The Costs You Are Not Measuring Draft 12-11 MC

On the wall, the projector glowed with the familiar comfort of the P & L. Columns of numbers. Rows of cost centers. Footnotes that reassured the audit committee that the rules had been followed. The CEO looked at the first slide and felt something close to irritation. “We are not a disaster,” she thought. “Margins are still respectable. Cash generation is fine. The rating agencies are calm. Why does it feel like we are losing a race we cannot see.” The CFO watched her face. He knew what the board pack looked like. On paper, the business was under control. Revenue pressure, yes. Cost headwinds, yes. But nothing that screamed “crisis.” He had lived through real crises. This did not look like one. It felt worse. Those crises at least had a clear cause. This one felt like a slow fade that no line item could quite explain. The COO had flown all night from a plant that had spent the last six weeks in a state the P & L could never describe. Overtime, expedited freight, rework, heroic maintenance, vendors pulled in on weekends, supervisors burning out. The numbers for the quarter would be close enough. The human cost would not be visible. The structural cost was not even measured. The meeting began the way these meetings always begin. A walk through the financials. Revenue versus budget. Margin bridges. Capex versus plan. Working capital. No restatement. No scandal. No smoking gun. Then the CEO asked a question that cut through the pageantry. “If this P & L is right, why does it feel like we are falling behind. Where is the line that shows me the cost of being late. The cost of capital tied to the wrong work. The cost of people spending their days firefighting instead of changing the system.” There was a long silence. Eventually the CFO answered with the only honest sentence available. “It is not in here.” You know this room. If you sit on a board, you have watched this scene. If you are a CEO, you have felt that irritation. If you are a COO, you have lived inside the gap between the numbers and the lived reality. If you are a CFO, you have defended a P & L you know is accurate and incomplete at the same time. This article is about that gap. The causal models behind it. the dense maps of triggers, operational challenges, mitigants, and outcomes. tell the story in stark logic. At the center sits one node. the decline or stagnation of manufacturing productivity. Around it, you see the quiet decisions that pushed you there. Governance that adds month after month to every meaningful approval. Investments that

reinforce yesterday’s architecture. Layers of work that maintain the status quo instead of changing it. Cultural reflexes that reward heroics and punish structural honesty. The problem is not that your P & L is wrong. The problem is that it is blind. It was built for stewardship in a slower world. It was not designed to see the real cost of not moving.

The P & L was built for a different war

The P & L is a tribute to an earlier kind of leadership. It assumes the job of management is to deploy capital carefully, operate reliably, and demonstrate compliance. It answers three respectable questions. Did we make money. Where did we spend it. Did we follow the rules. Those are not bad questions. They are simply no longer sufficient. Complexity, connectivity, and computational power have changed the battlefield. The questions that now decide advantage are different. How much value did we destroy by moving slowly. How much capital did we quietly chain to yesterday’s operating model. How much human capacity did we burn on work that protects the past instead of building the future. The P & L has no place to put those answers. Delay does not show up as “delay expense.” Misallocation of capital does not appear as “bad bet.” Structural complexity is scattered across IT, overhead, logistics, and “business support.” Cultural drag is disguised as overtime, turnover, and “engagement initiatives.” Accounting standards are not the villain. They are doing what they were built to do. The villain is our refusal to admit that a statement designed for control at human speed cannot, by itself, guide strategy in an environment where small timing differences compound into structural advantage or structural loss. Lincoln spent nights in the War Department telegraph office because he understood that the speed and quality of information shaped the fate of the country. Churchill slept in the Cabinet War Rooms because he knew that minutes of delay in decision could be measured in lives. Today, the equivalents of those telegraph lines and war rooms run through your operations, your capital committees, and your digital stack. Yet the primary instrument you use to judge yourself was designed for a world where a month either way rarely changed history. That mismatch is why you keep feeling something is wrong even when the P & L looks acceptable.

