You’ve Solved the Wrong Problem (ft. COMET, Pt. 2)

Most investigations don’t actually prevent the next incident. They document what happened, assign a cause, and then

close the loop. So on paper, everything looks complete. But then weeks later, months later, even a very similar

incident happens again. And it raises an even bigger question. Are we really learning from investigations or are we

just getting better at explaining incidents after the fact? Because if the outcome is always human error, if the

focus is always on what someone did wrong, then we’re probably just missing something much bigger in the system.

Welcome back to the safety brief by intellects. I am your host Monica and as always I am joined by my co-host Scott

and today we are joined by Craig from Comet. Craig, it is so great to have you

here. We had Mark on last time, so it feels like we’re meeting the uh other half of Comet. Do you want to just take

a moment to introduce yourself here? I’m not sure he would agree with that, but no, delighted to be here, Monica.

Thank you very much. My name is Craig Smith. I’m head of operations at Comet. And yeah, delighted to add to your

discussion today. Thanks for having me. Happy to have you here, Scott. How are you doing today?

Hey, I’m great. vice president safety and health here at Intellects and I’m always looking forward to these type of

discussions. So let’s get started. Lovely, lovely. So last time when we chatted with Mark, we got into why so

many teams are measuring the wrong things and how they can create a false sense of confidence when it comes to

preventing some serious incidents. Today we’re going to be taking that one step further because even when something does

happen, even when there is an incident and an investigation, a lot of organizations still don’t really get to

the real answer. So today’s episode is going to be all about that question. Why

do so many investigations fail to actually prevent the next incident? So Craig, I want to start here. In your

experience, where do you think most investigations go wrong?

Uh yeah, I mean it’s a great it’s a great introductory point I suppose. Um having having been involved in in different types of investigations now

for for almost um gosh where are we now? Uh 18 years. I know I don’t look like

that’s the case but yes h for for for quite a while now across different sectors. I think where we often see it

going wrong are super superficial outcomes uh versus real outcomes. Right. And and I think organizations that that

do this well will take a a strong uh close look at themselves in terms of

what what not not not just at what has happened but but why has it happened and how do we truly learn from that to try

and eliminate that repeat occurrence. I think that the alternative is where organizations um are a little bit more superficial

with their views on on what has happened uh and and probably opt for more easier shorter term fixes, right? which which

might not necessarily be wrong, but are they truly addressing the heart or the cause of the issue long term that then

becomes questionable. So I think it’s about not addressing the true underlying causation of the incident and looking at

things a little bit superficially that I mean I could see a number of things probably but I think that’s a nice way for me to kick that off is not going

deep enough often enough. Yeah, Craig. I mean it’s uh it’s a dead end, right? because

we’re looking to close an investigation at the exact moment when it should be

opening and and you know that’s how I always kind of think about this when we kind of talk about about it the way that

you’ve described you know once we’ve kind of gone through this process we’ve named a person as a cause you know the

system gets walked away right unexamined and uh many times right the cost of

stopping at that spot is at the same incident going to show up over and over

and over again. And the only thing that I cannot do is predict when it’s going to happen. You know, maybe wearing a

different set of clothes or it may be a different spot in a facility. Uh but it’s going to show back up, right? If we

really don’t find the root cause of what this thing uh happened right right there

in front of us. Well, well, well, yes, exactly. I mean the surely the purpose has to be not not

just to acknowledge the fact that this is a one-off right it might it might present as a one-off incident but

without looking properly at the characteristics of that event you’re

probably uh losing the opportunity for for learning from has that happened elsewhere and other situations could it

happen again that the objective has to be about try to eliminate repeat repeat occurrence and um you know if we don’t

have that goal then probably we’re not going to get the heart of what really calls it. Agreed. Agreed. I mean, that actually does get

me thinking and maybe we could take it one step back of what I’m hearing is that a lot of

investitionians are pretty good at documenting what happened but not why it

happened. Can we talk about that difference for a moment? I guess from both of your perspectives.

