Culture Survey: 10 Hidden Culture Problems Employee Engagement Surveys Miss

The strongest single item in this survey was about accountability. When employees rated “My leader holds people accountable for their performance,” the average was 5.60 on a seven-point scale and 60.6% answered favorably, which put it near the top of all 39 items. Any standard employee survey report would file that under strengths and move on.

The trouble is that accountability comes in two very different varieties and a single item can't distinguish between them. There's the kind where people know what's expected, own their results, and can tell you when something has gone wrong. And there's the kind where people know what's expected, own their results, and would never dream of admitting a mistake because of what happens to the person who admits it. Both of those produce a high accountability score. Only one of them is worth having.

So we paired the accountability item with three items about what happens when somebody speaks up: whether you can raise a concern with your leader without worrying about the reaction, whether you can report an error that could create a serious problem without causing problems for yourself, and whether people here focus on solving problems rather than assigning blame. Averaged together, those three scored 5.07 with 31.0% favorable, which is 0.53 points below the accountability item.

A gap of 0.53 points company-wide isn't much of a finding, and if that were the whole analysis I wouldn't be writing about it. The finding is what happens when you split it.

Reading down the chart, Operations sits at the top with an accountability score of 6.25, which is the highest of any department in the company, and a speaking-up score of 4.61, which is close to the lowest. That's a gap of 1.64 points among 232 people. Sales shows the same pattern at 1.08. At the other end, Finance and People & Culture both come out negative, meaning candor scores higher than accountability there, and nine of the twelve departments land less than half a point apart.

The manager-level picture is wider still. Among the 77 managers with at least a dozen responses, the gap ran from −1.02 to +2.28, and ten managers combined an accountability score above the company average with a speaking-up score below it. If you're a CEO looking at a 0.53 company number, you have nothing to act on. If you're looking at ten named managers whose teams say they're held to a high bar and can't safely report a mistake, you have a coaching agenda.

The three speaking-up items don't behave the same way

When we broke the speaking-up composite back into its three parts and compared each one separately against accountability, the composite turned out to be hiding a good deal of variation.

Raising a concern with your own leader trails accountability by only 0.25 points. Reporting a serious error trails it by 0.45. But whether people focus on solving problems rather than assigning blame trails it by 0.90 points, which is three and a half times the first gap. Only 2.1% of employees gave a favorable accountability score while giving an unfavorable score on raising a concern with their leader, while 10.0% did so on the blame item.

That pattern is worth sitting with, because it says the relationship with the immediate manager is mostly intact and the thing that breaks is what happens organizationally once a problem surfaces. The correlations point the same direction. Accountability correlates 0.63 with being able to raise a concern with your own leader, and only 0.40 with the organization solving rather than blaming, so the accountability employees describe travels closely with their own manager and the blame response doesn't. The item doesn't identify who does the blaming, so I won't name a level, but whatever produces it isn't the same thing people are describing when they talk about their boss.

Departments differ in shape and not just in severity. Operations is worse on all three items, which is a straightforward problem. Sales is fine on reporting errors and poor on the other two, which is a different problem with a different fix. People & Culture is actually negative on two of the three, meaning candor beats accountability there, and it's the department you'd want to study rather than repair.

When a strategy stalls, the usual diagnosis is that people don't understand it, and the usual remedy is more communication. Another town hall, another deck, another set of cascaded talking points. Sometimes that's the right call. Often it's money spent on the one thing that was already working.

Executing a strategy asks five separate things of an individual employee, and our culture survey measures each of them with its own question. The person has to understand how they're expected to think and act differently. They have to believe the goal can be reached without compromising the company's values. They have to feel empowered to act when they see a problem or an opportunity. Their department needs enough people to meet the expectations placed on it. And when they find a process that's wasting time, they have to be able to get it changed.

Reported one at a time, those five items look like an organization in reasonable shape: 63.7%, 53.5%, 72.2%, 48.4% and 47.1% favorable. Two of them are soft, three are respectable, and none of them looks like an emergency. But strategy execution isn't an average. It's a chain, and a person who fails any one link can't execute regardless of how they scored on the other four.

Start with all 1,354 employees. 862 of them understand how they're expected to work differently. Of those 862, only 604 also believe the goal is reachable within the company's values. Of those 604, 503 also feel empowered to act. Of those 503, 373 also say their department has enough people. And of those, 310 can also change a wasteful process when they find one. That's 22.9% of the workforce with everything required to actually execute.