The cost of delay, and why you cannot see it

Return to that room with the CEO, CFO, and COO. Imagine that instead of the standard P & L, the first slide had shown a simple question. “For the last three years, when did we know enough to act, and when did we actually act.” The gap between those two dates is the cost of delay. It is painfully real. Your accounts cannot name it. A throughput project sits in PowerPoint for a year while committees argue about scope, risk, and who will sponsor it. By the time shovels hit the ground, twelve months of contribution margin has quietly evaporated. The P & L records that year as “business as usual.” A chronic quality issue becomes part of the scenery. Everyone understands it. No one owns it. Meetings are held, task forces created, root cause slides polished. Actions drift. Meanwhile you pay out in warranty, concessions, and lost trust as customers re qualify competitors. The P & L calls this “COGS” and “sales discounts.” A new planning tool could shrink inventory and stabilize schedules, but it requires role changes and uncomfortable conversations with middle managers. The project slows. The company keeps paying for excess safety stock, premium freight, and overtime. The P & L calls it “logistics expense.” In the causal models, delay is not a single node. It is a bloodstream. You see it in high decision latency, in overloaded governance forums, in risk functions that treat every change as hazard, in legacy ERP that turns each modification into open heart surgery. Those choices accumulate into stagnation even when each meeting and gate looks prudent on its own. From a finance perspective, delay is seductive. It keeps the quarter tidy. There is no check labeled “lost option value.” There is only an opportunity quietly leaking away, one month at a time. Economically, delay is as real as a broken machine. A project worth twelve million dollars a year does not become less valuable because your capital committee took an extra year to approve it. You simply donated twelve million to your competitors and called it prudence.

Misallocation of capital. the quiet engine of decline

When people hear “misallocation of capital,” they picture spectacular failures. vanity headquarters, doomed acquisitions, grand automation projects that never work. Those exist. They are not the main story. Everyone notices them, and eventually you write them off. The more dangerous misallocation is quiet and cumulative. It is what happens when individually rational projects collectively reinforce the wrong architecture. Each proposal in isolation passes the spreadsheet tests. Together they anchor you deeper in a model that cannot win. Look at the triggers and controls in the causal map. You see a pattern.

Projects that add more SKUs, more local variants, more “flexibility” that is really just strategic indecision expressed in product form. Automation that locks in broken flows instead of simplifying them. Digital programs that build prettier dashboards while leaving the permission staircase unchanged, so decisions still climb the same slow ladder of escalation. On the P & L, this pattern looks respectable. Capex as a percent of sales sits in benchmark range. IT spend is “aligned with peers.” Overhead is “well managed.” The misallocation is architectural, not numerical. You are funding the comfort of interpretive visibility instead of the hard work of causal improvement. In behavioral terms, it is classic Thaler. You overweight what is easy to explain and underweight what actually changes long term payoff. A dashboard is easier to defend than an uncomfortable redesign of roles and decision rights. A new line for a new product variant is easier to argue for than the disciplined killing of low earning complexity. What you are really doing is placing a leveraged bet that tomorrow will look like yesterday, just with better reporting. In a world of rising complexity and agentic technology, that is not prudence. It is slow motion self harm.

The missing layer in your accounts. work on the work

Every company has three kinds of work, whether it names them or not. Work that touches the product or customer. Work that supports that work. Work that changes how all of that work works. The first two are familiar. Direct labor, materials, logistics, sales, service. Then planning, maintenance, quality, IT, HR, finance. These are the lines the P & L was designed to capture. The third category is the one that determines whether productivity compounds or erodes. Call it “work on the work.” Structural simplification. flow redesign. operating model architecture. cross functional problem solving. capability building. removal of complexity that no longer earns its keep. In the causal models, when work on the work is starved, the system begins to fray. Operational challenges multiply. Mitigants turn into band aids. Leaders rely on heroics to bridge gaps that should have been closed structurally. Noise rises. Fragility increases. Eventually productivity flatlines or falls. At that point, the P & L produces its verdict. Margins squeeze. Investors demand action. Leaders reach for the lever they have been trained to pull. They cut overhead. Which nearly always means cutting the very people whose job was to work on the work. That is how you end up in the loop we see in too many industrial enterprises.

