Wednesday, May 27, 2015

Anatomy of an Error – What You See Depends On What You Choose to See

It’s one thing to talk about “human error”, it’s quite another to commit the error and to feel the effects of that error. Recently we had a situation where we had to deal with the fallout from a mistake one of our project managers made when scheduling services for a client. The project was an assessment where we had multiple assessors onsite conducting focus groups for employees. Each assessor was given a sheet that listed their activities, including location, contact person and time, for each day. Unfortunately one of the locations was put in incorrectly, resulting in two assessors going to the same focus group, and one set of client employees without a facilitator for their focus group. The supervisor for the employees became upset and cancelled the rest of the focus groups for that day for his employees. The focus groups would have to be rescheduled for another day and we essentially lost a whole day, where we had one assessor on site with nothing to do.

Now, obviously this event is not directly related to safety, as no one was injured, nor were they likely to be injured. However, the story is pretty similar to one you hear a lot following an accident or incident – someone screwed up and we have to pay the price for it. If only the project manager had paid more attention while setting up the schedule the failure would not have occurred. Therefore we can say that the cause of the accident was “human error”.

How do we deal with it? Well, the answer is obvious. If the project manager had only paid more attention we wouldn’t be in this mess, so we need to tell the project manager to pay more attention next time. Perhaps even some discipline is necessary to let them know the consequences of their mistake. We could dock their pay for the lost revenue from having an assessor do nothing all day. That would show them. If only we could find a way to get a picture of them in the act that we can use in training to shame them and others into doing the right thing, that might help them not make that same mistake.

Does any of the above sound familiar? It should, because it’s the same basic way that most organizations respond when a failure occurs and it can be traced back to some sort of mistake. The thing is that the above seems like it makes sense. However, what you see depends on what you choose to see. Or, as Erik Hollnagel says, what you look for is what you find, particularly after an accident. Many are content with seeing a failure event like the one above as another story of people not doing what they are supposed to do. So we need to fix the people (the project manager in this case) and the problem will go away. But if you choose to look at the event differently a different lesson and a different set of interventions present themselves.

To illustrate, let’s look at the same event, but let’s look at the actions of the project manager in the context they were in. First off, the client was the one scheduling the individual focus groups and they did not finish the schedule until late Friday afternoon, the week before the assessment. So the project manager had to construct the schedule for his assessors over the weekend. This was no easy task, as the project manager had to take the schedule from the client and ensure that each focus group had an assessor assigned to facilitate and some of these focus groups overlapped and were miles away from each other (it was a large site). So the project manager needed to ensure that each assessor had time to get to their focus groups, and then put this information into a format that was easy to understand.

Normally we have quality controls in place to check people’s work, but special circumstances made this challenging. First, it was a weekend, so most people were not available. Second, those that were normally available over the weekend, were not this particular weekend, as a result of their family dog passing away that Friday. Oh, and by the way, the project manager was constructing the schedule after helping bury the dog that afternoon. So he was doing complex work, while tired, potentially distracted, and without crosschecks to ensure accuracy.

Finally, the assessors who arrived at the same focus group could have realized that it is not normal to have two assessors in the same focus group. So that could have alerted them to the problem, which they could have then figured out that one them was supposed to be somewhere else. But the two assessors happened to be those with the least experience with the client and with the focus groups we do. In fact, one of the assessors was not even supposed to be there, as we had informed the client that we had a limited number of assessors available that morning, so we needed to limit the number of focus groups scheduled, which may have been lost in translation. But the customer is always right, so we pulled in an assessor who was unfamiliar with the project at the last minute to accommodate.

Here’s the thing – it is true that if only the project manager had not made the mistake then the failure would not have happened. But the fact that we focus on that piece of the story and ignore everything else should say something to us. Here’s some more “if only” statements that are equally true:

  • If only the client had not scheduled too many focus groups that morning, the failure would not have happened.
  • If only one of the two assessors happened to be one of the more experienced assessors who were onsite they might have noticed the issue earlier and started asking questions that would have led to a better solution.
  • If only the family dog didn’t pass away, the quality control programs may have been in place to catch the problem during the weekend.
  • If only the client had delivered the finished schedule earlier, the project manager may not have had to construct the schedule over the weekend, when he was tired and likely to make a mistake.
  • If only the supervisor had not gotten angry at having missed the first focus group, we could have accommodated a make-up for the one focus group and easily conducted the rest of the focus groups for that day as scheduled.


