The seven questions executives ask before they know what to call the problem.

Because profit is an accounting outcome and cash is an operational one, and the gap between them is built by decisions no finance function ever sees. Profitability is reported after the fact; cash position is set by how long each function holds value before releasing it. A forecasting bias in the demand review becomes over-ordering in procurement, becomes safety stock layered across every facility, becomes a working capital line the CFO funds without ever seeing its origin. Nothing in that chain is an error. Every link is a defensible local decision, and the enterprise still funds the sum.

This is trapped working capital: cash held by operational behavior, policy friction and handoff delay rather than by an accounting mistake or a financing problem. It does not appear as a variance because no single function owns the chain that creates it. It appears as a credit line.

The Trapped Cash Analyzer quantifies it directly from your transactional records. On a $50M operating footprint the baseline diagnostic isolates $6.23M in addressable working capital across receivables terms, invoicing lag and inventory you did not decide to own. Seven days from raw ledger export to a board-ready recovery roadmap, with every dollar attributed to the function that is holding it.

By measuring margin twice instead of once. Most organizations measure it at the point of sale and never again on the way to the bank, which means every concession granted after the quote is invisible to the number the board reviews. Gross margin erosion is not one leak; it is a sequence of them, each granted by a different function against a target that function is not measured on.

The pattern repeats across industries. A volume discount is granted against lot sizes the plant will never run, converting a commercial courtesy into permanent unit-cost erosion. A delivery date committed to close the quarter forces partial shipments and spot-market expedites, and the deal books at standard margin while settling at pocket margin. Rework, returns and warranty claims travel backward through the P&L as credits scattered across four cost centers, so no single owner ever sees the total.

The Margin Leakage Analyzer runs the pocket-margin walk end to end. On a $210M footprint at a 35% target it traces $73.50M of quoted potential down to $54.97M actually realized, names every leak in between, and separates the $10.01M that is recoverable from the $8.52M that is the cost of operating. The distinction matters more than the total: it is what stops a recovery program from attacking spend that is doing real work.

Because the enterprise is organized to optimize vertically and value moves horizontally. Each function is measured on its own utilization, its own service level, its own cost per unit, and each one can hit its target while the handoff between them loses days. Local optimization is not a failure of discipline. It is what a well-run silo does, and it is precisely why the constraint survives every departmental review.

The mechanics are unglamorous. A cost-reduction mandate transmitted as a spreadsheet becomes local batch-size decisions that protect each department's utilization metric and starve the enterprise constraint the mandate was meant to relieve. Finished goods that beat their production schedule then queue three days for waves, slots and tenders. Throughput is won on the floor and surrendered in the yard, invisible to OEE and fatal to OTIF.

The Flow Constraint Analyzer measures the system rather than the assets. On a $210M capacity footprint the throughput walk traces 100% of available capacity down to the 64.2% that physically ships, because three zone yields of 84.6%, 84.1% and 90.3% multiply rather than average. It then separates the $37.24M that is recoverable from the $37.89M that is the cost of operating a real plant.

By pointing it at the relationships between functions rather than at the records inside them. Most enterprise AI programs stall because they are aimed at a single function's data and therefore can only improve what that function already controls. The loss lives in the white space between functions, where a decision made in demand planning becomes a cost carried in treasury, and no dashboard in either function contains both halves.

The Enterprise Reasoning System is decades of operating judgment encoded into multi-agent reasoning that reads your own transactional records against a causal model of how one function's decision becomes another function's loss. It does not report what a table said last month. It reconstructs how demand influences supply, supply affects production, production affects inventory, inventory affects fulfillment, and fulfillment affects cash, then translates each interaction into quantified financial impact.

That is what makes the output an EBIT number rather than an insight. One reasoning core, three levers: cash, profit, throughput. Identified in days, recovered through focused execution, and held afterward by an optimizer that re-runs the same walk against live data. Programs built this way typically move EBITDA by two to eight points, funded from the capital the first diagnostic released.

You do not align them with a workshop. You align them with a figure none of them can dispute and a business case that survives a capital review. The greatest barrier to enterprise transformation is not analysis and it is not technology; it is that a finding everyone accepts still has no owner, no funding line and no cleared path through the allocation committee.

The Alignment Mandate is the two-to-four-week intensive that converts an accepted finding into an authorized program. It works three axes at once: vertically to the capital allocators, the CFO and the board who must clear it against competing calls on the same capital; horizontally across the division heads whose functions each own a fragment of the value chain and none of the whole, which is exactly why the leak persisted; and technically down to engineering and IT security, where execution either survives contact with the existing estate or quietly does not.

It is invoked only where it is needed, when the organization has accepted the number and still cannot move against it. It is funded from what the diagnostic already recovered, which removes the net-new capital allocation that most alignment work never gets approved for. The output is a board-cleared, self-funded execution roadmap with a named owner in every function that has to move.

It works because nothing is installed and nothing is integrated. The diagnostic runs against a read-only extract of transactional records your systems already produce, from the ERP, WMS, CRM and general ledger your teams work from every day. There is no connector to build, no schema to agree, no production footprint and no change to any system of record.

Ingestion operates inside your network boundary. Your security lead is asked for read access to data that already exists, not for a deployment, and the questions that lead will ask are answered in a working session rather than a datasheet. The practical demand on your IT function is measured in hours of export, not in a project on next year's roadmap.

The seven days are fixed scope. What used to take a discovery team three months of interviews is a reasoning system reading the records at machine speed against an encoded causal model, which is why the timeline compresses without the rigor doing the same. You receive a definitive walk: every leak named, sized and attributed to the owning function, in a form your finance committee can interrogate line by line. A comparable Big Four diagnostic runs eight to twelve weeks with four to six consultants on site.

We execute. The analyzer is the entry point, not the product. Our principals are operators who ran supply chains, plants and P&Ls before they encoded that judgment into the instruments, and they stand beside your teams through the recovery rather than handing over a document addressed to them.

There are two routes, and which one applies depends on whether you already know what is wrong. Recover starts with the analyzer: the number in seven days, a practice executing against it, and an optimizer holding the baseline afterward. Days to the first figure, weeks to the first recovery. Transform starts where the direction is already known and the organization is the constraint: a Readiness Assessment or Alignment Mandate establishes capability and commitment, then a practice rebuilds the capability itself across business model, revenue engineering, or operations from manufacturing through quote-to-cash. Weeks to a mandate, quarters to a rebuilt capability.

Both routes end at the same place. A recovery no one measures afterward is a recovery you will run again in eighteen months, so the instrument that produced the number is the instrument that watches it: the same walk, the same segments, the same attribution, refreshed on your cadence against the baseline you recovered to.