Yeah, I mean I’ll Yeah, absolutely. We I’ll take this initially, Scott, and you can you can add some some some some some

forensic detail that will add a lot more value, I’m sure, than I will, but um the fundamental principle is the what to the

why in any investigation. And um really really not just looking at it from a surface perspective. Sometimes, you

know, even even if you have something relatively lower level in impact, there can be a habit to say, well, we only

really need to fix what happened there and then, right? And maybe from a short-term perspective, that’s okay, but but we’re still missing the why. And the

why has to move away from whatever that particular system failure was or whatever that human error might have

been but but understanding what and why that was allowed to happen. What were the error inducing conditions or traps

or influences that allowed that to happen. So yeah, a good investigation has to acknowledge the what is merely

the starting point. That’s the easy bit. H we have to get right and without doing that well you know good luck uh in

fixing that in in the fullness of time. Yeah, I I you know I I totally agree

with with Craig. The what is just easier and you know I I’ve been practicing for 36 years and I started my career

probably looking at what happened versus why it happened. And you know you can timeline it, you can photograph it, you

can diagram it, but the why really requires judgment. And I I I’ve often

kind of characterized this. It requires curiosity, right? It requires a comfort

with a level of ambiguity uh that a lot of organizations are just

not up for. You know, they they go straight to the why. And many times I use, you know, this as an example is

that I don’t know how many investigations that I’ve looked at that they put the guard back on the machine

after someone had been injured. And my entire question set was, well, why did we take the guard off? Right? So, so

there’s a difference between what and why and it feels more comfortable doing the what because it’s easy to put the

guard back on the machine. Uh, as a metaphor, right, in what we’re talking about, I think there’s also a speed

problem. You know, leaders want reports closed. We don’t want to spend a lot of time. The timeline of events can be

written in a day. Understanding why those conditions existed takes maybe

months or weeks or days. And and so I I think that there’s a time element of

this. I know when I started I had 48 hours 48 hours to start an investigation

and complete it. And I don’t know how many colleagues that I have that have been in that same spot of of these

artificial frameworks to understanding what really happened. And I think you know to kind of benchmark or or to to

base this conversation from what Craig said, it’s politically a whole lot safer, right, to just tell you what

happened. So I guess my test of all of this is is if your report can be copy

pasted into any other incident in your injury in your industry, you probably

haven’t gotten down to the Y level. And and that’s really what I think as a practitioner I really need to understand

to make sure I can eliminate it or mitigate it sustainably, you know, from the work system.

So Scott, is it safe to say that this is kind of chocked up to a time constraint

issue or would you say it’s more of a mindset issue? I I I think that we tend to I think it’s

a little bit of both. I think there’s some artificial frameworks that we have developed for organizations to work in.

I think when you look at the complexity of investigations, uh you’re you’re trying to get to an end

and in in a what I would say an efficient manner. And that’s why many times you stop at at what I would call

the immediate, you know, causes of of of an output, which are the symptoms,

right? It does feel good to put the guard back on. And and when you kind of peer below that, you start asking

different questions. And that’s when it gets very very uncomfortable. You and I, you know, said this just a few moments

ago. Uh when we start looking at the why, it requires judgment. It requires

curiosity. It requires ambiguity uh with with the organization. But what it

really requires us I think and Craig you can talk more about this is an uncomfortable part of our organization

where maybe there was production pressure or maybe preventive maintenance was not done or accomplished on that

particular asset that allowed some of these things to happen. So I I think there’s a lot of things and organi

organizational dynamics that kind of play into this. uh but I think for a strong practitioner right a lot of our

responsibility is to make sure that we not only understand the what but the why

that these things happen and being able to have that conversation with senior leaders in an organization.

Uh yeah I mean I would add to that I mean I’m busy taking my own notes on that Scott. That was really that’s

really interesting. So I’m I’ve got no doubt people that are watching and listening will get some get some takeaways from that. It’s really useful.

But I I you know dare I make a comparison here from if if we go back you talk about you talk about outcomes

being uncomfortable right and and how do we really know we’re getting to the heart of where we need to improve from

and I think that I think that that that view is absolutely right if you an investigator walk in to the you know the

management team to present those findings after investigating for a week and you’re really happy and joyous

because you know that what you’re presenting back is going to cause no issues no discomfort and it’s going to go swimmingly well then probably you’ve

not dug deep enough or you you’ve maybe avoided something for that particular reason, right? Because typically these things are not that comfortable. And

actually I remember and and data makes comparison to when we responded to COVID, right? I hate to take the

conversation back to that unnecessarily, but I always remember watching one of our government do things on COVID and our first minister at the time spoke

about if you don’t feel uncomfortable in your day-to-day life now with the changes that we’ve introduced, are you

doing enough? are you doing enough to change to adjust your lifestyle to adjust your behaviors? And that really

resonated with me because I knew I wasn’t. I was sort of thinking, you know, this is okay. I’m still going out for walks and I can do this and that.

