When your policies and practice don’t match, or when what you say doesn’t align with what you do.
I wrote the other day about Psychological Safety, following the FMA media pieces.
Here’s the thing.
If I ask you, as an organisation, to show your commitment to a healthy, psychologically safe culture, I might well get a list. There’s a wellbeing programme. There are mental health first aiders. An anti-bullying day. You very likely have a code of conduct, a values statement, and even some training modules.
It’s all genuine and real. Some of it may be worthwhile (research says otherwise on some things but that’s another story), but it doesn’t really answer the question. These are intents, effort even, but Psychological Safety is measured in behaviour, and specifically what happens when someone takes an interpersonal risk.
Apples and oranges
You can have a perfect policy suite and a culture where nobody dares admit a mistake. Or you can have a perfect policy suite that isn’t followed. Documents and lived experience might be only loosely coupled, and it’s the experience that will dictate whether people speak up. They will work from observed behaviour, evidence, history, advice from others, anecdata, and seeing how the last moment was handled, and much less from the document that describes how things should be handled.
Detert and Burris’ study makes the point from the other direction. What predicted whether employees actually voiced concerns wasn’t the presence of formal mechanisms, it was the leader’s openness in practice, operating through peoples’ perception of what speaking up would cost.
After a Government agency was in the news in 2025 for their culture, low trust in complaints processes, little faith in HR, and other complaints around bullying, meaningless engagement surveys, policy gaps, policies not being followed, and more, I wrote to the CE. I got a reply from HR, you know, the ones people had no faith in, telling me that they were comfortable their internal capability could progress the work. Obviously, they hadn’t been able to progress the work but clearly had faith in themselves that their people didn’t. Ironic.
The mismatch is obvious
An open door policy just says the door is open. It doesn’t, nor can it, make any accurate claims about what happens once you’re inside the door. That’s evidence that only behaviour can provide, and it’s the behaviour that people will respond to. It’s here that you see why culture programmes and psychological safety may diverge so completely. A programme can measure what the organisation has in place; what systems have been installed. Safety measures what it does when tested. The two things are different. The test, typically, is the behaviour of one person, or a few people.
This is not to say that policies are pointless. They aren’t, and I’m not cynical about them. I’m definitely cynical about some things, but good policies aren’t on that list. They’re important for setting expectations, providing recourse, signalling intent, and having them is a thousand times better than not having them. You may have a speak-up channel, but it’s weaker without a policy.
Side note: The NZ Public Service Commission has what they call Te Whakapuaki Kōrero, which is Model Standards: Speaking Up, along with Ngā Wāhi Mahi Pai, Haumaru anō hoki, Model standards: Positive and safe workplaces. Have a look.
Where it goes awry
Where it might go awry is in treating them as the same measure. Having a policy doesn’t produce the thing you need. Installing a machine doesn’t automatically mean it will produce the thing you want. You need people able to operate it effectively. A wellbeing programme might be an input, but it isn’t evidence of an outcome. What I see is these things being presented as if it were. There is a bunch of reasons this is wrong, but I’ll start with just one.
The FMA, or any organisation under scrutiny (the Big Four accounting firms come to mind, especially in Australia, see the Broderick review of EY and the Switkowski review of PWC) routinely points to its programmes as proof that things are okay. Policies are readily to hand, there are photos of events on social media, and each thing can be verified. And every item on the list can be in place, while the actual culture, which is what people experience, aligns with what critics describe.
As we’ve said, they co-exist because the measure different things. Pointing at the first tells you nothing of the second. The tricky part is not that they’re different, but that pointing at the first can mislead people into believing that second exists. Further, if I’m the one doing the pointing, I may be misleading myself, or taking my own visible effort as evidence of something I haven’t actually checked.
To be clear, I’m not saying that every organisation that has good policies also have bad cultures. Plenty have both, and the policies help. I am saying you can’t infer culture from policies. Moreover, behaviour is genuinely harder to measure. Policy audits and inventories are much easier, which is why we point to them in the first place. However, measuring the easy but wrong thing is no substitute for the harder but right thing.
Now what?
So the practical shift is from auditing what we’ve installed, to observing what we do. Not, ‘Do we have a Speak Up policy?’ but is it being used and, if it is, what happened the last three times someone used it? Not, ‘Do we run wellbeing programmes?’ but what do people actually experience if they raise concerns, or voice disagreements, or challenge some senior leadership sacred cows?
Those questions are harder, and they may not make a tidy list for your annual report. But they are way, way more valuable.
Photo by Gowtham AGM on Unsplash
References
Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350-383. https://doi.org/10.2307/2666999
Detert, J. R., & Burris, E. R. (2007). Leadership behavior and employee voice: Is the door really open? Academy of Management Journal, 50(4), 869-884. https://doi.org/10.5465/amj.2007.26279183
Edmondson, A. C., & Bransby, D. P. (2023). Psychological safety comes of age: Observed themes in an established literature. Annual Review of Organizational Psychology and Organizational Behavior, 10, 55-78. https://doi.org/10.1146/annurev-orgpsych-120920-055217