In the weekly planning meeting of a manufacturing site, the frequency with which a sequence confirmed on Monday is rewritten by Wednesday is a quantity that almost no operation measures. Reviewed across several quarters, the meeting minutes reveal a recurring pattern rather than a series of isolated incidents: the sequence fixed at the start of the week has been altered at least once by the third day, in busier periods twice or three times, and each alteration rests on a rationale that is entirely defensible when examined on its own — a significant customer has a line standing idle, a tender delivery date has been advanced, a quality escape has sent a batch back. Over the same span, the share of orders flagged as urgent within the total order book climbs from quarter to quarter while the demand arriving at the site does not grow at anything like that rate. The system, in other words, is generating urgency faster than urgency is being imposed on it from outside, and the source of that generation sits within the system itself.
The same pattern reads more plainly on the procurement side. Once a line item becomes date-critical, the supplier is called, the order is asked to be brought forward, and in the great majority of cases the supplier agrees. Agreement is possible not because additional capacity has been created but because the supplier has resequenced an existing queue; the job pulled forward has taken the place of another job. Where the supplier's customer portfolio is concentrated — the typical condition in industrial supply relationships, particularly for machined components, castings and custom electrical assemblies — the job pushed back is frequently another order belonging to the same buyer. The buyer thus finds itself, three weeks later, calling the same supplier to reclaim time it has taken from itself, and the tone of the second call is invariably harder than the tone of the first.
This loop is what is meant by the expediting spiral: an intervention made to pull one job forward disrupts the plan, generates a fresh requirement to expedite, and thereby feeds its own demand. Nothing at the origin of the spiral is irrational; on the contrary, intervening in the sequence at the moment of a genuine constraint is correct managerial behaviour, since the cost that a single missed delivery imposes on a customer relationship or through a contractual penalty typically exceeds the cost of disturbing the schedule. The difficulty lies not in the shortcut itself but in its persistence as the default operating mode long after the condition that justified it has disappeared. An exception repeated often enough becomes the rule, and an exception that has become the rule no longer manages anything as an exception.
At the core of the mechanism sits an arithmetic that is difficult to argue with. Where capacity is fixed over the relevant horizon, expediting manufactures no new time and only redistributes the time that exists; every hour pulled forward is an hour taken from another job. Layered onto that transfer is the cost of the resequencing itself — changeover duration, line setting losses, first-article approval, the material staging that must be performed a second time. Each intervention therefore deposits a net capacity loss alongside the delay it displaces, and the capacity lost lengthens the queue by a further increment. As the queue lengthens, the proportion of orders deliverable through the normal flow falls, and the falling proportion triggers a higher rate of intervention. That is the mathematics of the spiral, and it operates independently of anyone's intent.
A second layer forms as the planning signal degrades. Where expediting requests are reliably honoured, the delivery date entered into the system gradually ceases to represent the genuine requirement date and comes instead to represent the negotiating position of whoever entered it. Commercial teams learn to write dates somewhat earlier than needed, having observed that unadvanced dates drift toward the back of the queue; planning, knowing this behaviour, reads incoming dates at a discount; commercial, detecting the discount, advances the date once more. Once this reciprocal adjustment has completed itself over a handful of cycles, the data resting in the planning system is no longer a representation of real demand, and at that point the only mechanism capable of identifying which order is genuinely critical is the volume of the voice raising it.
The third layer is organisational and concerns the structure of incentives. Whoever initiates an expedite receives the outcome in visible form — the shipment has gone out, the customer has been calmed, the crisis has closed — while the cost is never posted to that person's account. The premium freight lands in the logistics budget, the overtime in the labour line, the changeover loss in the efficiency ratio, and the delay of the displaced order in another manager's delivery performance. The organisation thus rewards, systematically and without deliberation, a behaviour whose benefit is concentrated and whose cost is dispersed; for as long as that asymmetry holds, there is no structural reason to expect the frequency of expediting requests to decline of its own accord.
