incident, it might take you quite a long time to prepare for it. If it’s something really minor, you could literally probably write your plan on the back of a cocktail napkin, but it’s still good to have a plan. So, you need to have a plan about what you want to talk about or sure, and make sure that you’re going to be going and doing your investigation in the right place, where no one’s going to be listening in.

It’s a neutral environment, just setting yourself up for success, I think. And that’s before you even ask a question. Then you go into the next phase in the peace model is engage and explain, where this is your initial interactions with that person. And this can make or break a conversation. Even if it’s, like I say, a 20-minute conversation over something minor or a three-hour interview over something really serious, set your stall out right from the start. First impressions count. So, explain why you’re there. Tell them what the purpose of the interview is. Ask them if they’ve got any questions. Make small talk, because a lot of people, the wall is up when you go in. Of course. If you can speak about the weather, speak about the weather going on their vacation this year or what sport and events they’ve been to. Just get people chatting about anything not related to the interview. And It works really well. I think building rapport is really important. Certainly, in the law enforcement environment, if police officer, I say, having to go to speak to a child or a very vulnerable person, they sometimes spend days on an engagement explain to build up a relationship.

Now, we’ve probably only got five minutes at the start of an interview. But if you shoot the breeze with someone for five minutes about nothing really that important, all of a sudden you go, right, should we get into it? And they’re like, yeah, let’s go to it because they’re just Great. Then the meat of the interview, the center of the model is account and clarification. That’s the long bit. And that’s basically just working through your ran, going down those rabbit holes, finding out what they’ve got to tell you. I always suggest as well, when I’m writing my interview plans, it’s very easy to write a plan where you’ve got no scope to talk about anything else. So, you could almost imagine it like if you tell Google that you want to go somewhere and Google says, okay, go 100 yards, turn left onto whatever street, go 1.3 miles, go right to the roundabout, and eventually you get to your destination. And to be honest, you hardly need to look where you’re going because it just tells you. You probably don’t see much along the way. I think an interview plan should be more like a list of waypoints.

Here’s where we’re starting. Here’s our destination. Here’s five places you might want to go on the way. And if you choose to go and visit some other places, if you want to take the scenic route, if you want to go the long way around, if you want to take a shortcut, feel free to do that. But choose any way you want to go. You’ll have a much more interesting journey You’ll get a lot more out of it than just following this direct question. So, I think it’s important to make sure you do just let the conversation flow. As I say, go down rabbit holes, pull on the threads and see where it takes you, because you never know what’s going to fall out of people’s heads sometimes. So, once you’ve done that, and that’s the longest part of your interview, I think, we do what we call confirm and close. So, we basically just… This is where the emphasis on speaking and listening changes. Obviously, when you’re doing the first part, the account clarification, it’s 90 % speaking, 10 % you. That will then flip, and you basically just go back what through what they’ve said.

And it doesn’t take very long. It can take you a couple of minutes just to confirm that you’ve captured it the way they hoped. And then try and close out your interview on a positive and say, all right, thank you very much. Any questions for me? Usually, it’s, Am I going to get into trouble? No, we’re not here to pin blame. We’re just looking to get to the root causes. And then I usually say to people as well, look, if I’ve got any more questions, can I get in touch with you? Which they’re usually pretty cool with. And then the final bit, and this is often something that’s quite overlooked because you often got quite a few people to interview as well. So, the final part is the E in the peace model is evaluating. And all that means is Because usually, if you’re doing some investigation, you’ve built a timeline of some sort, a graphic timeline. Go back to your timeline and look at it and say, Right, what has this interview told me? What gaps have been filled? What I have to find out. And then what questions might I need to ask somebody else along the way?

Sure. Again, it’s often overlooked. And then you go, and you do three interviews, and then you read back the first one, you go, I forgot to ask this person about this. I forgot to ask that person about So after you’ve done any interview, as I say, whether it’s a 20-minute or a three-hour, just take a breath, look over it and say, has it meant what I needed it to do? Has it given me what I needed it to give me? 

It You talked about root cause and getting to root cause. How do you know you’re done interviewing, that you’ve got all the insights that you need to get? Because often what I see is an investigation or you’re trying to capture the learnings, and then you go fairly deep, but you don’t necessarily get to all the system factors that contributed to the event, mostly because people don’t have the time or it’s convenient to stop at a certain point.