But the reality is I was pushing those boundaries. The comparison I would make there is actually, yeah, if we’re

investigating properly and we feel a little bit uncomfortable about where we are, probably we’re in the right place.

Um, and we have to be brave enough as a as a as a practitioner. We have a duty

when we’re investigating on behalf of the organization to report properly. Um, we often get asked to go in as third

party independent um, experts to help organizations out. More often than not, we present findings that aren’t that

well accepted, but they are the true findings. So, we have a responsibility to do that, right? And it’s the same

with internal investigation teams as well. Discomfort should be seen as a good place to be. And I would I just

want to pick up on that specific point, Scott. I think that’s a really good take. No, totally. And for those of you who

don’t know what root or sorry, what comet does really well is that root cause analysis and really looking into

what things look like in practice. So, building on your point, what does strong root cause analysis actually look like

and building on that discomfort? Well, I mean, strong root cause analysis

has to be first of all, it has to be consistent. I mean not organiza we hear a lot from organizations that have got a

lot of you know crossf functional teams different different verticals different business areas and they all do things

differently and and they all probably think they’re doing it the best h and maybe they are maybe they’re doing it really well but there’s no real

consistency and so when we talk about root cause analysis it’s it’s two things first of all are you are you working

with a methodology and a structure that gives you outcomes when you need them so

when you deploy it for a single investigation at a point in time at a particular site are you getting the

findings and the outcomes that helps you fix that issue. That’s the first thing. That should be the easy thing. Um the

second one is well what are we learning from that? Okay. So for a site at an organizational level, what are we doing

with that information? What are we doing with that data? How does that how does our root cause analysis system approach methodology help us learn at scale? So

it’s not just about having something that you can take out of the cupboard when you’ve got to use it and dust it off and get some learnings and move on

from it. It’s what are we doing with that all the time. So how is that root cause analysis machine giving us the

insights organically regularly to say okay well actually this is becoming a trend because one investigation at one

site will learn one thing it’s amazing what other sites are also learning if we’re not joining the dots on that

information and that intelligence uh as we would describe it then really are we are we really benefiting from root cause

analysis so it can be easy to think we’re effective if we’re operating in silos but and there might be good reason

for that but if we are doing that we’re probably not learning at more a macro at scale level and we are definitely losing

out on a lot of opportunity. So, so effective RCA looks like doing it well on one-off occasions uh Monica, but you

need to have the ability to look at it as a whole program or discipline as well. Many organizations aren’t as good

at that as as they might think. Yeah, I I would add on to that and I

think it’s one of the uh the most positive things about our partnership with with Comet is that I don’t know how

many colleagues that I’ve had throughout my career and now working with

intellects where I see a lot of our clients using some of those those same uh frameworks. I I I would just, you

know, and not to pick on fivey, but fivey has been fine. [laughter]

Well, I I I would pick on it this way is that if I were to lead a 5Y during an

investigation process, I’m very convinced that I could do it very well because I know the right questions to

ask to get us to the beginning to understand what root causes were. But

what I what I’ve seen is that we have used that process and we’ve asked the wrong questions to try to get down to

the root cause. So it’s very incomplete. I think it’s one of the things and I’ll embellish this is why we strategically

partnered with Comet because we needed to have more framework, more methodology

of getting down to why things actually happen. And it really kind of goes down to in my mind to where the work actually

happens. uh you stand where the worker stands when you’re doing root cause analysis properly. You talk to the

people who do that job every day. When you talk about this uh with the benefit

of understanding root cause, you ask about workarounds. You ask about time pressure. You ask what normal actually

looks like versus what the design said it should look like. So we just keep

pressing through these comfortable answers that many times we accepted

right over time. So failure to follow a procedure is never going to be a root cause. Uh it’s a prompt to ask why that

procedure did not work for them. So that’s root cause and and I I’ve seen this happen over and over and over

again. And I think it’s because it feels like it’s easy. Uh but it’s not easy. If we don’t find the root cause that

actually sustainably helps us mitigate or eliminate that issue, then we’re we’re shortsighted, right? In in what

what we should be doing as practitioners. And I I think it names the blunt end. You know, Craig and I

have talked about sharp end, blunt end, right? If every corrective action lands on the worker, the analysis did not go

deep enough. That’s a good point. I think what a lot of uh industries are missing, it’s not

actually finding that root cause. It’s taking it one step further and actually realizing that they need to go deeper.