In the income statement, the loop appears nowhere under its own name. The freight differential created by emergency shipments dissolves into total logistics expense; overtime and supplementary shifts merge into personnel cost; the setting scrap and rework produced by resequencing remain inside cost of goods sold. Examined individually, none of these items looks unexplainable, which is precisely why no report ever aggregates them. Unless the costs are accumulated in a dedicated account with a defined trigger, the expediting spiral never surfaces at the board table as a cost question at all, because the accounting architecture the company has built for itself is configured, without anyone having intended it, to render the phenomenon invisible.
The trace left on working capital is considerably more diagnostic. In businesses where the spiral has matured, on-time delivery performance deteriorates while inventory levels rise in the same period; as planning loses credibility, every function constructs its own buffer, raw material safety stock expands, work in process accumulates between operations while awaiting sequence, and finished goods fills with items produced against the wrong priority. Inventory turns slowing and on-time delivery falling are, under ordinary conditions, indicators that move in opposite directions; a picture in which both deteriorate together points to the dissolution of internal sequencing discipline rather than to demand volatility. The resulting extension of the cash conversion cycle, even when it escalates into a discussion of credit facilities, is habitually framed as a financing question rather than as the operational question it is.
In a transaction, these items become a distinct surface of negotiation. Where a seller proposes to add expediting-related costs back to normalized earnings as non-recurring and extraordinary, the buyer's diligence team, upon establishing that the same items appear in every quarter of the trailing three years, will decline the adjustment; a cost that recurs is by definition not one-off, and the finding reduces directly the base against which the multiple is applied. The same evidence tends to settle, in the buyer's favour, the question of whether the working capital peg should be set on a normalized average or on a period-end snapshot, and where single-source dependency is identified on the supply side, it generates additional pressure on the scope of representations and warranties and on the escrow percentage.
The intervention that breaks the spiral is not a call for individual discipline but a system design resting on four separable components. The first is recording the expediting request at the moment it is raised rather than the moment it is approved, with the requester, the stated rationale and the job being displaced captured in the same record, so that the price of the intervention does not remain anonymous. The second is a frozen planning window combined with a deliberate narrowing of the authority to enter it; where that authority is unrestricted and free of consequence, request frequency will not fall. The third is accumulating expediting cost in a dedicated account and charging it back to the result of the requesting unit, so that benefit and cost finally meet in the same ledger. The fourth is elevating the expedite ratio — the share of total orders subjected to unplanned resequencing — into an indicator tracked on the same dashboard as delivery performance and inventory turns.
BEIREK typically begins this intervention not by rewriting the capacity plan but by constructing the record of the decision. The first mechanism established is an expediting log in which every request is captured as it is raised and the displaced job is a mandatory field; once that log has accumulated across an observation window of four to six weeks, the requests are classified by root cause — supplier slippage, internal quality loss, commercial commitment, forecast deviation — and the question of what is feeding the spiral ceases to be a matter of opinion. A weekly review rhythm then runs on top of the log, and the subject of that meeting is not which order should be pulled forward but which jobs last week's expedites displaced and whether those displaced jobs have returned this week as requests in their own right. Where the same record is mirrored on the supplier side, the expediting premium quietly embedded in a supplier's quoted price also becomes a negotiable line rather than an assumption.
The meaning of the mechanism differs by role, and the design has to respect that difference: for planning, the question is not the preservation of the schedule but whether the date entered into it continues to carry information; for the commercial function, it is whether the promise given to the customer is genuinely deliverable, which requires the capacity to declare urgency to be managed as a scarce resource rather than a free one; for procurement, it is the avoidance of competition among a buyer's own orders inside a single supplier queue; and for finance, it is rendering visible, as a named line, the source of the margin erosion that otherwise remains unexplained. What indicates the health of an operation is not how quickly it can intervene in a crisis, but how many weeks later the same crisis returns in a form the operation itself produced.