Culture Problem or Perception Problem? How to Tell the Difference

You can tell whether a culture problem is real by testing what people say against what people do. If the same behaviours are described in specific terms, reported by independent groups, visible in hard data and rewarded or tolerated by the organisation, the problem is real. If the concern rests on impressions, a single event or a few loud voices, it is more likely a perception problem.

Most leaders arrive at this question with a signal but no verdict. Survey scores have slipped. A respected colleague has resigned and said something on the way out. Leadership believes things are fine, while employees describe something quite different. The instinct is to pick a side, and both sides carry risk. Calling it "just perception" can dismiss a genuine problem. Calling it "toxic" can trigger an expensive response to what is really a misunderstanding.

This article sets out a practical way to answer the culture problem or perception problem question before anyone commits to culture transformation work. It is written for leaders, HR teams and advisers who need a decision rule rather than another list of warning signs.

Culture Problem or Perception Problem: What Is the Difference?

A culture problem is a pattern in how people behave and in what the organisation rewards or tolerates. A perception problem is a gap between what is happening and what people believe is happening. The two overlap more than most people expect, because perception is often the first place a culture problem shows up.

Culture, at its simplest, is the everyday way work gets done. Gallup uses a similarly plain definition in its guide to diagnosing a broken culture. That makes culture visible in behaviour: how decisions are made, how bad news travels and what happens to someone who disagrees with the boss.

Perception sits one step removed. It is what people conclude from the behaviour they see, the stories they hear and the events they live through. Those conclusions can be accurate, partly accurate or wrong. Sometimes they are wrong because a change was poorly explained. Sometimes a single vivid event has outweighed months of ordinary experience.

Here is the part that matters. Perception is not the opposite of reality. A belief that speaking up is punished will make people stay quiet, and that silence then becomes a real feature of how the organisation works.

Research points the same way. An MIT Sloan Management Review analysis combined attrition data with themes in employee reviews across large US employers between April and September 2021. It found that toxic culture was the strongest predictor of industry-adjusted attrition, around ten times more powerful than compensation. That describes statistical patterns across companies, not any single workplace. Still, it shows that what employees say about culture can track what they later do.

So the useful question is not whether the problem is real or a matter of perception. It is what the perception is anchored to.

Why Leaders and Employees See the Culture Differently

Leaders and employees often describe the same culture in different terms because they observe it from different distances and through different filters. Senior people see the organisation through reports, meetings and prepared conversations. Employees see it through their own manager, their own team and small moments that never reach a dashboard.

The distance can be wide. Gallup reports that 43 percent of leaders strongly agree they feel connected to their organisation's culture, compared with 20 percent of individual contributors. A survey by Arbinger, described in a Forbes Human Resources Council post by its managing partner, found that 43 percent of executives believed their organisation excelled at fostering collaboration, while 18 percent of employees agreed. On innovation the figures were 37 percent and 21 percent. That survey comes from a consultancy and appeared as an opinion piece, so it works best as an illustration rather than a benchmark.

These gaps do not prove that leaders are wrong. Sometimes leaders hold information that employees lack. More often, a gap suggests that information is being filtered on its way upward, which is a finding in itself. Structured 360-degree feedback can help show where a leader's intent and their impact diverge. A gap is not a verdict. It is a reason to test.

How to Tell If a Culture Problem Is Real: Five Tests

The most reliable way to settle a culture problem or perception problem debate is to test what people say against five things: examples, consistency, data, correction and consequences. A concern that passes most of these tests is real, whatever anyone prefers to call it. One that fails most of them is more likely a perception problem. These are practical checks rather than a validated instrument, so I would use them together and treat none of them as decisive alone.

Test 1: Can People Describe Specific Behaviours?

A real culture problem can be described in examples. A perception problem is often described in adjectives. "The culture is toxic" is a label. "Decisions agreed in the meeting get reversed by email that evening" is a behaviour.

When you gather feedback, ask for the last time it happened, who was there and what followed. If people answer easily and the examples resemble each other, you have a pattern. If the answers stay general or trace back to something someone heard, you have a lead to investigate, not yet a finding. This is not about dismissing feelings. It is about knowing what you are being asked to change.

Test 2: Do Independent Groups Report the Same Thing?

If people who do not talk to each other describe the same pattern, it is probably real. If the concern appears in one team or under one manager, it is probably local. Both answers are useful, but they lead to different actions.

Compare across levels, functions, locations and length of service. New joiners can be especially useful, because they have not yet adapted to local norms. A pattern that appears only among people who share a manager, a recent restructure or a group chat is a different problem from one that appears everywhere.

The local case matters more than it might seem. Gallup estimates that managers account for at least 70 percent of the variance in engagement scores across business units. That figure concerns engagement rather than culture, but it points to something practical. What employees call "the culture" is often the experience of one team. A local problem is still real. It is simply not a company-wide one.

Test 3: Does the Hard Data Show a Footprint?

A real culture problem usually leaves a trace in numbers that are not opinion. If the data shows nothing, the concern may rest on perception. If it shows something, it is unlikely to be only perception.

