Workplace culture: what it is, and how to actually change it

Key takeaways

Culture isn't your values poster. It's how it actually feels to work somewhere, what gets rewarded, and what gets quietly tolerated.
Engagement is a measurement. Culture is the thing being measured, plus everything that happens after you see the result.
Measuring culture without acting on it doesn't just fail to help. It teaches people that feedback goes nowhere, which makes the next survey worse.
Changing culture for real takes three moves done on repeat: listen continuously, act visibly, close the loop.
You don't need a culture overhaul. You need one visible change, made quickly, that people can point to and say "that came from what we told them."

A new client once told me, with a straight face, that her culture was "collaborative, innovative and people-first." I asked her team the same question a week later. Three words came back: "political," "slow," and "exhausting."

Both descriptions were true. One was what leadership believed. The other was what people actually lived. That gap is the whole problem, and it's where I want to start.

What culture actually is

Culture isn't the poster in reception. It isn't the three words on your careers page. It's the answer to a much blunter question: what's it actually like to work here, day to day, when no one's performing for the survey?

Two things build that answer.

What gets rewarded. Who gets promoted, and for what? Does the person who hits their number but burns out their team get ahead faster than the person who quietly makes everyone around them better? People watch this closely, far more closely than they watch your values statement.

What gets tolerated. Every workplace has a line between "we don't love this, but we live with it" and "this genuinely isn't okay." Culture is where that line actually sits, not where the handbook says it sits. A manager who talks over everyone in meetings and never gets pulled up on it is a culture signal, whether anyone names it that way or not.

Put those together and you get culture: the lived pattern of what's rewarded and what's tolerated, which people learn fast and repeat because it's how you get on here.

That's why culture is stubborn. It's not a slogan you can rewrite. It's a set of incentives and tolerances that took years to build and won't shift because you changed the wording on a slide.

Engagement versus culture: they're not the same thing

I hear these two words used as if they're interchangeable. They're not, and mixing them up leads to the wrong fix.

Engagement is a measurement. It's the score you get when you ask people how connected, motivated and supported they feel right now. It's a snapshot, useful, real, but a snapshot.

Culture is the thing being measured, plus everything that happens next. It's the conditions that produced that engagement score, and it's also what you do once you've seen it. A low engagement score with no follow-up doesn't touch culture at all. It just confirms what people already knew and adds "and nobody did anything about it" to the pile.

This is why an organisation can run engagement surveys every year for a decade and still have a culture problem. They've been measuring the temperature without ever opening a window.

	Engagement	Culture
What it is	A score, at a point in time	The lived pattern behind that score
What it captures	How people feel right now	What gets rewarded and tolerated over time
Can you fix it with a survey alone	No	No
What moves it	Nothing, on its own	Visible action, repeated
Time horizon	A snapshot	Built and rebuilt continuously

If you only ever look at the left column, you'll keep being surprised by the right one.

Why measurement without action makes things worse

Here's the part most leaders don't want to hear. Running a survey and then doing nothing with the results isn't a neutral, low-cost move. It actively damages trust.

Think about what you're asking people to do when you send out a survey. You're asking them to be honest about something that might be uncomfortable to raise directly, on the promise that it's safe to do so and that it'll matter. If nothing visibly changes afterwards, you've just taught them two things: their honesty didn't matter, and the next survey isn't worth their time either.

That's why response rates decay year over year in organisations that survey without acting. It's not apathy. It's a rational response to evidence. People are pattern-matchers, and "we asked, they answered honestly, nothing happened" is a pattern they learn quickly.

This is the trap I call action theatre: running the measurement ritual, publishing a results deck, maybe forming a working group, and calling it culture work. It looks like effort. It produces almost nothing, and it burns the trust you'd need for the next round to work.

If you want the detail on why the traditional annual survey is especially prone to this, I've written about it in annual engagement survey vs continuous listening.

