The Friction Tax Hidden in Your Processes

Nobody in the room had decided this should happen. A request from a customer-facing team needed sign-off from two other departments before it could proceed. Each department had, independently and reasonably, established a process for reviewing requests like this — a weekly meeting, a standard checklist, an assigned owner. None of these processes was badly designed in isolation. Combined, they produced something closer to a relay race with three separate starting guns: the customer-facing team submitted the request, waited for department one's weekly review cycle, then waited again for department two's, then convened a fourth meeting simply to reconcile what the first two had each decided, because neither had been designed with the other's process in mind.

Six weeks passed between the original request and the work actually beginning. No individual in this chain behaved unreasonably. Every review step existed for a defensible reason. And the organization, as a system, took six weeks to do something that any single competent team, working alone, could have started in days.

This is the phenomenon this article investigates, and it deserves a more precise name than the usual complaints — "we're too slow," "too much red tape," "death by meetings" — tend to give it. Call it the friction tax: the aggregate cost, largely invisible in any single team's own metrics, of coordinating work across the boundaries between people, teams, and functions. It is not the same as inefficiency, and it is not simply the price of doing things properly. It is what remains once every individual step in a process has been justified on its own terms, and the organization has still failed to ask the one question none of those individual justifications was designed to answer: what does this cost, in total, once every reasonable step is added together?

This connects to concerns already on a CEO's own agenda, not only an operations lead's. Customer commitments depend on whether a promised delivery date survives contact with every internal boundary it has to cross before the work can actually begin. Cost structure depends on how much of total headcount capacity is effectively spent coordinating rather than producing. And competitive speed depends on whether a genuinely urgent priority can move through the organization faster than its ordinary priorities do, or whether every request, urgent or not, is routed through the same fixed sequence of reviews regardless of what it actually needs.

The instinct is to treat this as an operations problem, delegated downward to whoever owns process improvement. That instinct undersells what is actually at stake. Coordination friction sits directly beneath four things a CEO is genuinely accountable for: how fast strategic priorities actually move from decision to delivery, how much of the organization's total capacity goes toward producing outcomes versus managing the handoffs between the people producing them, how reliably a customer-facing commitment survives contact with the rest of the organization, and how much slower the whole system gets as headcount and complexity grow, independent of whether any individual team has gotten worse at its job.

None of this shows up cleanly in the metrics a CEO typically reviews. Revenue, margin, headcount, and individual project timelines each answer a real and important question — but none of them is designed to answer the specific question this article is about: how much of the organization's total effort is being spent on the coordination required to get work across its own internal boundaries, as distinct from the work itself? A team can hit every deadline it controls directly while the organization as a whole is quietly losing weeks to friction at every boundary that team touches, and the team's own metrics will never show it, because the metrics were built to measure the team, not the seams between teams.

This is a different question from the one this Journal's earlier research on sustainable performance addressed. That research asked whether current output is being financed by depleting future capacity. This article asks something adjacent but distinct: whether a meaningful share of current effort — not future capacity, but effort happening right now — is being spent on coordination that produces no proportionate progress, in a way ordinary performance metrics have no mechanism for surfacing.

It is worth being precise, because "friction" as ordinarily used in business conversation blurs together several genuinely different things. This article uses it specifically to mean the coordination cost incurred when work crosses an organizational boundary — between two teams, two functions, two levels of a hierarchy, or two decision-makers who each hold part of what's needed to move forward. That cost is real even when every process involved is working exactly as designed. It is not synonymous with a broken process, a badly designed meeting, or an unnecessary approval — those are specific causes of excess friction, not friction itself.

