PAPod 622 - When Safety Programs Go Bad: Enshittification and the Iron Law

So probably nobody on earth tried to set up a safety program that was bad. You know what I mean? I mean, let's just give the benefit of the doubt. Probably nobody is that big of a sociopath. So everybody intends for their safety program to be effective. The challenge is, is that over time and tested by metal, sometimes they aren't, as effective as they could be.

Hey, everybody. Todd Conklin, Pre-Accident Investigation Podcast. How are you? It is fall where I am. In fact, it's Halloween where I am. Boo. I tried to sneak that in. And it's cold and wet, which is kind of not normal for New Mexico. It's cold sometimes. Don't get me wrong, but cold and wet, that's crazy.

So now we've gone from the arid Southwest to the quite moist Southwest, but I was traveling like crazy this week. So I went and saw my friends in Brazil. That was a great experience. And then I gave a speech in Bergen, Norway, and that was really fun. So yeah, lots of trying not to travel, which sounds like such a lie.

Every time I say that people look at me like I'm crazy, but I really am. I'm trying not to travel because airports are not that fun. I have such an airport story. I try not to tell airport stories because Because... You know, it gets tedious. It's like rock bands who do songs about being on the road.

After a while, you're just like, okay, got it. Oh, man, your life is terrible. A million dollars a day and no green M&Ms or whatever the story is. But yeah, no, life is grand. It's just fall. I mean, and so now it smells good. The leaves are changing because, you know, I live in the mountains. And it means winter's on its way. So you'll hear me complain about winter now for months. That's probably what's going to happen next. I mean, I'm kind of predicting the future.

Today we should have a chat, though, because today I want to talk to you a little bit about this idea of what happens to safety programs. And really, when you think about it, it's kind of an interesting idea. And it all kind of started when I read this concept a couple years ago called inshittification, which sounds dirty, but I checked it's on the safe for work list.

Now, there's a root to that word that probably will get you in trouble somewhere, but the whole word itself, and that's the idea that as we create systems, we tend to move them towards less useful, not more useful. So we tend to take things away from those systems as opposed to adding things to those systems. And systems get crappier and crappier and crappier, harder to use, harder to interface with.

And there's some reasons why that exists. Now, I'm not sure this has any direct correlation to. To a safety program, but it might. And I've thought about it a lot because over time, some things are kind of guaranteed to happen to a safety program. Number one, just right out of the chute, is all the incredibly interesting metrics, that we've designed to measure so we can determine our status.

Because as you know, what is, if it's not measured, it's not monitored. What's measured gets done, you know, all the measurement stuff. But one of the interesting phenomenons is, is we consciously try to create metrics that are really effective and forward looking, you know, maybe, maybe not predictive, but maybe predictive.

Maybe they're leading metrics. If such a thing exists, I've not found them, but maybe you have. But ultimately we consciously try to make the best metrics we can. And then what happens is as metrics become metrics, the process of inshittification clearly happens because what takes place is everything that was once seen as a leading data set a metric that was measuring progress towards getting better.

Eventually over time becomes a lagging data set, and becomes a metric that must be measured because we're accountable to that measurement and so kind of like when you put in barriers or mitigators or controls in a system. Every time you put a control in a system, you can guarantee that that control, will in fact be used eventually as a production advantage.

The example I think of all the time is if you bought a saw stop, right, that's the machine that if it senses human skin, it puts a dog in the blade and stops it. That technology's been around a while. It's really interesting and it's really changed. It's much more advanced. It's very interesting. Eventually, the users of that saw stop will become so normalized to the fact that that saw stop will stop and it's really, really difficult to cut your hand, that their hands will get closer and closer to the blade.

Or the use of a pusher stick will become less important because the saw, in fact, saves their fingers. That's a really good example of a control of the barrier that you've designed in there, the process barrier, becoming eventually... Production advantage. Or they'll use the high limit alarm to tell them when the tank is filled when they're filling it, right? So these are interesting problems.

And that very same thing happens with metrics, is that metrics over time go from leading to lagging. And every metric will eventually become some kind of performance hurdle in some kind of key performance indicator watch list or dashboard or chart or submitted slide or whatever you have to do. And that over time is kind of normal.

And so that's a part of what happens and it kind of takes place. And that moves us into that direction. The other thing about safety systems is they're really interesting in that safety systems actually sort of over time become encumbered with their own bureaucracy. And the word for this, if you want to look it up, I think I did a podcast a million years ago on this.

It's called the iron law of bureaucracy. And the iron law of bureaucracy says any bureaucratic system will always, underline that word, always reinforce itself. And so the notion that we somehow won't create bureaucracy in a safety program, that's foolish because we will. There will be some kind of bureaucratic underpinning. There will be some kind of reportability, some kind of accountability. That's all pretty normal, right?

The problem is, is that the bureaucratic systems that we create often serve to reinforce the bureaucracy of the system that's created. And you see it all the time. If you've ever tried to get something, no matter where you live in the world, from the government, right? There's a bureaucratic process that moves over and over again.

And our friends Drew and Sydney in Australia really tried to tackle this idea by talking about safety decluttering. And what they were really talking about is how can we remove, the parts of the bureaucratic process that drift into the system over time by the system reinforcing itself in order to make that system less bureaucratic and more streamlined.

You can tell I'm trying to struggle because I want to say simplified. I don't think simplified is the right word. Streamline is probably the word because a lot of these systems, if they're complicated, they'll simplify pretty well. But a lot of our systems over time because of the interfaces and how they have touch points throughout the organization, a lot of those systems become.

