What the alignment gap actually is
“Alignment” is usually treated as a single condition an organisation either has or lacks. In practice it is two separate flows, moving in opposite directions, that can each fail independently. One is upward: does information about what is actually happening at the front line reach the people who set direction, in a form accurate enough to act on. The other is downward: does the intent behind a decision survive its journey outward, or does it arrive at the front line as something narrower, vaguer, or subtly different from what was decided. An organisation can have one of these flows working well and the other badly broken, and from the inside it will look like a single, undifferentiated problem — “we’re not aligned” — when it is really two distinct failures with different causes.
This distinction matters because the popular framing of the problem — that leadership and the front line don’t understand each other, or don’t trust each other, or aren’t “on the same page” — assumes a communication deficit that better messaging would close. Research on strategy execution complicates that assumption directly. A multi-year study of more than 250 companies and several thousand managers by Donald Sull (MIT Sloan School of Management) and Rebecca Homkes (London Business School), examined where execution actually breaks down. Their finding was not that vertical alignment mechanisms were weak; in most of the firms they studied, those mechanisms were sound. The sharper problem, in their data, was coordination across units at the same level — different functions, different teams — rather than clarity moving up and down a single chain of command.
That finding does not make the leadership–front-line gap this guide examines any less real. It reframes it. The gap this guide is concerned with is not primarily a values problem or a trust problem. It is what happens to information and intent as they pass through layers, roles, and time — a structural property of organisations with any distance between decision and execution, present even where the culture is open, the leadership is approachable, and nobody is hiding anything.
Why it happens even when nothing is hidden
The most counter-intuitive feature of this gap is how rarely it requires bad faith to explain. Several distinct mechanisms produce it independently, and they compound.
The first is interpersonal risk. Amy Edmondson’s field study of 51 manufacturing work teams, published in Administrative Science Quarterly, found that team psychological safety — not team confidence or team efficacy — predicted whether teams engaged in learning behaviour, including surfacing problems and errors. Team efficacy did not predict this once psychological safety was accounted for. In other words, a team can believe strongly in its own competence and still stay quiet about what it sees going wrong, if the perceived cost of speaking up is high enough. This is one route by which front-line reality fails to travel upward: not concealment, but a rational calculation about what is safe to raise.
The second mechanism does not require any perception of risk at all. Information can degrade purely through successive, individually reasonable acts of summarising. Each person passing a problem upward compresses it slightly — drops a caveat, rounds a number, softens a description — because that is what summarising for a busy audience requires. No single step distorts much. But a report that has passed through three or four such steps can arrive materially different from what was first observed, without anyone along the chain having done anything an outside observer would call dishonest. This is a property of the chain itself, not of the people in it.
A third mechanism works on the downward flow rather than the upward one. Steven Kerr’s analysis of reward systems — examined across a manufacturing company, an insurance firm, and settings in politics, medicine and academia — found a recurring pattern: organisations frequently reward one set of behaviours while explicitly hoping for a different one, and behaviour reliably follows the reward rather than the stated hope. Applied here, this means leadership intent can be communicated with complete clarity and still fail to shape front-line behaviour, because the measurement and incentive system the front line actually operates under is pointing somewhere else. The front line is not misunderstanding the message; it is correctly reading the system underneath the message.
A fourth cause is more structural still. Michael Beer and Russell Eisenstat’s research on strategy implementation, published in Sloan Management Review, identified a set of mutually reinforcing “silent killers” of implementation and organisational learning — among them a top-down or laissez-faire senior management style, unclear strategy and conflicting priorities, and poor vertical communication itself. Their central finding was that these barriers reinforce one another, so that addressing any one in isolation does little; a change in vertical communication practice, for instance, tends not to hold if the senior team’s own style or the clarity of stated priorities has not also changed.
There is also a timing dimension worth naming separately from cause: the forums, channels and habits that once carried information reliably between levels are frequently built for an organisation of a particular size. Larry Greiner’s account of organisational growth observed that the practices which resolve the problems of one stage of growth routinely become the problem of the next — what he called the irony of watching “a major solution in one period become a major problem in a later period.” A weekly stand-up that let a 40-person leadership team hear directly from operations does not scale to 400 people; nobody decided to stop hearing from the front line, but the mechanism that once delivered that hearing quietly stopped working as the organisation grew past the size it was built for.
Finally, ownership ambiguity compounds all of the above. Research by Paul Rogers and Marcia Blenko, based on a Bain & Company survey of executives at more than 350 global companies, found that decision-making in most organisations stalls at one of a small number of recurring points — between global and local levels, between the centre and business units, between functions, or between an organisation and its outside partners. Where it is unclear who actually holds a given decision, a front-line observation has nowhere reliable to land; it is passed toward whoever seems likeliest to own it, adding another layer of interpretation and another opportunity for the signal to attenuate.
How it shows up — and where it doesn’t
The gap rarely announces itself directly. It shows up as a leader who is repeatedly surprised by a problem that had been visible, in some form, weeks or months earlier — not because nobody knew, but because what was known did not survive the trip upward intact. It shows up as an initiative reported as “on track” until it very suddenly is not, because status reporting had been quietly optimised for reassurance rather than accuracy. It shows up as a written policy that almost nobody follows in its literal form, with a parallel, unwritten version that everyone actually uses — not out of disregard for leadership, but because the unwritten version is what the incentive and workload realities of the job actually permit. And it shows up as an “exception” that was granted once, for good reason, and then quietly became the default route, without anyone formally deciding that it should.