From exactly what you’ve just said there, that comes back to the desire of the organization, if people haven’t got time. Linking root cause into interviewing. Interviewing is usually a formal interview. It’s something I do towards the end of my investigation, so very much just before root cause. So, what that means is by that time, I’ve already built up quite a detailed picture of evidence. I’ve already maybe started to identify where these system factors exist through my initial evidence gathering around documentation and records and maybe examining the scene and maybe some initial conversations with people and subject matter specialists. My actual interviewing of witnesses more often happens towards the end of the process, which means you’re in a much better position from a preparation point of view to know what those system factors potentially are and jump into them. It shouldn’t be a go and blind thing. Often what happens is that initially after an incident has happened, you’ll maybe note a few initial accounts from people. All they’ll usually tell you is the five minutes before the incident happened, because that’s what I was thinking about at that point. And that helps you build up your picture.

But then towards, as I say, the end, these more detailed conversations mean that we can get into those system factors. We can get into those more managerial management system problems in a bit more detail, or let’s just say, be a bit more aware of them, I think. Going back to your question about how deep you go, how do you get to the root cause, again, it comes back to that desire by the organization. But I think the main theme here tends to be proportionality in terms of what’s happened, consequence versus potential reoccurrence, your standard risk assessment type thing. What the injury is, what the outcome is. It would be easy for me to say, oh, you should get to the nth degree on root cause analysis on every investigation. Hey, wouldn’t that be lovely? But who has the resource for that? Nobody. So, you pick and choose. Sometimes you may go to a very superficial level of analysis on a very low-level incident. You may be just getting to root cause categories or root cause subcategories. And that’s absolutely fine. Like I said, we’ve not got the resource, nor has anybody probably got the interest to do a six-person investigation team onto somebody who scratch their finger.

But this, for me, is where the data is really, really important about gathering data about the types of incidents you’re having, but also gathering data about your root causes that you’re having. And I guess we’ll probably speak about root cause, taxonomy, and lists, because This is, I suppose, if we go back to the law enforcement life, this is where we say it could be intelligence-led. You might have what on the face of it is a fairly minor incident. But if you’re able to look at your data and go, we’ve had six or seven of these very minor incidents in the last year, then maybe you say, Right, we’re going to put more resource to get to a deeper level of root cause. If you can even go the other way and say, well, we’ve had what is maybe a quite serious incident. Maybe it’s a high potential near miss, maybe it was a serious injury. But we’ve genuinely assessed it and found that the likelihood of that happening again is very, very low. So, we’re not going to go into as much detail. But having that data and risk assessing allows you to make a decision on what level of root cause you want to get to because you’re not going to go super deep every time.

Let’s go a little bit deeper on this taxonomy piece. How do How do you start building the right taxonomy for the organization? Is it something that is system-driven, that you just copy and paste from somewhere else, or does it need to be adapted from the organization to truly work?

A mixture. When you say copy and paste from someone else, there’s lots of lists out there that vary in length. You can go on Google and probably find a list very quickly. I’m sure the likes of ChatGPT would write you a really good list at the moment as well. I’ve been playing around with that actually, and it’s quite cool, some of the outputs you get. Again, over the course of my career, the last 10 years of really being into root cause analysis, I’ve seen different variations. There’s a really good tool that I used to… The company I used to work for a good tool, and they had, I think, 64 root causes in their system. Another American company I actually worked for, they had like 103, I think, in theirs. There’s one company who I did some consultancy work for to develop a list, and they actually basically took the list from a BP list, which is out there on the internet somewhere where 180 plus root causes. And that’s probably overdoing it a little bit. So, what I do is I have a list that I use as a starting point. It’s quite broad in many ways.

And then what I’ll do with the client is I’ll sit down with them, and I’ll say, Right, this is our starter. Let’s adapt it to make it suit you, because I think that is really crucial. Got to have a list that suits the organization. I’ve used a couple of them, but I don’t believe in a one size fits all approach. If you really want to get that buy in, it’s got to look and it’s got to feel like the organization, whether that’s the terminology, the way things are phrased, certain areas of focus, if certain types of activity they’re involved in, then you maybe have certain different root causes in there. But I use this starting list, and then I’ll work through a bit of a process of some mock RCA’s based on case studies, and I’ll say, let’s try and use this list and we finesse it and we adapt it so that what we’re looking to get is consistency in the sense that two or three different people could look a case study, maybe even interpret it slightly different ways. But when they use the list, their outputs, their root cause selections are pretty much similar.  

It’s never an exact science. They get pretty close to where they need to be. Sure.

I think that makes good sense because if the list is too long, then you’ll never find trends. And if the list is too short, you’re going to find false trends in your data as well.