Gerardo Okhuysen and Beth Bechky's comprehensive review of coordination research, published in 2009 in The Academy of Management Annals, frames coordination as the process of interaction that integrates a set of interdependent tasks — and argues that coordination mechanisms such as routines, meetings, plans, and schedules serve organizations by producing three specific conditions: accountability, predictability, and common understanding across the people who depend on each other's work (Okhuysen & Bechky, 2009). This framing matters because it makes explicit what the opening scenario's three separate review processes were each, individually, trying to produce — accountability for a decision, predictability of a review cycle, shared understanding of what was being approved. Each mechanism was doing real work. The friction cost emerged not from any single mechanism failing, but from three legitimate mechanisms operating without reference to each other.

This is the conceptual move this article asks a reader to make: friction is not the opposite of good process. It is the coordination cost that exists whenever interdependent work crosses a boundary, and the executive question is never whether that cost can be eliminated — Okhuysen and Bechky's own framework treats some coordination cost as the unavoidable price of accountability, predictability, and shared understanding, none of which an organization can simply do without. The question is whether the coordination currently happening is proportionate to what it produces, or has quietly grown past that point.

It is worth being equally precise about what friction is not. It is not simply the presence of more than one team in a process — most valuable work in any organization of real size requires more than one team, and treating multi-team involvement itself as the problem would argue for a level of consolidation most organizations correctly reject for other good reasons. It is also not simply slowness — a genuinely complex, high-stakes decision that takes real time to coordinate correctly is not friction; it is coordination doing exactly what Okhuysen and Bechky's framework says it exists to do. Friction, specifically, is coordination cost that exceeds what the underlying interdependence and stakes actually require — which is a comparison against a baseline, not a judgment about speed or team count in isolation.

Friction is structurally difficult for any single person or team to see completely, for a reason connected directly to this Journal's earlier research on organizational blind spots: each team involved in a multi-step coordination chain experiences only its own segment of it. The customer-facing team in the opening scenario saw its own submission and its own eventual approval. It did not directly experience the internal review cadence of either downstream department, and neither downstream department directly experienced how long the requesting team had already been waiting before the request reached them. The total six-week cost existed only at a vantage point none of the three teams individually occupied.

This produces a specific measurement problem worth naming precisely: visible process time and hidden coordination time are not the same quantity, and an organization that only measures the former can have an accurate, well-instrumented view of every individual step while remaining blind to the sum. Each department's own dashboard might correctly show its review cycle completing on time, every time — and none of those individually accurate dashboards would ever show the six weeks the request spent, in aggregate, moving between them.

There is a second, compounding reason friction stays invisible even to people who directly experience its cost: each individual instance tends to feel minor in isolation. A two-day wait for one approval, a single clarifying round-trip on one specification, one meeting scheduled to resolve a misunderstanding that a clearer initial handoff would have prevented — none of these, encountered once, feels like a systemic problem worth escalating. It is only in aggregate, across many such instances accumulating at the same recurring boundary, that the pattern becomes a genuine cost, and aggregation across many small instances is precisely the kind of pattern individual, day-to-day experience is poorly equipped to notice on its own.

It would be a mistake to read this article as an argument that coordination itself is the problem, and it is worth being explicit about why, because the more useful distinction is not between coordination and its absence, but between coordination that produces proportionate value and coordination that consumes capacity without it.

Paul Adler and Bryan Borys, in a 1996 Administrative Science Quarterly paper, offer a precise conceptual tool for exactly this distinction, applied to a closely related phenomenon: bureaucracy and formal process. They propose that organizational formalization comes in two genuinely different types — enabling formalization, which gives people the tools, context, and shared reference points to coordinate effectively while preserving their judgment, and coercive formalization, which constrains behavior through rigid rules and monitoring regardless of whether it actually helps the work get done (Adler & Borys, 1996). Critically, their argument is that the same underlying mechanism — a checklist, a review step, a standard form — can function as either, depending on how it is designed and how people experience using it, not on some fixed property of formal process itself.