So Craig, do you have any advice for organizations that they can pick up on

signals where they’re stopping too early or they haven’t gone deep enough or they don’t even know that it’s a problem? Is

there things that they can look out for? Uh yeah. No, I think that’s a I think that’s a again a useful kind of segue

into a conversation that that that really allows us to talk about, you know, what is the true purpose again of

doing a doing a root cause analysis. I mean, often often some of the challenges that come with that are are knowing

knowing where to start, knowing where to end, at what point do we move on from from the investigation to saying, well,

we’re establishing what happened. When do we move on to finding out the root cause? When’s the right time to do that?

When do we actually how do we know our actions are going to be effective? An investigation contains so many unknowns.

Uh and and and you have to have you have to have a structured methodology that

signposts what you’re doing, when you’re doing it, and why you’re doing it, right? And there are and and of course I’m going to sit here and talk about

comet, but but there are many many different ways at which that that of course can be done. So I think some of the signs that people

have got to look out for are if you get to a point when you’re doing a root cause analysis, right? Because doing an

investigation versus a root cause analysis are different things. People might not think about that but they are and often we want to jump too quickly

into root cause analysis because we want to get the outcomes and Scott mentioned that all that leadership want the results on my desk and 24 hours and

that’s not always feasible right so we do we do jump ahead one of the things that we often see is when you’re doing a root cause analysis

and you’re trying to get to that why if at that point in time you are still using assumptions you are still using a

little bit of opinion information at that point uh or you have a process that allows you to kind of ask certain

questions to get the right outcome of the root cause analysis and you don’t necessarily know the answers to those factually, you’ve probably not done the

investigation right. Okay. So, that kind of input is really really important. So, I think that’s probably one of the first things is if you if you just don’t know

and you’re still making assumptions at that stage and you’ve gone there a little bit too quickly, then you’re really going to go go to go back a step.

So, when it comes to root cause analysis, it should actually be a case of validating what you’ve learned in the investigation and and and establishing

these are the root cause outcomes that we’re now going to fix. So, you shouldn’t really be learning anything new when it comes to that stage. It

should be validating what you’ve learned in the investigation in a structured way that allows you to then say, “Okay, we

we’ve worked through all of that. That’s the input. That’s how we’ve we’ve broken it down. These are the these are the

four root cause outcomes that we’re now going to go and address with two or three really good actions.” That’s a

really simple linear method that allows you to do it, but the checks and balances keep you right. So the longer you go in that process, if you still

have questions at the end about why it happened and how it happened and what happened, then you’ve probably not actually done that initial investigation

properly in the first place. Whilst there will always be gaps and challenges, um, we have to have an

honest view of that. So that’s one of the things we often see is, well, Craig, how do we know when to move on or when we’re finished?

Well, you’ll know by by the status of your outcomes and how easy it was to get to that point. That that’s our advice.

Yeah. Yeah. I I would just add to that is that, you know, if you’re if you’re finished with your investigation and

your corrective action looks like retrain, rewrite, rem, uh you didn’t get you didn’t get to the

end point of where you needed to to get to. And I’ve often said this as I’ve counled practitioners, if you see repeat

incidents, that’s always a signal that the prior investigation stopped too

short. you know, there was a system condition that caused that first in incident that you never addressed. So,

it’s always going to to be a big need to go to that next stage of of

understanding of really why did we see that and why are we seeing it again,

right? So investigation quality I I think it it’s it’s the single biggest

lever most organizations don’t pull very well and it costs nothing but it does

require some discipline and I think Craig said it well is that the purpose of framework and methodology is

discipline so we can do these so we can repeat them over and over and over again and when I led a very big organization I

taught everyone methodology of going from an immediate cause to a root calls.

Why? Because it turned them into risk managers. They started understanding not only what could happen, but why it could

happen. And it’s a really really important lever to pull because it affects the culture of the organization.

So doing this well is really a culture driver if you if you really spend the time and the discipline to do it.

A topic that we chatted about on the first episode was utilizing AI to help

kind of um provide some of that information. Craig, do you think that AI

should be implemented in these to provide that data and make sense of it or do you think it’s more of a hindrance?

