We all know culture when we see it, but defining it is another matter
It was my first day on a new project, and I was excited to meet the team. As I walked into the office, I couldn’t miss the floor-to-ceiling banners: workers in hard hats and hi-vis, 20 metres high. Otherwise, it was a standard corporate space: bright, airy, bustling with activity.
At reception I collected my pass, and the project lead whisked me through the turnstile towards the stairs.
“Make sure you hold the handrail,” she said casually. “Oh, and if you grab a coffee, always take the lift.”
I must have looked surprised. She smiled and told me about a colleague who once dropped a laptop from the third floor. No one was hurt, but it was a close call.
“Safety is really important to us,” she said, and I could tell she meant it. This wasn’t corporate spin.
It reminded me of my twenties, working in wineries. People imagine wine as glamorous; in reality, most is made in sprawling, factory-like complexes. Every year, people die in accidents - vats produce carbon dioxide, which sinks, and if you fall in, you can lose consciousness instantly. In Australia, New Zealand, and the US, safety was taken obsessively seriously (France was a little more laissez-faire). There were signs, safety briefings, and stories about past incidents. In short: a strong safety culture.
Why defining culture is so tricky
We all know culture when we see it. We’ve all worked somewhere with a “great culture” or endured a “toxic culture”. But defining it is trickier.
Researchers have tried, and there are plenty of models:
- Cameron and Quinn’s Competing Values. This looks at four pre-determined drives in the brain: the need to bond, learn, acquire, and defend. A good ref for this is Nitin Nohria and Paul Lawrence's book
- Denison Culture Survey. The work focuses on mission, adaptability, involvement, and consistency (broken down into 12 components to describe the organisation's beliefs and assumptions).
- Schein’s Triangle. The triangle defines culture across three aspects: artifacts (visible symbols like posters and dress codes), espoused values (statements and behaviours), and underlying assumptions (what people REALLY think).
- Deal and Kennedy’s Model. This model is based on six interlocking elements: history, values, rituals, stories, heroes (individuals admired in the workplace), and the cultural network.
- Johnson and Scholes’ Culture Web. Sharing aspects of Schein and Deal and Kennedy's models, this mixes stories, routines, symbols, structures, controls, and power dynamics.
These are interesting lenses, but they share the same problem: they’re descriptive, often subjective, and hard to test. Ask the HR Director, a frontline worker, and the CEO the same survey question, and you’ll likely get three different answers.
Like horoscopes or reiki, you can’t disprove them. But that doesn’t make them useless, a lot of psychology is hard to test, and therapy still saves lives — but it does mean we should look for models that are replicable and objective. Two consultants assessing the same organisation should reach the same conclusion.
The problem with culture models
Culture is often invisible from the inside. Stories only make sense in context: ask people for a 'company story' and they might shrug, but forward them a mislabelled spreadsheet and suddenly you’ll hear the tale of when Finance once sent the wrong numbers to the board (true story).
Power structures are the same. You don’t see them day-to-day, they become apparent only when budgets are allocated or big decisions are made.
So what should a good culture model do? At the very least, it needs to be:
- Replicable: same input, same result
- Objective: not reliant on one individual’s opinion
- Representative: reflects the whole organisation, not just the exec team
- Diagnostic: identifies what’s working and what isn’t, in context
- Actionable: provides guidance on what to do next, with limited resources
Without these qualities, models may be interesting but not that useful.
So what’s the answer?
Culture is made by real people in real contexts. There'll never be that perfect model that explains everything. But a useful model must go beyond description. Information is only valuable if it helps you act.
That’s why we’ve designed our own Epion methodology: one that measures over 100 concrete factors to show you where you are, what it means, and where to focus your change efforts.
For more information, contact ruth.tank@epion.co.uk or follow Epion on LinkedIn for more articles like this.