This reframes the friction question usefully. The three review processes in the opening scenario were not obviously coercive in Adler and Borys's sense — each existed for a real reason, connected to real accountability. What made their combination destructive was not that any one of them was badly designed, but that none of them had been evaluated against what it cost once combined with the others. A CEO applying this distinction is not asking "should we have fewer approval steps" as a general policy — Adler and Borys's own research would caution against that blanket conclusion. The more precise question is: which specific coordination steps genuinely enable better decisions, and which have drifted into coercive formalization that consumes effort without a corresponding gain in accountability, predictability, or shared understanding?

This is also the point at which accountability and control are worth separating explicitly, since Adler and Borys's enabling-versus-coercive distinction is, at its core, a distinction between these two. Accountability means a decision or a piece of work has a genuine, identifiable owner, and that ownership is meaningfully reviewed. Control means behavior is constrained and monitored, whether or not that constraint improves the outcome. The two frequently travel together in a well-functioning process — a real owner, meaningfully reviewed, naturally produces something that looks like control from the outside. They diverge specifically when the monitoring persists after the accountability it was meant to serve has become nominal: a review step that no longer changes any decision, retained because removing it would require someone to formally take ownership of that removal, is control that has outlived the accountability it once represented. A related and equally common divergence concerns compliance and effectiveness: a process can achieve full compliance — every step completed, every box checked — while no longer being effective at producing the coordination outcome those steps were designed to protect, which is precisely how the opening scenario's three individually compliant review processes produced a collectively ineffective result.

This is also where the earlier distinction between bureaucracy and governance becomes concrete rather than semantic. Governance, in the sense worth preserving, is coordination that produces genuine accountability, predictability, or shared understanding proportionate to what it costs. Bureaucracy, in the pejorative sense executives actually mean when they use the word, is coordination that has drifted past that point — retained less because it still produces those three conditions and more because removing it would require someone to take ownership of a change no one currently owns. The two can look identical on an org chart. They are not identical in what they're actually accomplishing.

Before turning to where friction actually accumulates, several terms this article relies on deserve the same precision this Journal applies throughout, because each is routinely collapsed into an adjacent idea in ordinary executive conversation.

Process complexity and task complexity are not the same thing, and the gap between them is one of the clearest diagnostic signals available for telling healthy coordination from destructive friction. Task complexity is a property of the work itself — how much genuine uncertainty and interdependence the underlying problem contains, in Galbraith's specific sense (Galbraith, 1974). Process complexity is a property of the organization's response to that work — how many steps, approvals, and handoffs a process actually contains. The two should track each other reasonably closely in a well-designed system: genuinely complex, uncertain, interdependent work should carry more process complexity than simple, predictable work. When process complexity exceeds task complexity — when a low-uncertainty, low-interdependence task is routed through a process built for something considerably harder — that gap is close to a direct measurement of excess friction, independent of whether any single step in the process is individually defensible.

Efficiency and effectiveness are worth separating as well, since a process can score well on one while failing the other. Efficiency describes how little is wasted in executing a given process as currently designed. Effectiveness describes whether that process, however efficiently executed, actually produces the coordination outcome it exists to produce — Okhuysen and Bechky's accountability, predictability, and shared understanding (Okhuysen & Bechky, 2009). A highly efficient process that no longer produces genuine accountability, because everyone involved treats it as a formality rather than a real check, is efficient and ineffective at the same time — and efficiency measurements, which typically track speed and resource use rather than whether the underlying coordination purpose is still being served, will not by themselves reveal this gap.

Predictability and rigidity are the final pair worth distinguishing, because the instinct to treat them as synonyms is precisely what makes over-standardization difficult to notice from inside an organization. Predictability is a genuine coordination asset: knowing in advance how a recurring situation will be handled, by whom, and on what timeline. Rigidity is what predictability becomes once it stops adapting to situations the original design did not anticipate — the same fixed response applied regardless of whether it still fits. A process can be highly predictable and appropriately so, or highly predictable and quietly rigid, and the two are indistinguishable from the outside unless someone specifically checks whether the process has ever been revised in response to a case it handled poorly.