Um I I think it can be a hindrance if the user doesn’t know how they’re how they want to use the AI. Right? I think

there’s a lot of um talk now about okay, we have AI as an option but we shouldn’t be using it. Sounds good. Well, why?

What are you using it for? And there’s a big big difference when it comes to how you apply AI. There’s a big difference

between utilizing an open AI mechanism and applying it to an investigation versus using AI that’s been designed for

investigations. That’s a significant difference. We talk about AI and we utilize AI that’s that’s that’s in the

workflow. Right? So, if we think about that from a perspective and I know I know obviously, you know, you guys are really really across the kind of input

versus insights AI. And if you think about that from an investigation standpoint, you want to have an investigator that’s

assisted by AI that prompts, guides, suggests things, that highlights gaps

they might have missed, that the human brain might might have missed, but the investigator or that team still ought to

validate the outcome. So we we cannot become overreiant on AI. That’s a that’s a huge risk, right? And actually we

would talk about still practicing investigations without AI because we we we still um remain and retain that skill

set. So we don’t want to we don’t want to suffer from skill fade. If we become over reliant on that whilst it’s very it

can be very very effective and and I think in this day and age it’s probably becoming an organizational expectation.

So we should embrace the change. Um but it has to be utilized in the right way for the right reasons and and nothing at

all should ever prevent the investigation team from making the final decision particularly on root causes and

outcomes. It can guide what happened. It can guide your understanding of immediate causes and human factor

problems and some of the barriers or controls that might not have quite been been that effective. That’s all useful

context, right? But when it comes to validating root causes and and getting those right actions at the end of it, I

think the human still has to be responsible for that. So, we should be embracing AI from that from a perspective of let’s let’s get as much

help as we can. Let’s save 50% of our time. Let’s increase our quality output by 80%. Brilliant. But let’s not remove

the human from that workflow ever. Um, and of course, from a data perspective, well, brilliant. What a great tool to to

ask a really simple question in terms of what’s my you know what’s my top trending root cause in the last three

months across all facilities and what are the what are the common theme human factors and barriers that are allowing

that to happen. What a great way to interact with the tool and you get that information instantly. So we have to

embrace the the the the power that that gives us but the human must retain all decision- making about what we do with

that data and how we use that AI and investigation. So, it’s a big yes for me, but with caution, right? With

caution. Yeah, I totally agree with that. I I use AI every single day, but mostly from

from the side of helping me do my job better, right? Or to check me on some of

the things that I think, right? Uh are there two things that I I may have forgotten, right, in in this path of of,

you know, thought that I want to have I think it’s good for that. I think it’s good for helping me be better. Is it is

it a replacement for me? No. And I don’t think it is. Uh I I think it’s exactly

how Craig has has has talked about it. It’s to help us be better, be more

efficient, to think about things that maybe we didn’t think about. But especially for this type of work where

majority of our investigation time is with a person, right, that’s had some

type of output that that we need to understand. And and I’ve told practitioners because it’s one of the

big things that I talk about at conferences. It’s one of the questions that I get. Is AI going to replace me?

Uh, you know, one of the things that AI is not good about is understanding that that person is having a relationship

problem. That person is going through some production pressure to try to perform at a level above his or her

capability. So, those are things that AI is not good about. And that’s where it kind of comes down to the gut of a

practitioner to see those things, to read those things intellectually and then being able to create that story or

to absorb that story much differently than what technology can do. And it’s the same thing, Monica, that we’ve done

with intellects, right? We’re we’re we’re not positioning AI as the answer to every problem that you’ll have. What

we’re trying to do with AI is to position you to make you more efficient, to make you a better practitioner, to

get you to those decisions that we really think that you need to make as a practitioner leading a program.

No, I agree. And I think a lot of people have this fear that AI will replace their jobs or they’re going to become

redundant because AI can do it better. And I think you both made such a great point that it’s a tool to help you do

what you do better, not what you do period. Um, on the podcast, I love to

have real world examples of stories. So Craig, starting with you, do you have an

example or a story about an investigation that led to a change and

how that was different? How was it approached? Anything you want to share? Yeah. No, absolutely. without without

without divulging details of course of of of who and when and all the rest of it but um yeah I mean I suppose I’ve

been depends on how you look at it fortunate or unfortunate to have been involved in many many industry based

investigations over the last decade I would say and um often from our perspective when we when we come in as a

as an independent third party to assist in in the process assist in getting the right outcome um it it you know every

organization does that a little bit differently and the ones that are certainly the ones that are open to to

