Why employers need to see people holistically, beyond their roles at work

The world of work exerts a powerful influence on people’s overall lives. A job is not just a means to an income— in fact, globally, only around 40% of those employed agree or strongly agree that “A job is just a way of earning money — no more.” In the UK, around 30% of workers agree or strongly agree; and in Nordic countries, fewer than 20% do. Or, to put it another way, more than 50% of people in the UK disagree or strongly disagree with that statement (that a job is just a way of earning money).

In the best case scenario, people in good jobs benefit from all the things that are important for human wellbeing: a sense of agency and mastery, positive social relationships, and meaning and purpose. Arguably, though, these “higher level” needs should only receive priority after some more basic requirements have been met — such as physical safety, respectful treatment, and a decent income.

Nevertheless, I’m frustrated that a lot of the discussion on “good jobs” has tended to lay out minimum standards as opposed to aiming to define what we really mean by “good”. Of course, it is not my place to define that, either. But we can use data to illuminate “what really matters” to people. [Note: I’m aware that making “average” statements about this is somewhat questionable— but at least these “averages” are based on fairly robust data, not someone’s personal opinion of what matters.]

So, yet again I turn to my trusty friend, machine learning, to tease out from a large number of variables what best predicts a “good job”. I offer this as a complement, not an alternative, to the already existing careful econometric studies (e.g., here) as well as causal studies using various designs (e.g., here). I have used data from the UK’s Understanding Society survey (wave 10), with a total of 17,300 respondents for whom relevant data was recorded.

The chart shows the top 20 variables that are most predictive of someone’s response to the following question: “All things considered, how satisfied are you with your (main) job?” This is an intentionally subjective question: I think we should let people decide for themselves, not assume outside-in, how much satisfaction they get (or don’t get) from each aspect of their job.

The results are not quite what I expected. Previous studies have suggested that social relationships and, especially our relationships with our managers, are a critical driver of job satisfaction. I’m pretty sure that’s still true — after all, that finding was trying to isolate the causal links between variables and outcomes, and this analysis is not. That earlier analysis was also focused on looking at things within the work domain. But I wanted to look across a broader set of variables and, indeed, there are other things going on, too.

In particular, the results show how intertwined “job satisfaction” is with the rest of our lives.

But before we get to that, to me, the output of the model also suggests something else: how negative for people’s overall wellbeing uncertainty can be. The biggest predictor of someone’s discontentment with their job is their lack of job security (first column on the left). The fourth most important predictor is whether they receive regular pay increases or not (fourth bar from left). This is entirely consistent with the literature that links financial insecurity to a number of mental health issues — and, by the way, one of the arguments in favour of guaranteed basic income.

So what about the rest? Well, unsurprisingly, the ability to have autonomy over job tasks pops up as an important driver. The need for agency and freedom to make choices is a well-established requirement for people’s ability to thrive (in work, or indeed in their lives). But perhaps more intriguingly, a larger proportion of the drivers that show up in this analysis are about someone’s life in general than they are directly about someone’s actual job.

Anyone who is a manager or leader out there, trying to motivate and support their team members, will probably be going, “well, we didn’t need a fancy piece of analysis to show that, did we?” But I think we did. Why is that? Well the non-believers may not believe my analysis, but they certainly won’t believe someone simply asserting such a point, either. And yet, I really think employers everywhere need to get real about this intertwine — not deny it.

So if your employees are depressed (for whatever reason), or unhappy with their residence (for whatever reason), or think they are not doing enough for the environment (for whatever reason), or are feeling constantly under strain (for whatever reason), or indeed, are strugging with their finances (for whatever reason), they will also be more likely to be unsatisfied with their jobs. And I’m reasonably certain that they are also less innovative, productive, and so on (see e.g., here). Even more reasons to take a more holistic, human-centric, perspective to your employees.

There is, of course, a counter-argument to this: some aspects of these factors are surely outside employers’ control. And I have some sympathy with that. But just like global warming is essentially entirely outside of most companies’ control, that doesn’t mean they shouldn’t try to understand it and its consequences, not least for their own business, and act on it. Besides, I can see many ways in which employers could try to positively influence the factors here, in ways that would not even cost much (if any) money.

But that’s for another post. A starting point is here: encourage your managers to have real conversations with their people. You’ll be surprised by what rewards it brings.