The chart makes the drop points easy to see. The single largest loss happens right at the start, where 36.3% of employees don't clear the understanding question, so communication genuinely is part of the problem. But the second-largest loss is the values question, which costs another 29.9%, and the staffing question costs 25.8%. Those two have nothing to do with clarity. You could explain the strategy perfectly to somebody who thinks hitting it requires cutting corners, or to somebody whose team is three people short, and they still won't execute it.

This is the analysis I'd put in front of a CEO first, because it converts a vague complaint about execution into a count of people and a named blockage. If most of your loss is at the understanding step, communicate more. If most of it is at staffing, no amount of communication will help.

Ambitious goals are supposed to be motivating, and in this organization they mostly are. The item “Our most ambitious goals push me to find better ways of working” scored 5.25 with 50.3% favorable, which is a healthy result and the sort of thing a leadership team would be pleased to see.

What that item can't tell you is the price people think they're paying. Understanding how you're expected to work differently to hit the goal came in at 4.95, which is already a step down, and three further items about the cost of the goal all scored lower than that. Whether employees believe the goal can be achieved without compromising the company's values came in at 4.55. Whether they can express concerns about the goal without negative consequences came in at 4.35. Whether they can miss an ambitious goal without fearing for their job came in at 4.30.

The combination that matters is the one where somebody is driven by the goal and has a problem with how it has to be reached. In this organization, 18.4% of employees agreed that ambitious goals push them to work better while also disagreeing either that the goal can be reached without compromising values or that they can raise concerns about it without consequences. That's 249 people. This is the one analysis on the page that uses a wider rule than the rest: agreement here means 5 or above rather than 6 or 7, because the group worth worrying about includes people who are only moderately driven.

The survey doesn't observe anybody actually cutting corners, and that distinction matters. What it identifies is the population in which corner-cutting becomes more plausible, since these are the people motivated to hit a number they don't believe can be hit cleanly, or who can't say so if they think it can't.

Splitting that 18.4% shows three different situations rather than one. 5.4% doubt the values question but say they could raise the concern if they wanted to, 8.0% say they can't raise concerns though they don't doubt the values, and 5.0% have both problems at once. The two groups that can't raise concerns are the ones I'd worry about first, because somebody who sees a problem and can't say so is the mechanism by which an ambitious goal quietly becomes a scandal.

By department, the share of employees in that at-risk group ran from 10.4% to 26.2%. That range is a reminder that pressure isn't distributed evenly by the company. It's distributed by the manager and the operating rhythm of a particular group.

Most employee surveys ask about the immediate supervisor and about senior leadership, and then report both. What they don't usually do is compare them at the individual level, which turns out to be where the interesting result lives.

Across four immediate-manager items, the average was 5.35 with 40.5% favorable. Across the four matching senior-leadership items, the average was 4.54 with 18.8% favorable. That's a gap of 0.81 points and 21.7 favorability points, and the two sets of items correlate at only 0.48, which means how somebody feels about their own boss tells you comparatively little about how they feel about the executive team. (Two further items in those sections, both about accountability, sit outside these composites because they carry their own analysis further down the page.)

The clearest single comparison is a matched pair. “When I share my work problems with my leader, they respond constructively” scored 5.55 with 59.1% favorable. The senior-leadership version of the same question, “Senior leadership responds constructively when employees raise problems,” scored 4.55 with 32.2% favorable. Same behavior, two altitudes, a full point apart. 10.9% of employees rated their own manager favorably and senior leadership unfavorably on that identical behavior.

Eleven of the twelve departments point the same direction, with the gap running from 1.46 in Product down to 0.41 in Finance. Sales is the one exception at −0.26, where senior leadership rates a little higher than the immediate manager, and that fits something we'll come back to later, since Sales also has the lowest score in the company on raising a concern with your own leader. Everywhere else the pattern holds, and the consistency is itself the finding. A gap in one department is easy to explain as a local personality conflict. A gap running the same direction in eleven is much harder to dismiss that way.

The practical risk here is that good management masks the problem. A company with strong frontline managers will post decent overall leadership numbers, and the executive team can read those numbers as a verdict on themselves. In this organization the frontline managers are carrying a senior-leadership score that would look considerably worse on its own, and because both sets of items get averaged into one leadership category on most dashboards, nobody ever sees it.

Almost every organizational culture survey asks whether people live the values, and almost every one of them gets a decent answer. That's because the question is usually asked about peers, and peers generally do behave well toward each other. The more useful question is where in the organization the values stop being binding.