Look at regretted attrition, internal transfer requests, exit interview themes, absence, grievance and escalation volumes, time to decision, rework and referrals from current staff. There is no universal benchmark for any of these, so compare each against your own history, across teams and, where you can, against similar organisations.

Two cautions apply. A footprint shows that something is wrong, not what. Turnover can come from pay, a difficult manager or a competitor's hiring drive. And a quiet dashboard is weak reassurance. Culture problems can suppress the signals that would reveal them, as people stop giving feedback, stop raising risks and quietly leave. If the indicators look calm but people have stopped talking, be careful about reading calm as health.

Test 4: Does the Concern Change When the Facts Are Clarified?

A perception problem tends to respond to better information. A culture problem tends to persist after the facts are clear. This is the closest thing to an experiment that most organisations can run.

Suppose employees believe promotions are decided by favouritism. Publish the criteria, show who was promoted and why, and give people a way to ask questions. If sentiment shifts and the questions become more specific, the concern was largely about information. If people accept the explanation and still describe the same pattern from their own experience, the behaviour is probably real.

Give this test time. Trust that has been damaged for a long while rarely recovers after one announcement, so repeat the check before you draw a conclusion.

Test 5: What Does the Organisation Reward and Tolerate?

What an organisation rewards and tolerates is its culture, whatever its values statement says. Look at who gets promoted, who gets forgiven and who is quietly moved aside.

If a high performer who treats colleagues badly keeps advancing, the behaviour is being paid for, and people have read the situation correctly. Their perception is accurate. If the opposite is true, and breaches of the values are challenged consistently and visibly, a negative story that lingers may be a legacy of an earlier period. This test is uncomfortable because it moves the question from what employees think to what leaders do. It is also the one that most directly shows what people have learned to expect.

Real Culture Problem or Perception Problem: A Quick Comparison

What you look at

• Real culture problem: Specific, similar behaviours described by many people
• Perception problem: General labels and second-hand stories

Who reports it

• Real culture problem: Independent groups across levels and locations
• Perception problem: One team, one manager or one social circle

Hard data

• Real culture problem: Attrition, absence or escalations that move with the story
• Perception problem: Indicators that look ordinary against your own history

Clarification

• Real culture problem: Concern persists after the facts are explained
• Perception problem: Concern softens once people have accurate information

Rewards

• Real culture problem: Poor behaviour is promoted or excused
• Perception problem: Behaviour is consistently challenged and the story is dated

When the Answer Is Both

A culture problem or perception problem is rarely a clean choice. When the tests give mixed results, treat it as a real problem with a perception layer on top, and deal with the behaviour first.

That order matters. A communication campaign that runs ahead of behaviour change can backfire, because people conclude the organisation is managing their opinions rather than the issue. Behaviour change with no communication has a different weakness: nobody knows to update their view.

Consider a hypothetical example. Staff comments say leaders never listen. Exit data shows departures clustered in two teams under the same director, while other teams report the opposite. That is a local, real problem with a wider story attached. A company-wide listening campaign would treat the story and leave the two teams unchanged.

Signs You May Be Dismissing a Real Problem as a Perception Problem

You may be dismissing a real problem if every explanation you hear places the cause in what employees understand rather than in what leaders do. Four signs are worth watching.

"They just don't understand" keeps coming up. Nobody has asked what people do understand, or what they have seen.

The concern survives clear communication. If the facts have been explained more than once and the concern remains, the explanation was probably not the missing piece.

The people raising it are hard to dismiss. When experienced, high-performing or well-regarded colleagues describe the same pattern, it becomes difficult to file it under a loud minority.

The feedback is getting quieter. Silence looks like improvement. It can also be the point at which people decide that raising concerns is pointless.

The opposite error also happens. A survey with a low response rate, one vivid incident or a handful of vocal critics can make a contained problem look organisation-wide. That is why the consistency test comes before any decision about scale.

What to Do Once You Know Which One You Have

Match the response to the diagnosis. A real culture problem calls for changes in behaviour and in what the organisation rewards. A perception problem calls for accurate information from people employees trust. A mix calls for both, in that order.

For a real problem, name the specific behaviours, give one senior person ownership and change the incentives that keep the behaviour going. Then track the same indicators you used to diagnose it, so you can see whether anything has moved. Incentives and decision rights sit within organisational design, and leaving them untouched is one reason culture transformation efforts fail.

For a perception problem, find out where the belief came from and correct it with specifics. Gallup describes the manager as the channel through which employees experience their employer's culture, which suggests line managers may carry more credibility here than a broadcast message. Treat that as a working assumption rather than a rule.

If I had to pick one place to start, it would be the fifth test. Asking what the organisation rewards and tolerates is uncomfortable, but it moves the discussion from what people think to what the organisation does. Everything else can be built from there.

Start With the Evidence, Not the Label

A culture problem or perception problem can look identical from the leadership floor. The difference lies in what stands behind the impression. If it is behaviour that repeats and is rewarded, the culture needs to change. If it is a misunderstanding, the information needs to change. If nobody has checked, then what the organisation has is an opinion, not a diagnosis.

Perception is not the opposite of reality. It is often the first place reality shows up.

The task is to check what it is showing you.