Yeah. Or just really superficial stuff. When I put the list in front of people your first time, I think I need to read through it, and I’ll say… And this list, I think I use is about… It’s a starting point is about 100 and 405. And like I say, sometimes we add more, sometimes we take some out. But the first thing I’ll say is, is there anything you noticed in there? And it covers I actually have it. It covers six broad categories. We talk about management system causes, process documentation causes, equipment causes, environment causes, task execution causes, and then personal causes. And look across those categories. It’s designed so that you don’t have to read all of them every single time. But often when I say, do you think there’s anything missing there? And they’ll go, I don’t think so. There’s quite a lot in there. I’ll say, what you probably won’t see in that list is terms like human error, complacency, recklessness, carelessness, because all they are termed subjective judgments of a behavior which really does nothing for me as a whole analysis specialist. It’s like, how do you fix complacency? I don’t know.

It’s on a root course. It’s looking back at something.

This episode of the Safety Guru podcast is brought to you by Propulo Consulting, the leading safety and safety culture advisory firm. Whether you are looking to assess your safety culture, develop strategies to level up your safety performance, introduce human performance capabilities, reenergize your BBS program, enhance. Enhance supervisory safety capabilities or introduce unique safety leadership training and talent solutions. Propulo has you covered. Visit us at propulo.com

How do we address system factors? We talked a fair bit about the importance of getting to system factors to dig deeper. How do you go about making sure we really find them and then address them?

This is where the list comes into play as well. Making sure that you’ve got plenty of system factors in your list. Because chicken and egg here, what the list actually does is it can focus your evidence gathering. So, if you know you’re going to have to look for or potentially identify system as in your analysis, then you should be thinking about that when you’re doing your evidence gathering in the very first part of the process. That list of categories that I gave you there, that’s the list that we… That’s the order with which we actually work through them. It forces people to think about the system first. I think, and again, there’s no perfect answer to this or no silver bullet, let’s say. Again, it comes back to the culture, the desire of the skill of the investigator as well, to ask the right questions as well, because that’s a big part of this is the interviewing. It’s all very well asking people what they did, but you’ve got to get to the, So, why did you do that? Was it because there was time pressure? Was it because the procedures you were using hadn’t been reviewed?

Is it because you usually have six people on the team, you’ve only got four today, we’ve got people off long term? Whatever it is, I think I would say from a personal investigator point of view, you get to system causes with practice. Practice, practice, practice. And again, that organizational desire to do that.

To go deeper.

To go deeper, I think, it starts at the top. Yesterday, I was down in England training a company down there who run a… They have a glass bottle factory, and they make hundreds of thousands, probably millions of glasses bottle a year. And they sell them to pharmaceuticals and the license and trade and brewing and all these different companies. It’s the third time I’ve been to see them. And I was there, I think I was there in October. And I did about 10, 15 people. I went again in December, and now I’ve just done the final course. Now, across those three courses, I’ve pretty much had the whole of the leadership team through the trade. I Yesterday, I think I had the marketing director, I want the finance director, the procurement director. I’ve had the quality director. That is an organization that is interested in this stuff from the top to the bottom. To have their executives, the time to attend this training, that’s the thing that really helps you get to those root causes, because that means that they’re then expecting to see them. I think that’s That probably is a really big driver. That doesn’t mean some organizations don’t do that.

Sometimes we’ll do maybe awareness training for a leadership team just so they understand terminology and things like that. But I’ll tell you what, if I had a pound for every time I’ve been in a training room and somebody said, oh, you know what? I wish my manager would go on this course. I wouldn’t be talking to you. I’d be lying on the beach in Tahiti because I’d have retired. I’d be a rich man. So yeah, I think getting management involved in this is a top to bottom thing. Yes, you should have people who skill it is. The understanding of the process, I think, has to run from the boardroom down to the shop floor.

Why don’t we share maybe some examples? You’ve done a lot of investigations. Tell me a little bit about what is it that really matters in terms of the investigation? You had two stories from a regulatory standpoint, very similar. Tell me a little bit about those examples and what made them different.

I think for me about transparency. Whether that is to the regulator, whether it’s to your customers, whether it’s to the rest of the team, just within the business, a transparency and an openness about when things go wrong, how you react to them. Basically, what occurred was this was a few years ago, I think 2023. I was involved in two investigations, two oil and gas industry investigations within a few months of each other. The first one was quite a serious incident. Now, they were both what you would, I suppose, call near misses in the sense that we know he got hurt. But the first one was quite a significant gas release. We’re talking significant enough that if there were maybe one control measure away from a multi-fatality disaster, we’re talking a serious question here. When this happened, they were fairly certain they were going to get a lot of heat off the regulator. It could have ended up in the media. First thing they did was instigate their own investigation, and they started a lot of detailed evidence gathering. Then they got a consultancy to come in, and it was my job then to go in and help them investigate the incident They did a pretty good job themselves, but then work through the root cause analysis program and process to make sure we got to the sufficient level of analysis.