We arranged four items as a ladder, from what colleagues do every day up to whether leaders themselves are bound by the standards. Each rung asks about the same values in a different place.

What people's peers do every day scored 5.55 with 59.1% favorable. Whether senior leadership makes decisions consistent with the mission and values scored 4.75. Whether the people being promoted into leadership reflect the values scored 4.45. And whether leaders at all levels are held to the same standards of behavior scored 4.40 with 28.8% favorable. From the top rung to the bottom, that's a drop of 1.15 points and 30.3 favorability points.

Taken together, those four numbers say employees see the values applying much more strongly to peer behavior than to promotions and to accountability higher up. That's a specific and fixable problem, and it's invisible if you report a single values score, because the peer item props the average up.

The sharpest version of the finding is the overlap between the top rung and the bottom one. 12.1% of employees, which is 164 people, gave a favorable answer on whether their peers consider the people they serve and an unfavorable answer on whether leaders are held to the same standards of behavior. Those are the employees telling you the values are real at eye level and stop being real above it.

Every organization has an official story about how people get promoted and an unofficial one that employees actually learn to work with. The gap between the two is a real cultural feature, and you can measure it without asking a single question about politics or favoritism.

The method is a correlation analysis. We took the item “I know what I need to do to advance here,” which 46.1% of employees answered favorably, and asked which of the other 34 items moves most closely with a favorable answer. These are plain correlations and none of them establishes cause, but whatever tracks career clarity most tightly is a reasonable read on what employees have concluded the real rules are.

Three items sit at the top and they're close enough together to be read as one answer: having opportunities to develop your skills at 0.606, the organization developing future leaders who preserve the culture at 0.604, and the people being promoted reflecting the values at 0.594. Then there's a clear break, and the fourth item comes in at 0.419. The employees who say they know how to get ahead are also the ones who say they can see a development path and that the promotion decisions are consistent.

The more interesting half of the chart is the bottom. Every one of the speaking-up items ranks in the bottom half. Being able to express concerns about the strategy ranks 31st of 34. Being able to report an error ranks 32nd. Whether people solve problems rather than assign blame ranks 33rd at 0.277, and being able to raise a concern with your own leader ranks dead last at 0.255. In plain terms, the people who know how to advance here are not disproportionately the people who feel safe raising problems, and the two things are close to unrelated.

I want to be careful about the interpretation, because a correlation of 0.255 doesn't prove that speaking up hurts your career. What it establishes is that career clarity and candor aren't traveling together, which means the informal advancement rules employees have inferred don't include candor as an ingredient. If leadership believes it's built a place where raising problems is a path to advancement, the data says employees haven't drawn that conclusion.

Empowerment is one of those words that surveys measure badly, mostly because the standard item asks about a feeling. “I am empowered to take action when I see a problem or an opportunity” scored 5.30 with 51.7% favorable in this organization, which sounds like a company where people can get things done.

The companion item asks about an outcome instead of a feeling. “When I find a process that wastes time, I am able to get it changed” scored 4.30 with 26.7% favorable. That's a full point lower, and the favorable rate is roughly half.

The department cut is unusual for this survey, because the gap is positive in all twelve departments rather than concentrated in a few. It runs from 1.16 in Facilities down to 0.15 in People & Culture, and nine of the twelve sit within a quarter point of a full point. When a gap shows up almost everywhere at roughly the same size, you're usually looking at something structural, like an approval process or a spending threshold, rather than at a handful of controlling managers.

Averages by themselves could be explained a few different ways, so the crosstab matters. 23.3% of employees answered favorably on both items, and those are the people who feel empowered and can point to something they changed. 5.8%, or 79 people, said they feel empowered while saying they can't get a wasteful process changed. That second group is small, and it's the group I'd want to talk to first, because they've been told they have authority and have discovered they don't.

This one matters for a practical reason. Empowerment initiatives are usually launched on the strength of the feeling item, and they usually consist of encouragement. When the feeling item is already at 5.30 and the outcome item is at 4.30, more encouragement isn't the intervention. Somebody needs to find out what actually happens when an employee tries to change a process, and who says no.

This is the same accountability item from the first analysis, pointed in the other direction. When employees rated whether their leader holds people accountable for performance, the score was 5.60 with 60.6% favorable. When they rated whether leaders at all levels are held to the same standards of behavior, the score was 4.40 with 28.8% favorable.

The gap is 1.20 points and 31.8 favorability points, though the more telling number is the overlap. 13.1% of employees gave a favorable answer on whether their leader holds people accountable and an unfavorable answer on whether leaders are held to the same standards. That's roughly one employee in eight explicitly describing an accountability system that runs downhill and, as far as they can see, doesn't run back up.

