How to Measure and Improve Company Culture

Measuring company culture means putting a clear read on something that feels intangible, by combining what people say with what they actually do. The most reliable approach triangulates three sources: direct feedback such as surveys and conversations, behavioural signals such as how problems are raised and how people treat each other, and operational data such as retention, quality and safety. No single measure captures culture, but together they turn a vague sense of "how things feel around here" into information you can act on. And measurement only earns its keep if it leads to change, so the point is always to improve, not just to score.

This article explains why culture is worth measuring, the methods that actually work, the traps to avoid, and how to turn what you learn into genuine improvement.

Culture is a set of norms you can actually read

I find that people over-mystify culture. Stripped back, it is simply a set of norms, the behaviours that have become normal and go unremarked in your business. That is oddly reassuring, because norms show up in what people do, and what people do can be observed. My training is in organisational psychology, and the discipline it gives you is not a magic instrument for reading minds. It is the patience to look at behaviour honestly and ask what it is telling you, rather than deciding in advance what you would like to be true.

So when I say you can measure culture, I do not mean you can reduce it to a single tidy number. I mean you can watch the norms in action and read them fairly, provided you are willing to hear an uncomfortable answer.

Why measure culture at all

Culture is often dismissed as too soft to measure, yet it drives the outcomes every business cares about: quality, safety, retention, customer experience and the willingness of people to improve things. If it matters that much, leaving it unmeasured means steering blind on one of your most important assets.

Measuring culture does three useful things. It surfaces problems you would otherwise only discover through a crisis or a wave of resignations. It gives you a baseline against which to judge whether changes are working. And the act of measuring, done sincerely, signals to people that their experience matters, which itself supports a healthier culture.

The methods that actually work

Because culture has no single natural metric, the best approach layers several methods so their blind spots cancel out.

• Pulse surveys. Short, frequent, mostly anonymous check-ins on how people feel about specific things, such as whether they feel able to speak up or whether their manager listens. Frequency beats length; a handful of questions every few weeks tells you more than an annual epic.
• Structured conversations. Regular one-to-ones, skip-level chats and honest exit interviews surface the texture that surveys miss. Why people leave is especially revealing.
• Behavioural signals. Watch what people actually do: how many problems and near-misses get raised, how many improvement ideas are suggested, whether difficult news reaches leaders early or late. These behaviours are culture in action.
• Operational and people data. Retention and turnover, absenteeism, internal promotion rates, safety incidents and quality trends all carry cultural signal, especially when you look at how they move over time.

The aim is triangulation. If the survey says people feel safe to speak up but almost no problems are ever raised and turnover is climbing, the behaviour is telling you the truth and the survey is being polite.

The signals worth watching

Some signals repay attention more than others, and a few are pleasingly counter-intuitive. The one I always ask about is whether people feel safe enough to report problems. It is tempting to treat a clean, quiet log of issues as good news, but a suspiciously quiet workplace often means people have simply stopped raising things. When you make an environment safer and more honest, reported problems frequently rise for a while, and that rise is a healthy sign, not a failing one. It means the norms are shifting towards openness.

A few other signals I would watch closely:

1. Who speaks in meetings. If challenge only ever flows downward and never up, the norm is deference, not honesty.
2. How bad news travels. Whether problems reach leaders early or only once they are unavoidable tells you a great deal about the real culture.
3. What happens after someone raises something. Ideas and concerns that visibly lead somewhere signal a living culture; ones that vanish signal a dead letterbox.
4. How people talk about the place when they leave. Exit conversations, taken seriously, are among the most honest data you will get.

What good measurement looks like

Effective culture measurement follows a few principles that separate it from box-ticking.

1. Measure things you will act on. Only ask what you are genuinely willing to respond to. Asking and then ignoring the answers is worse than not asking.
2. Favour frequency over grandeur. A regular pulse you act on beats a once-a-year survey that produces a report no one reads.
3. Protect honesty. Anonymity and a track record of no reprisals are what make the data trustworthy. If people fear consequences, they tell you what they think you want to hear.
4. Look at trends, not snapshots. A single score means little; the direction of travel over time is what matters.
5. Combine the numbers with the stories. Quantitative scores tell you where to look; conversations tell you why.

As a simple illustration, a business might track a handful of pulse questions, its voluntary turnover rate and the number of improvement suggestions each quarter. None is definitive alone, but watching all three move together gives a fair picture of whether the culture is strengthening or fraying.

The traps to avoid

Culture measurement can do harm when done carelessly, so a few traps are worth naming.

The biggest is measuring and then doing nothing. When people complete a survey and see no change, they conclude the exercise is theatre and disengage, leaving you worse off than before. Never launch a measurement you are not prepared to act on.

A second trap is turning culture into a target to be gamed. If a manager is judged purely on their team's engagement score, some will manage the score rather than the culture, leaning on people to answer positively. Treat the numbers as a diagnostic, not a stick.

A third is over-relying on a single metric or a slick dashboard. Culture is too rich to reduce to one number, and a tidy score can create false confidence. Keep the qualitative picture alongside the quantitative one.

Turning measurement into improvement

Measurement is only the first half; the value is in what you do next. The pattern that works is a loop: measure, share honestly, act visibly, and measure again to see whether it helped.

Sharing the results openly, including the uncomfortable ones, builds the trust that keeps future measurement honest. Then pick a small number of things to change, ideally the issues people raised most strongly, and act on them where the whole organisation can see. Visible action on real concerns does more for culture than any number of statements of intent. Finally, measure again to close the loop, both to check whether the change worked and to prove to people that speaking up leads somewhere.

Done consistently, this loop turns the raw data you gather into genuine information about your culture, and then into steady improvement.

Frequently asked questions

What is the best single metric for company culture?

There is no reliable single metric, which is precisely why triangulation matters. Measures such as voluntary turnover, engagement pulse scores and the volume of problems and ideas raised each tell you something, but each also has blind spots. The truest picture comes from watching several indicators together over time and combining them with honest conversation.

How often should you measure company culture?

Short pulse checks work well every few weeks or once a month because they let you spot changes and act quickly, while deeper reviews might happen quarterly. This beats the traditional annual survey, which arrives too infrequently to catch problems early and often produces a report that changes nothing. The right cadence is whatever lets you act on what you learn.

Can you improve culture without measuring it?

You can improve culture through good leadership without formal measurement, and many small businesses do. Measurement helps by surfacing blind spots, giving you a baseline and signalling that people's experience matters. Its main value is discipline: it stops you assuming the culture is fine simply because no crisis has forced your attention.

Where to go next

Measuring company culture well turns a vague sense of how things feel into information you can act on, and acting on it is what drives real improvement. ProgressXL, our operational-excellence platform, is designed to help you turn data about your business into a clear plan for improvement and is coming soon. Learn more at ProgressXL.