Productivity stalls. The P & L tightens. Overhead is cut. Complexity and latency rise. Firefighting crowds out structural improvement. The cycle repeats until decline feels normal. From the outside it looks like rational cost control. From the inside it is liquidation of the only asset that can compound productivity across cycles.

The other costs your P & L cannot see

Delay, misallocation, and the erosion of work on the work are the main characters. The causal maps add several supporting actors that do just as much damage. There is the complexity tax. Every new variant, node, exception, and “special” policy adds drag. None of it appears as a separate line. It hides inside planning effort, changeovers, training, IT configuration, quality escapes, and inventory buffers. Each decision that adds complexity is defended on its own merits. The cumulative effect is a system that spends more time explaining itself to itself than serving the market. There is decision quality degradation. In overloaded systems, important decisions are either rushed at the top or delegated without context at the edge. The cost of poor judgement surfaces as under investment in resilience, a portfolio of products that never quite earn their cost of complexity, and supplier choices that trade short term price for long term fragility. The P & L records the result as “cost of goods sold” and “customer churn.” It cannot tell you that you are paying a tax on shallow thinking. There is erosion of learning rate. Enterprises that win now are those whose rate of learning exceeds the rate of change in their environment. That rate never appears in the accounts. You see its shadow in recurring issues, in plants or regions whose performance diverges instead of converging, in improvement programs that remain busy but not cumulative. You are funding motion, not progress. There is resilience and fragility. The P & L is a photograph. Resilience is a film. A fragile system can look efficient right up until the moment it breaks. The cost of that break is booked as a “one time event.” The structural fragility that made it inevitable is rarely measured. There is the opportunity cost of talent. In every company a small fraction of people carry a disproportionate share of real problem solving capability. They are either building the future or compensating for the weaknesses of the present. The P & L treats both as salary expense. The difference is invisible. And there is cultural drag. The habits and stories that define what can be said, what gets escalated, what is rewarded, and what is quietly punished. You feel culture when small problems are hidden until they explode, when complexity is defended as “customer centricity,” when

pointing at structural causes is labelled “not being a team player.” Culture is not a soft side topic. It is a multiplier on every arrow in the causal model. If you lead from the C suite, you live with the consequences of these costs even when you cannot see them. The first act of leadership is to stop pretending that what is visible in the P & L is the whole truth.

Reconstructing the P & L so you can see causality

You will not persuade your auditors to create accounts called “complexity tax” or “learning deficit.” Nor should you. The statutory P & L has a job. What you can do is build a management view of the business that overlays causal economics on top of traditional accounting. Three moves change the conversation. First. create three capital ledgers, not one. Every significant dollar of opex or capex should be tagged internally as run, improve flow, or build new advantage. Run is spend required to operate the current model safely and reliably. Improve flow is spend that reduces decision latency, simplifies architecture, or increases the capacity of people and systems to solve problems quickly at the edge. Build new advantage is spend that creates new value streams or step change capability. When you do this honestly, two patterns usually appear. You are under investing in improve flow. And a surprising amount of “transformation” spend is actually run spend with modern branding. It extends the life of the old model instead of changing its economics. Second. put timing on the same page as NPV. For each major initiative estimate the economic value of earlier realization. If the full benefit is twelve million per year, each month of acceleration is worth one million. Put that number in front of the capital committee. Track decision latency from first credible proposal to approval and implementation latency from approval to first measurable impact on operations. Do not let these metrics be an appendix. Put them on page one of every investment paper. When a project shows an attractive NPV but a history of long delays, force the discussion. Are we truly willing to pay the hidden tax of waiting another year. If not, what will we change in the permission staircase. Third. build a productivity bridge beneath the standard variance bridge. Finance is very good at explaining earnings in terms of price, volume, mix, and cost. It is less good at explaining how structural drivers created those variances. Each quarter, ask operating leaders to quantify how much of the change came from complexity, schedule stability, rework, talent churn, asset availability, and decision latency.