The deepest theoretical account of why this cost exists at all, independent of anyone's competence, comes from Jay Galbraith's 1974 framework treating organizations as information-processing systems. Galbraith's central argument is that the amount of information that must be exchanged among decision-makers during task execution rises directly with task uncertainty and interdependence — and that an organization facing more of either must adopt one of a limited set of design responses: standardized rules for predictable situations, escalation up a hierarchy for exceptions, coordination through shared targets and goals, or direct lateral communication between the people who actually hold the relevant information (Galbraith, 1974).

This matters because it locates the source of coordination cost in the structure of the work itself, not in anyone's failure to communicate well. Two departments whose work is genuinely interdependent will generate real coordination demand regardless of how skilled or well-intentioned their people are — the demand is a property of the interdependence, not a symptom of dysfunction. What varies, on Galbraith's framework, is whether an organization has matched its coordination mechanism to the actual level of interdependence and uncertainty involved. A predictable, low-uncertainty handoff coordinated through a standing rule is efficient. The same handoff coordinated instead through an escalating chain of ad hoc meetings — Galbraith's more expensive, higher-touch mechanism — is friction: not because coordination was unnecessary, but because the mechanism chosen was more expensive than the actual interdependence required.

This has a specific implication for how a leadership team should read a request for "more headcount to handle coordination." On Galbraith's framework, that request is sometimes the correct response — genuinely high-interdependence, high-uncertainty work may need the more expensive lateral-communication mechanism his framework identifies, and understaffing it produces exactly the friction this article describes. But the same framework equally predicts that adding people to a boundary that was never actually high in interdependence or uncertainty will not resolve the friction — it will most often just add another node the coordination has to route through, potentially deepening the exact problem it was meant to solve. The useful diagnostic question precedes the staffing question: is this specific boundary genuinely as interdependent and uncertain as the coordination effort currently applied to it assumes?

A closely related confusion is worth separating out on its own, because it shapes how leadership typically responds once friction becomes visible: the instinct to treat standardization and adaptability as opposing forces, where more of one necessarily means less of the other.

Galbraith's own framework points toward a more precise reading. Standardized rules are, in his terms, the least expensive coordination mechanism available — appropriate specifically for the portion of an organization's work that is genuinely predictable and low in interdependence (Galbraith, 1974). Applying that same low-cost mechanism to work that is actually uncertain or highly interdependent does not produce efficiency; it produces exactly the coercive, rigid formalization Adler and Borys warn against, because the mechanism is mismatched to the task rather than because standardization itself is the problem. The opposite failure is just as real and just as costly: treating genuinely predictable, low-interdependence work as if it required case-by-case judgment every time, which manufactures friction of its own kind, through unnecessary meetings and ad hoc decisions where a standing rule would have done the job at a fraction of the cost.

The executive-relevant version of this distinction is not "should we standardize more or less," stated as a general organizational philosophy. It is closer to: for this specific boundary, is the underlying work predictable enough that a standard rule serves it well, or interdependent and uncertain enough that it needs the more expensive coordination Galbraith's framework associates with genuine complexity? Getting this match wrong in either direction — over-standardizing complex work, or under-standardizing simple work — is itself a source of friction, independent of how well-intentioned the process design was.

There is a useful test for telling the two mismatches apart in practice. Over-standardization tends to produce a specific complaint pattern: capable people quietly routing around a rule because it doesn't fit their actual situation, the workaround dynamic described earlier in this article. Under-standardization tends to produce a different pattern: the same basic question being re-litigated in a fresh meeting every time it comes up, because no standing answer exists for something that no longer requires fresh judgment. A leadership team hearing the first complaint pattern at a given boundary is likely looking at coercive over-formalization; hearing the second, it is likely looking at a predictable interaction still being coordinated as if it were novel each time.