Context changes how the gap presents. In a young or small organisation, leadership and the front line are often in close physical or organisational proximity, and the gap barely registers — not because the underlying mechanisms are absent, but because there are too few layers for them to compound. As layers are added, each of the mechanisms above gets more room to operate, and the gap widens even if every individual manager is behaving exactly as they did when the organisation was smaller. Sector matters too: a professional services firm, where authority is often distributed among partners rather than concentrated in a formal hierarchy, experiences a version of this gap that looks different from a multi-site operation coordinating dozens of physically separated locations through the same small leadership team, though both are instances of the same underlying pattern. Distributed and remote working arrangements tend to remove the informal, unplanned encounters that once caught some of what formal channels missed, which can make a gap that already existed more visible without necessarily making it larger.
A common misconception is worth naming directly: that a visible gap between leadership and the front line indicates a trust problem, and that the appropriate response is more communication — more all-hands meetings, more surveys, more cascading of messages downward. The research above suggests this often addresses the symptom rather than the mechanism. More communication volume, moving through the same layered structure, subject to the same summarisation and incentive effects, can produce more information without producing more accurate information. Some organisations discover this directly: reporting cadence increases, dashboards multiply, and the gap persists or widens, because the structural mechanisms producing it were never addressed by adding volume to the channel that runs through them.
This points to a genuine tension rather than a solvable inefficiency. Every mechanism that reduces distortion — shorter reporting chains, more direct access to the front line, fewer intermediary approvals — also adds a demand on leadership attention that competes with everything else leadership is responsible for. There is no configuration that eliminates the trade-off; organisations differ in where they choose to sit on it, and that choice itself tends to shift as the organisation’s size and complexity change, which is one reason a workable arrangement at one stage becomes an unworkable one at the next without anyone changing it.
What the research shows
Taken together, the research cited in this guide converges on one point and diverges on another. It converges on treating the leadership–front-line gap as structural rather than motivational: Edmondson locates one source of it in the interpersonal risk of speaking up; Beer and Eisenstat locate another in mutually reinforcing patterns of senior-team behaviour and communication practice; Kerr locates a third in the mismatch between stated priorities and what is actually rewarded; and Sull, Homkes and Sull found that in most of the organisations they studied, the vertical mechanisms themselves were sound — the sharper failure sat in coordination across units. None of these bodies of work frames the gap as a matter of individual honesty, effort, or intent.
Where they diverge is in emphasis and remedy-adjacent implication, which this guide does not resolve and does not attempt to. Kerr’s work points toward incentive design as the lever that matters most; Edmondson’s toward the interpersonal climate a team operates in; Beer and Eisenstat’s toward senior-team behaviour as the barrier that has to shift before others can; Rogers and Blenko’s survey work points toward decision ownership as the friction point; and Greiner’s account points toward organisational age and size as the variable that determines which mechanism is dominant at any given moment. A leader trying to diagnose a specific instance of this gap is choosing, whether consciously or not, among these different possible primary causes — and the evidence base does not currently offer a single ranking of which cause dominates in general, only that several distinct, independently sufficient mechanisms exist and typically operate together.
How this presents in practice
The following is an invented illustration, not a description of any real organisation.
Consider an organisation where a senior leader is consistently surprised by problems that had been known, in some form, at lower levels weeks earlier. The obvious explanation — that people are afraid to raise things with him — does not hold up well against what people who work with him actually say: he is approachable, and by every account has never reacted badly to bad news. What complicates the simple explanation further is that each intermediate level did pass the information upward. Each summary offered along the way was, on its own, a reasonable, professional compression of what came before it — slightly more measured, slightly more resolved-sounding, than the input it was based on. No single step in the chain distorted much of anything.
The deeper pattern is that the information degraded through several rounds of individually reasonable summarisation. Nobody suppressed anything, and nobody along the chain did anything an observer would call dishonest. The chain itself simply attenuates signal at every link it passes through, in the same way a message loses detail each time it is retold. The question this raises is not who failed to speak up, but how many summarisations sit between a problem as it is first observed and the person who is ultimately responsible for acting on it — a number that tends to grow, quietly, as an organisation adds layers, and that nobody is likely to be tracking.
For how this pattern presents in other kinds of organisation, see other sectors.
Questions worth considering
- How many layers of summarising sit between a front-line observation and the person who is ultimately responsible for acting on it?
- Where in the organisation is alignment currently assumed rather than actually checked?
- Which channels or forums for hearing from the front line were built for a size the organisation has since outgrown?
- Where does the measurement or reward system point in a different direction from a stated priority — and which one does behaviour actually follow?
- Is a given silence better explained by interpersonal risk, by incentive design, or by the ordinary mechanics of summarisation — and would the answer change what, if anything, is worth doing about it?
This guide sits within Leadership. A related guide: Decision Rights: The Written Rule and the Rule in Force.
Your team is facing similar challenges?
Subscribe to The Riddle Ledger
Perspectives are sent as they are published, with no promotional email, and you can unsubscribe at any time.