A recurring pattern surfaces in weekly cash meetings: when the question of how much should sit in the account is put on the table, three people produce three different figures, and each of the three can defend the figure with a reasonable rationale. The finance manager anchors on the week in which payroll and tax obligations converge, procurement points to the period where supplier maturities cluster, and the founder speaks from the residue left by a collection delay that occurred once, several years earlier. The spread between these figures is rarely trivial; it can widen to the equivalent of a full month of operating expenditure. The meeting nonetheless moves forward without resolving the gap, because the balance is sufficient that week, and for as long as it remains sufficient the question itself never enters the agenda.
The same question arrives from outside, in a considerably sharper form, during the opening weeks of an investment or sale process: how much of the cash on the balance sheet is required for the business to operate, and how much is surplus? It is not a neutral request for information. In a transaction structured on cash-free debt-free logic, the portion deemed necessary transfers with the company while the portion deemed excess remains with the seller. Confronted with a question the company has never posed to itself, the counterparty generates an answer from its own assumptions, and an answer generated in the absence of a definition is predictably calibrated against the company — bearing the cost of ambiguity being no part of the reviewing party's mandate.
That the definition is never built is explained not by neglect but by the shortcut having genuinely worked for a period. In a company whose cash balance has never approached a critical threshold, calculating a minimum cash requirement carries no visible payoff; the exercise consumes time, demands data, invites argument, and yields a number whose value becomes apparent only once something goes wrong. What substitutes for it is the average of historical balances: having operated within a certain band for some stretch of time, the company quietly converts that band into a norm, and the norm then functions as its own justification. The problem lies not in the shortcut but in its persistence after the conditions change; once revenue doubles, supplier terms tighten, or customer concentration increases, the historical band no longer carries the same risk.
On the measurement side, the most frequently observed structural error is assessing the minimum cash requirement against the month-end balance. Month-end is the date on which the accounting period closes and the bulk of collections has landed, producing a snapshot close to the highest balance of the month. The company's actual trough, by contrast, typically forms inside a narrow window where payroll, social security obligations, tax payments and supplier maturities fall within days of one another — a window that never appears in monthly reporting. A minimum cash assumption built on a twelve-month average balance therefore comes out systematically low, and the distance between that assumption and operational reality becomes visible only when a collection delay happens to coincide with the window.
The second structural error is the assumption that an undrawn credit line can stand in for a cash buffer. The proposition heard verbally is that the balance can be kept thin because the revolver can be drawn if needed; yet even a committed facility is a conditional source. Drawdown is conditioned, under most credit documentation, on representations and warranties remaining true as of the drawdown date, on covenant tests being satisfied, and on the absence of a material adverse change — which is to say that the conditions under which the facility is most needed and the conditions under which access narrows are the same conditions. The reviewing party understands this linkage and does not treat the commitment as cash-equivalent. The absence of that distinction inside the company indicates that liquidity management rests on an expectation about a banking relationship rather than on a written policy.
The first and most direct channel through which the gap reaches valuation is the setting of the working capital reference level. Where the minimum cash requirement is documented, justified and supported by daily data, the negotiation proceeds against a defined figure; where it is not documented, the buyer anchors on the highest observed necessary balance and pulls the reference level upward. Every unit of that adjustment materialises as cash that does not reach the seller at closing. This rarely presents itself as a discount discussion or a multiple discussion; it emerges instead within a technical line of the closing calculation, at the last and most exhausted stage of the negotiation, at which point producing the data that would move the counterparty's assumption is no longer feasible.
The second channel sits on the debt side. In a credit structure carrying a minimum liquidity covenant, where the relationship between the point of testing — period-end or throughout the period — and the location of the company's own cash trough has never been established, technical breach risk arises for reasons entirely unrelated to operating performance. A waiver request, even where it produces no separately measurable cost, permanently alters the lender's perception of management quality and is priced into subsequent negotiations. By the same mechanism, in the absence of a minimum cash definition, revolver utilisation is kept wider than necessary, with commitment fees and interest paid continuously against a buffer that is never actually required.
The third channel lies where the ownership and continuity dimensions intersect. In many companies the minimum cash requirement exists in practice — the payment sequence is known, which supplier can wait and which cannot is known, the critical week is anticipated in advance — but that knowledge resides in an individual's experience rather than in a document. When the reviewing party identifies this, the finding is recorded not as liquidity risk but as key-person dependency, and key-person dependency reaches the transaction as a structural condition rather than as a headline discount. Post-closing retention commitments, an earn-out spread across a longer horizon, an elevated escrow percentage, or a request for specific approval rights over treasury decisions are the characteristic responses to such a finding. Each of them pushes the seller's access to proceeds further out in time.
Building the structure is a matter not of settling on a single figure but of decomposing the requirement into separable components, each of which must be independently defensible. Four components carry the load: an operational buffer covering the timing gap between collection and disbursement in the ordinary course; a contractual and regulatory buffer imposed by credit agreements, letters of guarantee, public procurement obligations or licence conditions; a seasonal buffer absorbing the additional burden of the period in which demand or supplier payments concentrate; and an event buffer sized against defined exposures such as a delay by a single large customer or an advance-payment demand from a sole-source supplier. A threshold broken into components behaves differently in negotiation than a single aggregate figure, since the counterparty is then obliged to contest the specific component it finds unjustified rather than the total.
BEIREK's intervention in this area begins by moving measurement onto the correct surface. In place of monthly balance reporting, we construct a daily trough series derived from bank movements, showing for each month which day carried the lowest consolidated balance, which payment item drove it, and which collection expectation was outstanding against it. On top of that series sits a thirteen-week cash projection, refreshed weekly and recording the variance between forecast and actual on a retrospective basis; the variance record itself produces an evidentiary chain demonstrating forecasting accuracy, which carries separate value during a review process. The minimum cash policy is then built above these two data layers as a written document that justifies each of the four components individually and is submitted for board or shareholder approval.
The second layer is governance, since a threshold that is measured but unowned collapses on the implementation dimension. Ownership of the threshold is attached to a single role, and that role's decision authority and its limits are written down: at what balance level who is notified, at which threshold the payment sequence changes and under what rule, and in which circumstances a drawdown proceeds and on whose approval. What demonstrates that these rules operate is not assertion but record; a short decision note kept each time the threshold is approached shows which decision was taken on which rationale, and the accumulation of such notes becomes an institutional capacity that survives the departure of the individual concerned. This is precisely the evidence the continuity dimension of a review is designed to locate: the threshold existing as a procedure the company can reproduce rather than as one person's instinct.
Cash sitting in a company's account is, in accounting terms, a single line item; in negotiation terms it is two distinct assets, and the identity of whoever draws the line between them largely determines where that line falls. Where the company has not drawn it — with its own data, its own reasoning and its own approval mechanism — it will be drawn across the review table by the other side.