And as I say, it was very, very open. And that’s Funnily enough, that’s something I tend to find more often with the more serious incidents. People are quite open about it. But we did that, and we got a lot of good learnings out of it. Now, halfway through the investigation, our regulator, the Health and Safety Executive, actually came back into this company and pretty much said to them, folks, you’re doing a great job. You were really pleased with your progress on the investigation. It looks like you’ve already put some good measures in place. You’re clearly taking this seriously. Carry on with business. And I mean, this company was staggered. They were like, I mean, are you not coming to do your own investigation? Are you not coming to see us? And they were like, no, no, no. It seems like you’ve done a good investigation yourself and well done. And carry on. And that was pretty much the long and short of it. And I don’t think it went much further than that in the end. Contrast that to about three months later, they go into another organization where they’ve had a release as well.

Their release wasn’t It was a different substance. The level of what they released was much… It was so low that it was never going to be fatalities. Never. We’re talking maybe itchy eyes and a bit of coughing. But they had these releases, and they didn’t actually report them to the regulator at all. They deemed them non-reportable, but they were including them in their weekly reports to the general operational report to the regulator who said, came back to them and said, you said you have this release. Are you going to formally report it to us? And they were like, no, we don’t think we think it’s an operational incident. And the regulator came back to them and said, we maybe think you should look at that again. And they looked at it again. No, this is the asset. It’s mine. We’re not investigating it. It’s not an incident. We’re not looking at it. It’s just operational occurrence. And the regulator came back and said, Right, there’s five different things that we want you to check. We want you to tell us if you’ve got these five measures was in place. And they went, well, we’ve got these two.

We’ve got a little bit of this one, and we’ve actually not got any of these two. And the regulator basically slapped them with an improvement notice, a regulatory enforcement action. This company had never had one of them before. By the time we got in there, when I went in on the previous one, we were looking at the production department. See this one we went in? Basically, it was the managing director who called us and said, you can go anywhere, speak to anyone and do anything because we need to get to the bottom of what happened. It was great that eventually they got to that point. But wouldn’t it have been great if they’d done that in the first place and avoided the regulatory attention and the enforcement. So, I think the transparency and that willingness to say, hey, things have gone wrong. Let’s find out why and not have the same thing happen again is absolutely crucial. That’s your win. I think I think as well, the companies who do that well, competitive advantage for sure.

It’s transparency, but it’s also getting to truly trying to be a learning organization to trying to learn from the events as opposed to, let’s move on.

Yeah. I think as well, look, I know organizations, they have operational pressure. Everybody, you can’t avoid that. What I have noticed, though, as I say, over the last 10 years and in my wider reading and observations, and there’s no science behind this. It’s more of an observation. But what I tend to find is you go across all the different industries. The deciding factor for me in the transparency in the learning culture, as you say, comes down to how many people you can kill in one go. If you’re looking at aviation, if you’re looking at nuclear, if you’re looking at offshore oil and gas, where you’re 100 miles out to sea on a production platform made of things that don’t float full things that explode, and everyone lives upstairs. Those industries tend to be, by and large, really, really good, or let’s say, better. You go down the other end of the scale, things like construction and waste management and even medical, the medical world. How many people is a doctor ever going to harm in one go? One. They tend to be the ones who don’t tend to do quite as well as it and don’t have the open cultures.

But make no mistake, there are within all of those industries, across the board, you do get pockets of excellence where they really will investigate every single little incident that they have. Safety, quality, environmental, you name it. Because all those, the causes are exactly the same. They’re system causes.

John, thank you so much for sharing some of your insights on an incident investigation, some of your techniques from the police service and applying it in industry. If somebody wants to get in touch you, what’s the best way to do that?

They can get me on email. They can hit me up on LinkedIn. I’m on LinkedIn. If you search for John Richardson, Bywater. I’d be delighted to hear from anyone, even just to shoot the breeze and talk investigations. I’m always happy to meet new people and help people out.

Thank you, John.

Thank you for listening to The Safety Guru on C-suite Radio. Leave a legacy. Distinguish yourself from the past. Grow your success. Capture the hearts and minds of your teams. Elevate your safety. Like every successful athlete, top leaders continuously invest in their safety leadership with an expert coach to boost safety performance. Begin your journey at execsafetycoach.com. Come back in two weeks for the next episode with your host, Eric Michrowski. This podcast is powered by Propulo Consulting.