What makes this worth acting on rather than filing as a grievance is where it lands in the correlation rankings. When we ranked all 34 items by how closely each moves with knowing how to get ahead, whether leaders are held to the same standards came in 7th. Against the outcome items generally it ranks 2nd of 34. That's unusually high for an item most engagement surveys don't ask at all.

One caution on interpretation. Employees see their own manager's consequences and rarely see what happens to a leader two levels up, so some of this gap is a visibility problem rather than a fairness problem. That distinction matters for the fix, because the answer to a visibility problem is telling people what happened, while the answer to a fairness problem is something considerably harder.

Psychological safety gets reported as a single number on a great many dashboards, and that single number is usually an average of several items. Averaging is the problem, because in this organization the six psychological safety items ranged from 5.35 down to 3.70, a spread of 1.65 points, and the average of 4.61 doesn't describe any of them.

Raising a concern with your own leader scored 5.35. Reporting a serious error scored 5.15. Raising a concern with anyone regardless of position or department scored 4.80. Questioning the strategy scored 4.35, and missing an ambitious goal without fearing for your job scored 4.30. The lowest item, which is reverse-worded and asks whether there are people outside your department whose reaction you worry about, scored 3.70 after reverse-scoring.

Those six numbers describe a coherent situation rather than a random scatter. Safety is highest inside the relationship with the direct manager, drops when the audience widens beyond that relationship, and drops furthest when the topic is the strategy or a missed number. A composite of 4.61 would send a leadership team looking for a general trust problem, when what the data actually shows is a specific one about lateral relationships and about goals.

The departmental cut adds a second layer. Every department has its own profile across the six items, and the profiles are different shapes rather than the same shape at different heights. Sales scores lowest of any department on raising a concern with its own leader at 3.93 while scoring above the company average on raising concerns with anyone, which is an unusual inversion. Operations scores near the bottom on the strategy and missed-goal items specifically. People & Culture is strong on the manager items and weakest on the missed-goal item. Four departments have a different worst item than the company does, so a company-level psychological safety score would point four of your twelve departments at the wrong problem.

Silos is the word organizations use for at least three separate conditions, and because the word covers all of them, the remedy tends to be a cross-functional offsite that addresses none of them.

Our culture survey measures three different things that people call silos. Whether employees from different departments are working toward the same goals scored 4.50. Whether people in other departments follow through on the commitments they make to you scored 4.55. And whether there are people outside your own department whose reaction you worry about when you need to raise an issue scored 3.70 after reverse-scoring, with 47.1% of employees agreeing that such people exist.

The correlations are what separate them. Goal alignment and follow-through correlate at 0.64, which is well above the 0.42 average correlation between items on this survey, so those two really are one problem with two symptoms. But each of them correlates only about 0.44 with the wariness item, which is close to the survey-wide baseline. Statistically, cross-department wariness is behaving like a separate condition, so fixing goal alignment alone is unlikely to touch it.

That distinction has a direct implication for what you'd do about it. Misaligned goals and broken commitments respond to structural fixes: shared metrics, clearer handoffs, service agreements between departments. Cross-department wariness is unlikely to respond to any of that, because the employee isn't confused about the goal. They're avoiding a specific group of people.

Which departments people are avoiding

Because the wariness item has a follow-up asking which departments, we can name them. The 1,354 employees produced 970 mentions, and laying those out in a grid shows both who gets avoided and who does the avoiding.

Read down the columns and you get the departments people hesitate to approach. Compliance & Risk was named by 180 people in total, which is 13.3% of the entire workforce. Finance was named by 149, Legal by 122, and Information Technology by 116. The pattern is recognizable to anybody who has worked in a large organization, because those four are gatekeeper functions whose job is partly to say no, and being named here isn't automatically a sign that they're doing anything wrong.

Read across the rows and you get the departments doing the worrying. Operations, Customer Support and Sales fill most of the grid, and each of them names the same gatekeeper functions at the top of its own list, which suggests the three big operating groups are running into the same wall rather than each having a private feud.

The number that complicates the easy explanation is what those departments say about themselves. Inside Compliance & Risk, 69.8% of the 63 employees said there are people outside their department whose reaction they worry about, which is the highest rate in the company and well above the 47.1% average. The department most avoided is also the department most avoidant. That's harder to explain as a function of the role, and it's the sort of finding that would never appear on a survey report built from item averages.