feedback and and really seeing as an opportunity to learn tend to get tend to get more out of it. We had one

particular example it was actually now gosh 2018 so maybe seven eight years ago

now we were working with a an oil and gas operator uh in in in North Sea waters and uh

they’d suffered a particularly um nasty incident unfortunately uh one of the rig crew did uh during a lifting operation

and they um they suffered permanent disfigure injury through their on their left hand and um unfortunately the

reason that we were actually drafted in to help with that is because there were some there were late um administration

changes made within some lifting plans and um that that red flag to the onshore

team is what led us to get involved because they they felt there was there was some foul play involved. Fortunately, the the the investigation

that we assisted with discovered that that wasn’t the case and that initial assumption on the human error which was

they thought were trying to deceive the outcome of the investigation was actually just a genuine error at the time and had no bearing on the

investigation itself. And so they went into that with a real objective I think to not just deal with the issue in terms

of the incident, but they they were they were they were hunting somebody and their perspective in that investigation

which lasted about two to three weeks their perspective changed and that person kept their job rightfully so and

and they learned that actually the system that people were utilizing uh was was showing delays in certain um inputs

of data and that’s where that discrepancy came from. So they got two learnings. They they got learnings from the incident itself that the actual lift

and the operation. Okay, that was that was pretty straightforward. The real issue with the human interaction and the amending of these plans retrospectively,

which actually turns out wasn’t the case. They got more learnings from that and I think from a management team, they probably realized, well, we were very

close to making an assumption there that would have probably let somebody lose their job and that would have been um

entirely unfair in the circumstances. So that’s one that always jumps to mind because there was there was multitude of

learnings but there was a real awakening I think with the management team to say oh we’ve got to be a bit more careful about the assumptions we make and I

think anyone who goes into an investigation should always be careful of the assumptions that you make at the beginning because rarely do those

assumptions come true Scott any stories from your Oh boy quite a quite a few I I’ll I’ll

tell one um you know earlier in my career I did I worked a fatality

investigation. I’d led that work uh where a worker was killed while performing unplanned maintenance on a

piece of equipment that wasn’t isolated and uh the easy finding in all of that

was failure to lock out. Uh the real finding was a work system

that had all of this uncontrolled variability. Uh we had long complicated

procedures in that facility. We had inadequate guarding on this piece of equipment. We had insufficient

maintenance planning. Uh and really we just had this culture that had

normalized all these shortcuts. Why? Because this person was good, because

this person was fast, because this person had a lot of knowledge of the equipment. So we started normalizing all

of these different shortcuts in that facility because quote, we trusted that

that person could do that job successfully. And probably that person had done this many times before except

this time it caught him. Uh so once we named the system uh the change was was

quite structural you know equipment redesign uh we simplified procedures we improved training to make sure that we

were validating um you know their their skill level maintenance planning was overhauled supervisor accountability

right there was just a lot of things and I think there was nothing about that

event that was one person’s choice right it was an organizational choice

comprised of a lot of different factors and I think that that if there’s one thing that we get out of today’s podcast

it’s that is that there are tremendous amount of factors that contribute to

output that at times you know cause or create an incident that that was a

fatality in this case I would add this too Monica and I won’t talk about it with but I’ve seen near miss

investigations drive bigger change than serious injury investigations because

organiz ganizations just have the emotional space to look at it honestly. And I think that if I could tell anyone

before we end today, it’s that is that if we can start looking at things more

honestly, the best learning comes from those events that maybe didn’t hurt anyone.

It’s like you’ve done this podcast with me before. I always love [laughter] to end the podcast with a piece of advice

uh for the listeners. So, I’m going to pose it back to you, Craig. Is there one thing that organizations should do to

improve investigations starting tomorrow? And if so, what would it be?

One just one thing, Monica, is that the is that it? Just one bit of advice. If you had to two or three things.

Yeah, [laughter] a couple of things. Okay. Okay. Um I think I think um probably in organizations that take it

seriously that that are that are maturing in this space um you know we’ll probably sit here and think yeah look we

know root cause isn’t a human error you know we know that that we’ve got to look closer at ourselves but in reality

actually are we doing that so I think I think I think an action takeaway has to be let’s let’s you know let’s be serious

about our next investigation um let’s make a point of getting ourselves uncomfortable okay so taking it back to

that that point earlier I think there also has to be an a realization that can be very very difficult particularly for

investigation teams that are that are doing this as their not as their day job right because investigators or

organizations rarely employ full-time investigation resources that only do that it’s people that get that bolted

onto their day job and that creates additional pressures. Yeah. So, I think relieving that pressure a little bit and saying, “Well, actually, you know what?