A further and genuinely important distinction concerns what happens when the official process and the actual work diverge — a pattern executives often describe, with some frustration, as people "not following the process." The research suggests a more precise and considerably less judgmental account of what is usually happening.

Martha Feldman and Brian Pentland's influential 2003 reconceptualization of organizational routines, published in Administrative Science Quarterly, distinguishes the ostensive aspect of a routine — the official, abstract description of how something is supposed to work — from its performative aspect: the specific, situated actions people actually take when carrying it out (Feldman & Pentland, 2003). Their central argument is that routines are not simply executed faithfully or violated; the relationship between the ostensive and performative aspects is where real variation, adaptation, and — over time — genuine change originate, meaning the gap between the documented process and the lived one is not automatically evidence of noncompliance. It is frequently evidence that the people doing the work have adapted a process that was never quite specified precisely enough for the situations they actually encounter.

This gives a CEO a genuinely different and more useful question than "why isn't the team following the process." It is: when the actual, performative version of a process has diverged from its documented, ostensive version, is that divergence a rational adaptation to a boundary the official process handled poorly — a workaround worth learning from — or is it a breakdown that is quietly generating rework and inconsistency elsewhere? Both are real possibilities, and Feldman and Pentland's framework gives no automatic answer; it only insists that the question be asked with genuine curiosity rather than assumed to be a compliance problem by default.

A concrete illustration makes this less abstract. Picture a documented process requiring every cross-team request to be submitted through a shared intake form, designed to ensure nothing gets lost. In practice, a team discovers that urgent requests routed through the form take days to be triaged, while a direct message to a known contact on the receiving team gets answered in hours — and over time, urgent requests quietly stop going through the form at all, while routine ones still do. The ostensive process — everything through the form — has not changed. The performative reality has split into two parallel paths, neither of which the original process was designed to distinguish. This is not the requesting team disregarding the process. It is the requesting team's accumulated experience revealing a distinction — urgent versus routine — the original process never made, and Feldman and Pentland's framework would treat this split as exactly the kind of adaptive signal worth learning from, not simply correcting.

Bringing the preceding sections together, friction accumulates specifically at organizational boundaries — the points where work, information, or a decision passes from one person, team, or function to another, and where each side of the boundary has typically optimized its own side of the interface without full visibility into the other's.

Consider several illustrative, hypothetical boundaries, offered here as examples rather than documented cases: a Sales team that has optimized its own proposal process for speed, handing off to an Operations team that has, separately and reasonably, optimized its own intake process for accuracy — with neither side's optimization accounting for what the other side actually needs to receive a request efficiently. A Product team specifying a feature in language optimized for stakeholder clarity, handing off to an Engineering team that needs a different level of technical precision to begin work without a clarifying round-trip. A headquarters function issuing a policy update optimized for consistency across the whole organization, received by a local team whose specific operating context the policy did not, and structurally could not, fully anticipate.

None of these examples describes a failure of goodwill on either side of the boundary. Each side made a locally reasonable optimization. The friction is not a property of either team; it is a property of the interface between them — which is precisely why fixing it by exhorting either team individually to "communicate better" so rarely works. The interface, not either team's effort, is usually where the actual cost is concentrated.

A finance function reviewing a business unit's spending request through a standardized approval workflow built for the organization's largest, highest-risk category of spend, applied uniformly to a request an order of magnitude smaller and lower-risk, is a further recognizable instance of the same pattern — a mechanism correctly matched to one kind of interdependence, mismatched to another, producing friction not because anyone erred but because the interface was never designed to distinguish between the two cases it now has to handle.

A customer-facing team and a legal function present a further, slightly different version of the same boundary problem, worth including because it illustrates process complexity outrunning task complexity directly. A standard contract renewal, materially identical to dozens processed before it, routed through the same review cycle built for genuinely novel agreements, is task complexity that has not changed being met with process complexity that was calibrated for something else entirely. Neither side is behaving unreasonably — legal's review process exists for real reasons tied to real risk, and the customer-facing team's frustration at the delay is equally reasonable given what it can see. The friction lives specifically in the absence of any mechanism distinguishing the routine case from the genuinely novel one, which is a design gap at the interface, not a failure on either side of it.