The first cuts will be rough. That is fine. The discipline matters more than the precision. Over time, you will refine the measures and build a much clearer view of how your choices are shaping performance.

Protecting and weaponizing work on the work

If you accept that work on the work is the real engine of productivity, then you must give it a different status. Treat it as a protected investment class, not a scattered overhead category. Bring continuous improvement, industrial engineering, operating model design, and digital architecture into a single portfolio. Fund it as an internal productivity fund with explicit rules. Any plant or function can propose initiatives that simplify flows, remove non earning complexity, or move decisions closer to the edge. The central team provides expertise and methodology. Savings are booked back to the operating units, but a portion is automatically reinvested into the fund so that the flywheel keeps turning. Report this portfolio explicitly to the board. Show not only immediate savings but also the effect on key causal indicators. decision latency, schedule stability, skill mix, asset availability, learning rate. Let investors see that you treat productivity not as a one time program but as a compounding asset. You will notice culture begin to shift. When people see that structural improvement is funded, protected, and celebrated, they take more risks in the right direction. Hidden complexity surfaces. Latent ideas are brought forward. The organization starts to behave less like a brittle hierarchy and more like an adaptive system.

From first generation control to second generation causality

There is a larger context sitting behind all of this. Most enterprises are still running what we would call first generation information architectures. They use data primarily to interpret the past. Dashboards to argue about what already happened. Governance to exert control over change. Second generation enterprises are quietly doing something different. They are designing their operating models around causality and permission. They treat the statutory P & L as the scoreboard, not the playbook. The playbook lives in the way they connect sensing to acting, capital to architecture, and authority to the edge. Agentic systems and automated reasoning will not save companies that refuse to face the structure of their own choices. In fact, these technologies will widen the gap. If you instrument a fragile, complexity heavy, permission bound organization, you do not automatically get productivity. You get clearer evidence of your own stagnation.

That is why this topic belongs in the boardroom. The transition from first generation to second generation AI is not primarily about models or vendors. It is about whether your architecture of capital, permission, and work on the work is ready to exploit technology that collapses latency and automates explanation. If it is not, you will end up with smarter dashboards sitting on top of the same slow staircase.

The decision in front of you

The most dangerous belief in any leadership team is that time is free as long as the quarter looks clean. The second most dangerous belief is that if something does not appear explicitly in the P & L, it is not your problem. The causal models say otherwise. They show that the decline of productivity is not an act of God, or a random walk, or the inevitable result of “labor shortages” and “supply chain issues.” It is the predictable outcome of thousands of small choices about capital, permission, complexity, and the value you place on work that changes work. So the question for CEOs, CFOs, and COOs is blunt. Are you willing to rebuild your view of performance so that the real costs of delay, misallocation, complexity, cultural drag, and neglected work on the work become visible, measurable, and actionable. Are you prepared to protect and grow the work on the work layer even when the quarter is tight. Will you attach a real price to time so that every month of decision latency is treated as a strategic choice, not an administrative footnote. If the answer is yes, then the P & L can return to its rightful place. Not as the oracle of truth, but as one instrument in a bigger, causal score. A score that tells you not only whether you made money this quarter, but whether you are building an enterprise whose productivity compounds, whose people are working on the right work, and whose capital is allocated to the architecture of the future rather than to ghosts of the past. If the answer is no, then you should not be surprised when, a few years from now, the numbers still look tidy, the margins a little thinner, the language a little more defensive, and the enterprise a little less able to shape its own destiny. The cost of not moving will already have been paid. You will simply be reading the receipt. References This argument draws on and adapts our prior Chief Architect Network and The COO Council work rather than citing it verbatim, including Enough Intelligence. Shaping Destiny Without Digital Gods, The Line Between First Generation AI and Second Generation AI, The Architecture of Conversion, The World’s Best COOs Think Differently, The Question Engine, and the internal One Degree for Everyone and Everything and Architecture of Permission white

papers, as well as ongoing Industrial Productivity Index and Decision Velocity research within LNS Research and the COO Council.