Your culture isn’t on the wall. It’s in four places.

Most organisations have a values statement.

On the website or on the walls. Or both. (I remember seeing on a wall in M&S “To be carbon neutral by 2016” sometime in 2017.)

Researchers at MIT Sloan studied around 700 companies and found more than 8-10 published core values on their website.

Then they analysed over employee reviews to see whether it made any difference.

And guess what, It didn’t.

They found negligible correlation between the values a company emphasised and how highly its own employees rated it on those exact values.

A company that talked loudly about integrity was no more likely to be experienced as having integrity than one that never mentioned the word.

All that guff on the website? Who is it really for? New recruits, customers? Current employees?

If the research is to be believed then it’s mostly for the optics and for the truth seekers  (new recruits, new customers) the question isn’t “what are the values?” but rather “where can I find those ‘values’ at work?”

Why words alone can’t do the work

Edgar Schein, who spent a career at MIT studying this, described culture as operating on three levels. There are:
– artefacts, the things you can see, including the values on the wall.
– espoused values, what the organisation says it believes.
– underlying assumptions, the beliefs so taken for granted that nobody articulates them, because they no longer feel like beliefs. They feel like reality.

Assumptions form because something at some time worked.

A group faced a problem, tried a response, the response succeeded well enough, and it got passed to newcomers as the correct way to handle that situation.

Schein calls culture solidified learning.

Espoused values (the words you want people to believe in) are written by a small group in a room over a couple of days.

Underlying assumptions are built by everyone, over years, through direct experience of what happens. When the two conflict, experience wins, every time, and not because anyone is being cynical. People are simply believing the evidence in front of them.

The experts believe that culture doesn’t appear on the wall, it comes in 4 places.

Money. Permission. Meetings. How bad news travels.

Money
In 1975, Steven Kerr published a paper in the Academy of Management Journal with a title that has outlived most management writing: “On the Folly of Rewarding A, While Hoping for B.” He catalogued systems across medicine, politics, universities and business that formally asked for one behaviour while paying for a different one. In every case, people did the paid thing.

Fifty years on, nobody has overturned it, and organisations keep rediscovering it the expensive way.

A company boasts for example, that ‘collaboration’ is a core value and yet compensates entirely on individual quota.

It might then say that quality matters but pays a bonus on volume shipped.

Nobody in those companies is confused. The compensation plan is simply the clearer statement, because it’s the one with money attached, and people quite reasonably follow the money

You might ask “what does the bonus actually pay for, and is it the named behaviour?”

Permission
The second place to look is who can say yes without asking.

‘Decision rights’ sit at the heart of organisation design – new starts work out the real answer in about a fortnight.

During interviews and onboarding, they are told that the organisation trusts its people, values initiative, pushes decisions down. Then they ask for access to the shed in the back of the yard and discover it needs three signatures and a fortnight’s wait.

At that point the induction becomes a distant memory.

Approval thresholds are a statement about trust, made in a form that can’t be spun. If the stated culture and the sign-off matrix disagree, the sign-off matrix is what people will believe.

Meetings
Watch who speaks first, who’s in the room, and whether the ‘respectful challenge’ actually means anything.

David Marquet cites the example of the El Faro containership in “Leadership is Language” (blogged about before here) sailing into Hurricane Joaquin and sinking with all thirty-three crew.

The transcript from the bridge was recovered and reveals officers raising the weather in hedged, tentative language, and a captain whose replies steadily close the conversation down. Suppressing the discussion, as Marquet observes, did nothing at all to suppress the doubt — or the storm.

Take that to an office meeting where the most senior person opens the discussion with their view – it won’t be all that easy for anyone to challenge the leadership perspective after that.

Instead Marquet offers a “Vote first, then discuss” protocol, having everyone commit to a position, by show of hand or anonymous poll, before any conversation begins.

Drop the yes/no questions in favour of ‘on a scale of…1-to-5…’ questions.

An “Is this safe?” question becomes “how safe is this, on a scale of 1-to-5?” because a binary question invites agreement while a number forces a position.

He also suggests having a junior person chair the decision meeting, so the senior person stays in the role of evaluating a decision rather than defending one.

In your meetings, who talks first?

How bad news travels
Observe how information moves, and specifically what happens to that person who brings bad news.

In pathological cultures, messengers are shot. In bureaucratic ones, they’re tolerated. In generative ones, they’re trained.

A study of nursing units in hospitals showed that better-performing units reported more errors
It wasn’t about how often mistakes happened – it was whether you could survive admitting one.

Where bad news travels slowly, people are not being dishonest, rather they have worked what tends to happen to whoever raises it, and they’ve simply adjusted.

That’s not a character problem but a design problem and design problems can be fixed.

(I remember someone talking once about a company that introduced a “Joker card” which employees could play when they made a terrible mistake at work. Introduced as an incentive to encourage people not to hide errors or at least feel comfortable / psychologically safe in confessing. Management allegedly got suspicious when staff didn’t play their jokers each month.)

Something you can measure this week

Here’s something practical to take away (I hope anyway!) that will tell you more about your culture than a survey will.

Take a recent problem…something with a customer, a missed deadline, a process failure.

Find out when the first person knew and then find out when someone with the authority to act knew.

The gap between those two dates is a reasonably honest read on how safe it feels to speak up in your organisation.

If the gap is uncomfortable, don’t tell people to escalate faster because they most likely already know they should.

Ask instead what happened to the last person who did.

Thanks for reading

Sources

MIT Sloan Management Review / Glassdoor, Culture 500 research programme; “How Leaders Champion Culture: Six Essential Lessons” and “10 Things Your Corporate Culture Needs to Get Right,” MIT Sloan Management Review.

Schein, E. H., Organisational Culture and Leadership (5th ed., with Peter Schein).

Steven Kerr, 1975