This is the deeper mechanism underlying nearly every example in this article, and it is worth stating as its own principle rather than leaving it implicit: individually reasonable, locally optimized decisions can combine into a system that is collectively slower, more expensive, or harder to coordinate than any of its individual parts would suggest. This is not a claim that local optimization is a mistake — teams are, correctly, incentivized and evaluated on their own performance, and asking every team to optimize for the whole system instead of its own function would generally make things worse, not better, since no single team has visibility into the whole system to optimize for it responsibly.

The organizational implication is that system-level friction is nobody's job by default, in most organizational structures — every function has an owner optimizing its own segment, and very few structures have anyone specifically responsible for the coordination cost that accumulates at the seams between those segments. This is the structural reason friction tends to grow quietly rather than being caught early: it lives precisely in the gap between what any single team's own metrics can see and what the organization as a whole is actually spending.

This is not a new observation about organizations in general — it is a specific, well-recognized instance of a broader pattern in how complex systems built from locally optimizing parts behave, visible whenever the incentive structure evaluates the parts without a corresponding structure evaluating the whole. What makes it worth naming specifically in the context of process friction is that the fix most leadership teams reach for first — asking each team to also consider the whole system when making its own decisions — asks individual teams to do something they are structurally poorly positioned to do well, since no team has full visibility into every other team's constraints. The more durable fix is not asking every part to see the whole system; it is ensuring someone, somewhere, has the job of watching the seams specifically, with the standing and the visibility to notice when local optimizations have begun working against each other.

A further, easily overlooked cost sits inside the mechanism this article has already described rather than alongside it: coordination itself is work, performed by specific people, and that work is rarely represented on any formal org chart or job description, even when it consumes a genuinely significant share of someone's time.

In the opening scenario, someone eventually had to notice that three separate review cycles had produced conflicting outcomes, convene the fourth meeting to reconcile them, and carry the context from each of the first three conversations into the fourth. That person's actual job title almost certainly did not include "reconcile conflicting departmental reviews." They did it anyway, because the work needed doing and no formal role existed to do it — a pattern consistent with Okhuysen and Bechky's broader point that coordination mechanisms like meetings and plans exist specifically because someone has to actively produce accountability, predictability, and shared understanding; those conditions do not arise automatically just because a process document says they should (Okhuysen & Bechky, 2009).

This hidden labor tends to concentrate on specific people rather than distributing evenly — often whoever happens to sit closest to the most boundaries, or whoever has, over time, become known as reliably able to untangle a stalled handoff. This creates a specific and under-recognized organizational risk: the person doing the most coordination labor is frequently the least visible in performance metrics built around individual or team output, precisely because coordination work, by its nature, produces smoother outcomes for other people's metrics rather than a clear metric of its own. An organization that has never asked who is currently absorbing this kind of invisible coordination labor has no way of knowing how concentrated — and how fragile — that arrangement currently is.

A useful, low-cost diagnostic follows directly from this: ask, across a leadership team, who each person would call first if a cross-team commitment looked like it was about to slip. If the same one or two names keep coming up across otherwise unrelated parts of the organization, that is a direct signal of exactly the concentration this section describes — and a specific, investigable finding, distinct from any assessment score, that a leadership team can act on the moment it's noticed.

This returns directly to the concern this article opened with. Revenue, margin, headcount, and individual project timelines are not the wrong metrics — they answer real questions a CEO needs answered. What they cannot do, by their own construction, is reveal how much of the organization's total effort is currently going toward coordination rather than toward the work that coordination exists to support. A project can complete on schedule while having consumed several multiples of the coordination effort a comparable project required a year earlier, and none of the standard project metrics is built to surface that difference, because they measure whether the deadline was hit, not what it cost internally to hit it.