A grounded model for changing culture

I don't think culture change needs to be complicated. I think it needs to be consistent, and it needs three things happening on repeat, not as a one-off project.

1. Listen continuously

A once-a-year survey gives you one photograph of a moving picture. By the time you've analysed it, presented it and formed a committee, the picture has already changed, and the people who told you something important have often already decided what they're going to do about it themselves, which sometimes means leaving.

Continuous listening, short, frequent pulses woven into normal work rather than one big annual event, gives you a trend instead of a snapshot. You see a problem building in a team three months before it shows up in resignation numbers, not three months after.

2. Act visibly

This is the step that's usually missing, and it's the one that matters most. Acting on feedback only counts if people can see it happened. A change made quietly, with no connection back to what was raised, doesn't build trust. It might even go unnoticed.

Visible doesn't mean a company-wide announcement for every small fix. It means the people who raised the issue, or the team it affected, can draw a clear line from "we said this" to "this changed." Small and visible beats large and invisible every time.

3. Close the loop

Tell people what you heard, what you're doing about it, and what you're not doing and why. That last part matters more than people expect. You can't act on everything, and saying so honestly, rather than going quiet on the topics you're not tackling, is itself a trust-building move. Silence reads as "we didn't listen." An honest "not yet, and here's why" reads as respect.

Do these three things on repeat and culture starts to shift, not because you announced a big change program, but because the lived pattern of what happens after people speak up actually changes. That's the whole game.

Where to start, practically

You don't need a culture overhaul to begin. You need one thing.

Pick one issue that's come up clearly and repeatedly, something small enough to fix in a month, not a year. Fix it. Then tell the people who raised it, plainly, that you heard them and this is what changed.

That single visible loop does more for trust than a beautifully designed values framework ever will. It gives people proof, not promises, and proof is what gets them to be honest with you the next time you ask.

From there, the model scales. Keep listening continuously. Keep acting on what you hear, visibly. Keep closing the loop, even when the answer is "not yet." Culture built that way holds up, because it's built on evidence people can see for themselves, not a statement someone wrote once and framed.

How this connects

Culture isn't one lever. It's several things working together, and I've written each of them up on its own.

Annual engagement survey vs continuous listening, on why the once-a-year model keeps failing and what to run instead.
Psychological safety at work, on the specific condition that determines whether people tell you the truth in the first place.
Employee engagement benchmarks for Australia and New Zealand, on reading a benchmark without chasing a number that misleads you.
Pulse survey questions that actually tell you something, on asking questions specific enough to act on.

This is also the thinking behind Whyser Culture, the part of Whyser Work built to listen continuously, surface psychological safety and psychosocial risk, and keep the record of what you did about it.

Frequently asked questions

How long does it take to change a workplace culture?

Longer than a quarter, and probably longer than you'd like. But you'll see the first real signal, people speaking up more freely, faster than you'd expect, often within weeks of the first visible action you take on their feedback. The full shift is a compounding effect of doing the three things on repeat, not a date on a project plan.

Can a small business have a culture problem, or is that a big-company issue?

Culture exists in every workplace with more than one person in it, so yes, a business of ten people has a culture just as much as one of ten thousand. The good news is that small businesses can often move faster, because there are fewer layers between "someone raised something" and "someone with authority can act on it."

What if leadership doesn't agree there's a problem?

Show them the gap between what they believe and what the data says, in their own words if you can get it. The example at the start of this piece, "collaborative, innovative, people-first" versus "political, slow, exhausting," is the kind of contrast that tends to land harder than an argument does.

Do we need new software to do this, or can we start with what we have?

You can start today with what you have: a short set of questions, sent regularly, and a genuine commitment to act on what comes back and tell people so. Software helps you do it continuously and keep the evidence trail intact, which matters more as you grow, but the habit is what actually changes culture, not the tool.

If you had to describe your culture the way your team would, not the way your values statement does, what would you say?