Pretty complex. And the example I give you is I did an investigation. I helped out with an investigation where some guys were fixing a gas leak. And the gas leak started out small. And as they dug it out and uncovered it, it got larger, which is pretty normal. That happens pretty often. And they were in a position where they probably should have been using breathing apparatus.

And they had it on the truck. It was available to them. And it was functional. It was working great. But the company over time had created a set of rules and regulations that you could not use breathing apparatus without senior management approval, whatever that meant. And that's what it said, senior manager approval.

And it was a Friday night as the time when all crap happens. Nobody was available. Nobody was callable. So they weren't calling people. They couldn't call people. They wanted to call people, but there was nobody to call. Nobody would answer the phone. And so they ended up jumping into the pit one at a time holding their breath.

Now, that's a bad thing, obviously. I mean, but what's interesting is you could look at that and say, that's an exercise in poor judgment by the workforce. We could talk about this for a long time. Ultimately, however, it was really kind of interesting that they had the equipment. They wanted to use the equipment. They were fine using the equipment.

But the fear of being punished for breaking the rule overwhelmed the understanding, of the risk that they were encountering. And that's a pretty interesting story. And that's a story, you probably have something kind of like that in your organization. You may not know it. I mean, no one may have told you that story.

And if it ended up fine, nobody finds out. You may never know. But what's interesting is that's a function of bureaucracy that over time had drifted to a point, where it was really pretty burdensome to get a primary safety capacity off the truck and use it as designed for an emergent system, an emergent problem.

That is a really good example of how our systems get cluttered. It's also a really good example of the iron law of bureaucracy. Because my guess is, and I was having lots of conversations about this, is nobody ever intended the approval from senior leadership, whatever that means, to be a burden to actually providing safe, reliable worker support.

It was really a way to control access to what at the time, I think, was a piece of equipment that required special training, special certification, special maintenance. I mean, there's a lot of details to it that kind of made that story a big part of what happens. So that idea is a pretty important idea that over time, bureaucracy will reinforce itself and become bigger, fatter, heavier, more burdensome, crazier.

Pick any word. That was an option. You can have two of those words. You can pick any two of those words. That really becomes interesting. So that takes us to the third one, and I think the third one is the one I think about the most, and that is eventually all these programs become weaponized. So if you've got a Start Safe program, or you've got a Start With Sure program, or you're using an Energized Work program, a HECA chart, those are all really great tools to help the workforce start safely.

But over time, and often based upon events, the tools will become weaponized and people will be punished for not using the tool when they should have used the tool as determined retrospectively by the organization. So if you've got a tool and says you need to determine the energy and then put in the correct controls before work starts and something happens, you do all that and something happens, you know as well as I do, this is not even made up, that it's a relatively good bet that somebody in the organization will look back and say, these guys failed to identify this hazard, therefore they didn't use the tool correctly or incomplete, and therefore they're bad workers, and we need to reinforce

and retrain everybody on how to use the tool perfectly. My advice here is that pretty much everything that we put out there has the potential to become weaponized, but that's not the problem of the workforce. That's the problem of how leadership thinks about those tools. And when things start to get weaponized in the field, like accountability is a great example of that. It gets weaponized all the time. We use it like a weapon.

But in fact, accountability is really much more of an act of clarity. And what's amazing to me is that if we weaponize these systems. Then what happens is, is we've kind of made the workplace harder to be safe. Because remember, safety is not something you have accomplished that you will, you know, eventually get to.

Safety is something you do. That's why we talk about safety being a capacity is it's something that happens in real time. And the amazing thing about these weaponized systems is what it really points to is the fact that the organization is really out there auditing for obedience. And when you audit for obedience, then what happens is you lose the ability, to have improvisation or you hide the ability to have improvisation, adaptability.

You lose the ability for workers to problem solve in real time. And that's kind of significant. That's what I think about when I think about, well, what's next or what fall is going to look like, right? Is these ideas that over time, whether we want to or not, our systems become inshittified. Inshittification happens. It happens to everything.

And part of our job is to understand that and to do our best as an organization to constantly monitor, and adapt our systems so that we can combat against those three potential pitfalls. And they're real. They're very real. They happen all the time. You know they're real because you can give me examples from your organization right now.

That is a very interesting way to think about how the world's. So that's pretty much the conversation we're going to have. I mean, I think we could have more conversation, but that's really kind of what I wanted to talk about. I've been thinking about this a bunch, a bunch, a bunch, which I don't know what that means.

It might mean that I'm going to think about it enough to write some stuff down. I haven't even thought about that because every book that you write is the last book you're ever going to write because why would you do that again? It's not that fun. But this one I think is interesting because I think it's really important to nod to the fact that when we started our programs, when these programs start everywhere, when you started your program, the intention was to help the world, not hurt the world.

And over time what happens is that things move towards places, where honestly what was really pretty good leading information has now become kind of a lagging obligation that we have to get in by Tuesday and at 10 o'clock before the management meeting, and that challenge is a big part of what we want to think about and where we go with these ideas.

So that's the pod learn something new every single day. I bet you did today. Today was a fun day. It was a little shorty, but I don't mind doing short ones. That's not a problem for me. Learn something new every single day. Have as much fun as you possibly can. Be good to each other. Be kind to each other. Take care of each other, even the least among us.

I've really been thinking about that a lot. There's so many stories I have to tell you. We're just going to have to make some time to do so. And until then, for goodness sakes.

PAPod 622 - When Safety Programs Go Bad: Enshittification and the Iron Law

Kapitelklicken springt direkt hin

Shownotes

Transkriptvom Podcast bereitgestellt · Zeitmarken springen hin