Impotent behavioural lassos
The Safety Managers were throwing behavioural lassos.
-I think we need to send a stronger message about our safety standards, and what we don’t accept. Big poster, all around the place. No excuses.
Someone else added their solution:
-What about a commitment ceremony? When people walk up and sign on to a charter that they will follow procedures and policies, it’s so powerful.
I was getting uncomfortable. The Safety Director had asked me to join the meeting. But I wasn’t quite sure how I could help. The subject of the meeting was a series of breaches of the working at height policy, on a particular site, with a particular contractor being caught in the crosshairs. They’d been spotted walking on wings without proper fall arrest equipment, they’d been photographed hanging out the cockpit window doing maintenance tasks on the windshield, and there were stands on top of platforms. And then, there was a particularly damning photo of high-vis scotch tape across a cabin door, with no stairs positioned outside, presumably put there by the same contractor. Too many too serious infractions. Something needed to be done.
-How effective have those interventions been in the past? I asked, trying to not sound too critical.
After a short but telling silence, they turned the spotlight on me:
-Well, what do you suggest we do about it then?
-I don’t know what the solution is. But I know where I’d like to start. I’d like to go to the site and talk to the people involved. I’d like to understand what they are trying to achieve, what they’re trying to avoid, and what's getting in the way, what’s holding them back. To me, it’s a mystery why people who are committed to aircraft safety would do these things. I’d like to understand it better.
After some discussion it was agreed that I could go to site and talk with the people involved. But only under one condition: that I’d always bring out the working at height issue.
Not in a single conversation did I have to bring up the working-at-height situation. Everyone wanted to talk about it. The breaches soon started to make sense. Part of the problem was that there wasn’t enough equipment to go around. Another part of it was that the line maintenance employees and the base maintenance contractors didn’t get along or weren't a team anyway. They’d ‘borrow’ equipment from each other to get their work done. The stand surrounding the aircraft was not an aircraft-specific stand but a generic building scaffold, leaving gaps around the fuselage and a constant need to adapt to get the right height to stand on for the different parts of the aircraft. Compliance was a struggle.
After collating 150 quotes and examples of how work was playing out on site, I invited a cross section of the organisation to make sense of the situation. Some of the lasso-throwing managers were there, and so was the GM Maintenance for the airline. Site-based personnel paired up with corporate-based ones and read one story together at the time; they spoke quietly about each example. After 45 minutes, we gathered around and talked about what was shaping performance on site, and we explored options. The ultimate outcome was to create a site improvement team. While this may sound like a deferral to a committee, it was what that team did that transformed the site.
In weekly meetings between a group of 8 frontline staff and the most senior manager on site, they generated small improvement ideas to try. Most ideas could be done locally. When something bigger was needed, corporate was supportive. Building a stand to stack panels sideways rather than in a pile was one. Giving contractors the same uniform as the employees was another. Getting the aircraft type-specific stand was an obvious change. In the end, many ideas collectively generated a radical improvement in conditions and in performance. The problem child became the poster child. And no posters of what the company accepted, and not as far as the eye could see.
Static and adaptive controls
What looked like a failure of behavioural control turned out to be something else. The organisation had a working-at-height standard. And there was nothing wrong with the standard itself. What the organisation lacked was a way to continually see, understand and adapt the conditions around that standard, when reality didn't cooperate.
In the name of improving safety, most organisations reach for controls, or more controls when the initial controls are deemed insufficient. Policies, procedures, standards, barriers and training encode what we believe should happen. They are our best thinking about how work should be done. And organisations need to signal what matters and establish things that should not be left to chance.
But these controls are relatively static. Designed upstream and ahead of the actual work, they anticipate what could go wrong and outline what should and shouldn't happen. Once designed, however, they have little capacity of their own to respond to changing conditions. They don't notice. They don't inquire. They don't have conversations. They don't adapt. As such, they can also create another problem: the more we specify how work should be done, the less room we leave for the people doing the work to exercise judgement, initiative and ownership.
And that becomes a problem when the challenge is adaptive. The work changes. Demands compete. Equipment isn't available. Different groups need to coordinate. Something unexpected happens. The conditions are different from those imagined when the control was designed.
I’m not arguing for doing away with static controls. But we need another kind of control as well. Something which is more alive to what’s going on. Controls that help an organisation notice what is changing, make sense of it, and adjust what happens next, while placing some of that adaptive capacity with the people closest to the work. I'm using “control” here in a broad sense: something that helps shape what happens, not simply something that constrains behaviour. From that perspective, metrics and information are controls too. What we choose to measure, report and make visible shapes what people attend to and what gets acted upon. But information doesn't become adaptive simply because it is current. It becomes adaptive when an organisation can interpret what it is seeing, decide what it means, and change something in response.
A tentative list of adaptive controls might include (click here for a slide format):
Being in the work. Staying close enough to changing conditions to know what is actually happening.
Job design. Changing the conditions of work as we learn what makes good performance easier or harder.
Coaching and connecting. Helping people navigate competing demands and coordinate across boundaries and different parts of the system.
Supported autonomy. Putting the capacity to adapt close to where the variability occurs, with the authority and resources to act.
The conversations that happened on site were part of boosting such controls. The site improvement team was one of these controls. It created a mechanism for the organisation to discover where the standard was colliding with reality, and to do something about it.
Such adaptive control requires leaders to relinquish some direct control in order to create more organisational control. The safety managers initially wanted more control over behaviour. The site improvement approach created more control over the conditions producing the behaviour, while simultaneously giving frontline people more agency. Which is richer than simply “empowerment.” Ownership becomes one of the ways an organisation maintains control when it can't prescribe everything in advance.
When we’re pushing to make controls more visible, another option worth exploring is how to make the organisation more present, maintaining a living connection with the work: seeing what is happening, listening to what people are experiencing, making sense of what it means, and being able to change something as a result.
And it changes the safety leadership challenge. Static controls need leaders who can define, communicate and reinforce what we have decided should happen. Adaptive controls need leaders who can notice, inquire, connect, enable and adjust to what is happening now.
Most organisations need both static and adaptive controls. The leadership challenge is knowing when the answer is to send a stronger message about the static control, and when the answer is to strengthen the organisation's capacity to adapt around it.
/Daniel
PS I’ll be in Dallas on 15 October, running a training workshop on Safety Conversations for Leaders. More information here.