This is not an argument that financial and delivery metrics are unreliable. It is an argument that they were built to answer a different question than the one this article is about, and mistaking one for the other is how an organization can look, by every visible measure, like it is performing well while quietly accumulating exactly the kind of coordination cost the opening scenario described.

This is also why repeated measurement matters more here than a single snapshot would suggest. A single assessment can show that Processes & Collaboration currently sits below an organization's other dimensions; it cannot, on its own, show whether that gap is widening as the organization scales, holding steady, or already improving in response to a change no one has formally evaluated yet. The trajectory — not the single number — is what would actually tell a leadership team whether the coordination cost this article has described is a stable, tolerable feature of how the organization currently operates, or a condition quietly getting more expensive with each additional layer of complexity the organization adds.

The preceding sections describe a genuine measurement gap: friction accumulates at organizational boundaries, in a form neither team on either side of the boundary can fully see, and in a form standard financial and delivery metrics were never designed to capture. This is the specific gap Klarwerk's Processes & Collaboration dimension is built to address, and it is worth being precise about exactly what that means.

The relevant survey items ask participants directly about their own experience of cross-team handoffs, whether agreements between teams hold up in practice rather than only on paper, and whether the same coordination failures recur at the same organizational boundaries without being corrected — a structured, anonymized signal of how coordination is actually experienced by the people doing it, aggregated across the organization. Where department-level comparisons show one function's Processes & Collaboration score diverging materially from the company-wide figure, or where the platform's Executive Tensions logic identifies a material gap between a strong Execution Index and a weaker Processes & Collaboration signal, that combination is surfaced as a specific, disclosed pattern worth investigating — consistent with, though never proof of, the kind of boundary friction this article has described.

This connects to a specific, already-implemented pattern in the platform's Executive Tensions logic: where Processes & Collaboration scores materially stronger than Strategy & Alignment, or an Execution Index stronger than the Processes & Collaboration dimension feeding it, the platform surfaces the divergence as a named pattern rather than leaving it buried inside an aggregate. Neither pattern proves a specific boundary is the cause — consistent with everything this article has argued about the difference between a disclosed signal and a diagnosis — but each gives a leadership team a considerably more specific place to start looking than "things feel slower than they should."

It is equally important to state precisely what this measurement cannot do. Klarwerk does not measure process cycle time, does not calculate the financial cost of coordination friction in dollars, does not identify which specific handoff or which specific individual is responsible for a delay, and does not observe operational bottlenecks directly — it observes how people experience coordination, which is a genuine and useful signal, not a substitute for operational data the organization does not currently have. This is, precisely, the difference between detection and diagnosis this article has returned to throughout: the platform can detect that a pattern consistent with excess coordination cost exists at a specific boundary. It cannot diagnose why — whether the cause is a mismatch between process complexity and task complexity, an accumulation of controls no one has revisited, or a genuinely necessary response to real interdependence that simply feels costly. What the research in this article tells us is the general mechanism by which friction accumulates. What Klarwerk can observe is a structured, self-reported signal consistent with that mechanism being present. What would require additional operational data — process timing, workflow logs, direct measurement of handoff duration — remains a distinct and separate investigation the platform does not perform.

The following uses the same illustrative Meridian Logistics scenario referenced elsewhere in this Journal — a hypothetical, demo-based example, not a real customer, used only to make the mechanism concrete.

Picture Meridian's Execution Index registering in a strong range — priorities translating into daily work, deadlines generally holding — while its Processes & Collaboration score sits meaningfully below its other dimensions. Read separately, the first number is reassuring and the second is a moderate concern, the kind most organizations could point to on some dimension. Read together, the combination is more specific: an organization whose people are working hard and hitting deadlines despite, rather than because of, how work currently moves across its internal boundaries — consistent with the opening scenario's six-week request, where every individual step succeeded and the total system still took far longer than any one part would suggest.