We’re going to do our best here, but we might not get it right or perfect every time, but we’re going to be honest. We’re going to be transparent.

Uh, and we’re going to we’re going to dedicate that through a really, you know, really healthy reporting culture. Um, outcomes that lead to people losing

their jobs are rarely outcomes that are going to the organization are going to learn and benefit from. So, I think I

think there has to be maturity in that space as well. And again, you know, we’re we’re in the year of 2026. I’d like to think that would be it. So my my

two bits of advice would be let’s let’s make a point on the next swing of doing things a little bit differently. Uh and

let’s make ourselves uncomfortable. The second one would be let’s accept that we’re not always going to fix everything

in one go. That’s not possible. Deal with what’s in front of you. Uh deal with even the lowhanging fruit and

accept we might not get it right. But you know what? That’s okay. Because the important point with this is it’s where you begin and where you start, not where

you end, right? That’s just a journey. And um if we start honestly and we look at ourselves honestly, then you’re going

to get you’re just going to open up the door to to to to better insights and and feedback. So the more you involve people

in that, you have that kind of honest reporting culture. Uh it’s a it’s a win-win situation, but it won’t happen

overnight either. Well said. Well said. Scott, any final words?

Well, I I would say this is Yeah, I’ve always got a final word, right? and it [laughter] usually spurred on by by

something that that Craig mentioned, but I I would say the strongest safety and

health programs that I’ve been a part of. Uh they treat data, they they treat reporting, they treat investigations,

they treat auditing and and inspections all as one system. You know, we don’t look at all of these things as separate

workflows owned by three separate people or three separate parts of of of the organization. And and many times then

you know when I work with clients even at intellects you know they may buy just the incident management application from

intellects and that and they’ll never they’ll no never go further than that and I often wonder because you could use

the same framework the same methodology that that comet has partnered with us to to demonstrate on auditing on

investigate on on inspections right why am I seeing these same problems over and over again when I audit an area Those

are really clear signs that we need to understand data much differently. So I

think reporting feeds data. Data reveals patterns. Patterns sharpen

investigations and investigations produce learning. And then learning changes the work. Right? So that’s

really how I look at at at all of this is that if we want actual change, uh we

want to to tighten that loop of what goes on in the work process. And and so

you know, I would say this and I think Craig has said it well is that there’s a lot of organizations that break this

loop somewhere, right? They collect data, they don’t analyze it or understand it. They investigate but they

don’t share learning or go deep enough with that learning. they share name don’t change the work right so I I think

if I had one challenge to every other practitioner that that has shared my chair look at your loop very very

honestly and see where it’s broken and that’s where your next improvement is hiding and I think Craig is exactly

right you don’t have to figure this all out tomorrow right but you can start tomorrow right you you can make

incremental change in your process and grow from there well said I mic drop I guess and I think

that is the biggest takeaway. The moral of this episode is that it’s not just about collecting data or completing

investigations. It’s actually about learning in a way that really changes the system. Um, Craig, Scott, thank you

both. This has been such an insightful conversation. I’m so happy we were able to continue it from the first episode.

For everybody listening, if you are interested in learning more about investigation methodologies and the work

that Craig, Mark, and team are doing, you can visit commonanalysis.com and explore everything that they’re doing

further. Uh, we also did do a master class with them, uh, which was also

equally insightful. But I think this episode really hit the nail on the head for so many important conversations that

need to happen. And it starts with just taking a deeper look, being uncomfortable, uh, taking that data and

actually looking into it with a little bit bigger lens than what you do now. Um, Craig Scott, thank you again. I I

don’t even know how to end this episode. It’s just such a great conversation. You guys covered so many great things and

um, if we have more conversations coming up, I’d love to have you back. Um, and

Greg, anything you want to conclude the episode with? No, if you’ll if you’ll uh if you’ll

have me back, I’ll be I’ll be delighted to be here. But yeah, I think I think I think Scott and I could probably um sit all day and talk about this stuff. So,

you’ll be glad to know we’re not right now. No, thank you for having me. Much appreciated. Amazing. All right. Well, thanks

everybody for listening. We will catch you on the next one.