There are more “if only” statements we could identify. This isn’t just a long, convoluted excuse (we apologized profusely and took full responsibility to the client, making up the focus groups and making everyone happy). Rather, what we are trying to show is two-fold. First, as Drew Rae says, when you tell the story of an event, especially a failure event, like an accident, you are making choices. The choices you make are how you tell the story, what facts you share (and what you don’t share) as relevant, and who you highlight was important (and not important) to the story. All of this affects what you get out of the story. If all you do is tell the story of “human error” that involves a deficient person doing something stupid, then it makes sense that you would be led to the conclusion that you need to blame and fix that person. But if you choose to tell the whole story and context, if you choose to see the world through the eyes of the people involved (as much as it possible), a different picture emerges that leads to different conclusions.

And our second point, is that is this has real world consequences. Because if the second version of our story is true, the idea that we can tell the project manager to be more careful next time and that this will fix our problem is just silly. The project manager didn’t make a choice to make the mistake, so tell him to choose to not make the mistake next time is like telling your pet dog to try and do better at math next time. The project manager felt terrible for his mistake already, so what is us making him feel even worse going to do?

The point of choosing to tell the whole story of the event is not to remove blame just for the sake of letting someone off the hook. Instead, it’s about really solving the problem. By understanding the story you can build in more defenses that actually improve your system. For example, if part of the issue is related to inexperienced assessors, telling the project manager to not make a mistake next time doesn’t make that situation better. It also doesn’t fix communication issues with the client in terms of scheduling needs.

So next time you are faced with a situation in your organization where someone made a mistake, remember that you have a choice. You can choose to focus on that one “if only” statement, or you can choose to see the larger picture and you just might actually make your system safer in the process.

Monday, May 18, 2015

Real Safety Leadership

Recently in conjunction with a management system assessment at a large international airport, we were holding focus groups with line employees, discussing their perceptions of how safety is managed. These focus groups are not only one of the most important pieces of the assessments, but they are also so much fun, because you get to interact with the people who really make safety happen, day in and day out. You get to see an expression of who the organization really is as the employees share stories and interpret those stories. It’s just fascinating.

At one particular focus group, the employees were discussing issues with upper management, where they felt management misunderstood the role of the employees and often blamed employees for problems that weren’t really the employees’ fault. The supervisor for the work crew spoke up and mentioned how often the issues are far worse than the employees may know. He often is called into the offices of upper managers and given a tongue lashing for the workers not getting the work done fast enough or not meeting expectations. The expectation is that the supervisor will then trickle down the discipline to his employees, where the problem (supposedly) really is.

But the supervisor in this case mentioned that his employees didn’t realize that this was happening as often as it was happening. Why not? Because he had deliberately chosen to bear the brunt of the production pressures and not share that with his employees. He didn’t want his employees to feel that production was more important than working safely. When asked why he did this, he simply stated that it was his job. Now, we’ve looked at the job descriptions in this organization. Nowhere does it mention in this guy’s job description that his job is to shield his employees from undue production pressures. So this supervisor, on his own, because of his concern for his crew, decided to bear this burden on himself.

Folks, that is safety leadership. In the safety profession there’s a lot of talk about how we need safety leaders and the idea of leadership is often glorified to be a person out in front of the organization, pointing them in the right direction against all odds. But we don’t think that’s really leadership. That seems more consistent with hero worship, than leadership. Instead, leadership is a social process used to influence others towards the completion of a common task. There’s nothing in that definition that requires someone to be in front of the organization. It’s about using the tools in your disposal to influence people to do the things that they already want to do. It’s about making it easy for people to achieve their goals (which you also happen to share).

That supervisor from the airport, by removing potential negative influences to his employees, is creating an environment that enables his employees to do what they already want to do – do a great job without getting hurt. How many of our safety programs are designed with this process in mind? How many safety systems focus on enabling safety rather than ensuring safety? Think about that. In one case (ensuring safety) we are dragging the organization kicking and screaming to do what they apparently don’t want to do (or else they wouldn’t be kicking and screaming about it). In the other case (enabling safety), we are making it easy for our people to do what they already want to do. It’s about identifying and removing barriers that make it hard for people to execute their tasks safely.