The responsible next step is not to conclude that a specific team or process is at fault — the assessment does not, and structurally cannot, identify that. The responsible next step is the same kind of specific, falsifiable question this Journal's research has modeled throughout: where, specifically, does work most often stall or get duplicated between two particular teams, and is the coordination happening there proportionate to what it produces, or has it quietly grown past that point?

What would validate this reading, and what would challenge it, is worth stating explicitly, consistent with how this Journal has treated every illustrative scenario before this one. What would validate it: a direct conversation confirming that the same one or two boundaries account for most of the coordination effort teams describe, and that process complexity at those boundaries has not been revisited since the underlying work last changed. What would challenge it: discovering that the pattern traces to a single recent reorganization still settling into a new shape, or that the teams involved consider the current coordination genuinely proportionate to real, current risk. The score can only ever point toward the question. It was never going to be able to answer it.

The following questions are intended for a direct leadership conversation, each built to surface a different mechanism discussed in this article — attention and interdependence, the accountability-versus-control gap, process complexity outrunning task complexity, and the hidden labor concentrated at specific boundaries.

1. Where does work most often stall or get duplicated between two specific teams, and would either team, asked separately, even see the full cost of that stall?

2. Which of our current approval or review steps genuinely improve accountability, predictability, or shared understanding — and which have drifted into a cost no one has re-examined recently?

3. Where has a team quietly built a workaround to something the official process handles poorly — and is that workaround worth learning from, or is it generating rework we haven't traced back to its source?

4. If we measured not just whether our last three major initiatives hit their deadlines, but how much coordination effort each one consumed to get there, would the trend be improving or quietly worsening?

5. As this organization has grown, has our coordination mechanism at each major boundary kept pace with how interdependent that boundary has actually become?

None of the mechanisms this article has described point to any individual or team behaving badly. Every example in this article involved people making locally reasonable decisions, and the friction that resulted from combining those decisions was never any single person's fault to fix alone. That is not a reason for a leadership team to feel less urgency about it. It is the reason the fix cannot be delegated to exhortation — asking people to communicate better, or to follow the process more diligently, addresses a cause this article has specifically argued is usually not the real one.

The more useful and more genuinely actionable question a CEO can ask is not who is responsible for the friction, but where the organization's boundaries currently carry more coordination cost than the interdependence crossing them actually requires — and whether anyone currently has the vantage point, and the mandate, to notice when that gap widens. That is a solvable, investigable condition, not an indictment of anyone currently doing their job.

The advantage belongs to organizations willing to look, deliberately and on a repeated basis, at the seams between their own well-functioning parts — not because those parts are failing, but because the space between them is exactly where an otherwise well-run organization can quietly spend more than it realizes, for reasons no single team was ever positioned to see on its own.

This is also why the distinctions this article has developed — process complexity against task complexity, accountability against control, compliance against effectiveness — are worth more than their definitions alone. Each gives a leadership team a specific, answerable question to ask about a specific boundary, in place of the vague and largely unanswerable question "why does this feel harder than it should." A system that quietly makes good work harder than it needs to be is not a mystery requiring a diagnosis from outside. It is, on the evidence this article has reviewed, an ordinary and largely predictable consequence of how coordination mechanisms accumulate over time without anyone specifically tasked with revisiting them — which means it is also, on the same evidence, a genuinely solvable one, one boundary at a time.

There is a specific kind of relief available in seeing the mechanism this clearly, distinct from the vague frustration "we're too slow" usually produces. Vague frustration has no obvious next step beyond asking everyone to try harder, which rarely survives more than a quarter. A located, specific question — which boundary, which coordination mechanism, proportionate to what — has an obvious next step: a conversation with the people who actually experience that boundary, followed by a real decision about whether what's currently there is still earning its cost.