We think this is a seemingly small, but potentially revolutionary shift in how safety management systems are structured, and how safety leadership is conceptualized. How can you get started?

Step 1 is to go out and talk to your employees. Ask them how work really gets done and what makes getting the job done challenging or difficult. Ask them what surprises them when they do jobs. Make a list.

Step 2 is to do an assessment of your safety management system. Identify the gaps, the imperfections, the places where goals are competing, where work is complex, etc. Once identified, list out what your organization is doing to fill those gaps, or mitigate the risk from them. Chances are many of them are filled by employees, not because that’s the best option, but just because no one else is dealing with them. In our experience, these issues often seem less problematic on paper than they are in the work environment. List all these out too.

Step 3 is to devise plans to remove the barriers your employees identified and fill the gaps identified in the assessment with system fixes. Involve your employees in identifying those fixes.


Step 4 is simple – follow through. By removing the issues your employees face you will influence their behavior toward the common task of getting the job done safely. You will be enabling safety. Like the airport supervisor, you will be a safety leader.

Thursday, May 7, 2015

Accidents As Investments

To a recent post of ours, an important safety researcher in the field, Dr. Richard Cook left an interesting and important comment speaking of a specific case study that was referenced in our post: 
It is, perhaps, telling that the argument that I used with this company that finally prevailed after the accident was this: "Your company has just made an unplanned and irrevocable investment in safety. It is up to you to decide whether you will get any return on this investment."
 I meant this quite literally: they had spent millions of dollars in the lost production and people time spent managing the accident but had done almost nothing to extract any value from the event. And this investment was much more than they spent on "safety" programs. Changing the direction was a matter of pointing out to them that the accident was a huge investment (albeit unplanned) in safety and it was in their business interest to treat it as such. Accidents are so disruptive and damaging that it is hard to think of them as investments. But the entire accident, including its direct and indirect costs (e.g. increase in insurance premiums) can and should be considered an investment. Reversing the language from “cost” to “investment” is key to getting the attention of management.
What an interesting and important shift in perspective. The traditional narrative in safety management following an accident is one of costs. We talk about what accidents cost the organization in the hopes that those costs drive action toward safety improvement. We debate the ratio of direct to indirect costs of accidents, with indirect costs taking up the bulk of the costs, at least that’s how the theory goes. What we’re really looking for is learning. We’re hoping that our organizations see the accident (whether real or potential) as something to avoid and change their behavior as a result.

But what if part of the problem is how we’re framing the problem? If the narrative we bring to management is all about costs, we create an environment where the only option is an escape from loss. This is inherently demotivating much of the time. If, as Dr. Cook recommends, we change the discussion from loss to potential gain (i.e., investment) there may be more motivational capital there. The thing is that after an accident happens, the organization has already invested the resources (time, money, etc., all those things that we call direct and indirect costs). The choice is now the organization’s whether they want to protect their investment by learning and improving, or if they will waste their investment.

As Dr. Cook points out, this shift in language may seem somewhat odd, or perhaps even perverse to some. Accidents are bad things, after all. This is true and no one is arguing that we want more accidents. However, accidents do continue to happen. The organization can choose to see the accident as merely a cost, which, like other organizational costs, can be simply budgeted for year after year (we’ve seen many organizations that do just that! Talk about a habituating practice!). Or, the organization can choose to see accidents as opportunities to get better at what they do. To put it another way, using another concept in safety management, although accidents are traditionally viewed as a lagging metric of safety performance, learning is so crucial to enabling safety, the way organizations respond to accidents are a leading metric (and yes, it’s something you can measure).


Now, this might seem like a subtle, perhaps meaningless change. After all, all we’re saying is that you should just change the words you use when talking about accidents. What difference can that make? A profound difference actually. As the Appreciative Inquiry community is fond of saying – words create worlds. But don’t take our word for it, just try it. Just start referring to accidents as investments and see how people in your organization respond. The results might surprise you.