In a sales meeting, the average discount granted during the final two weeks of a quarter sits materially above the average granted across the first six, even though customer profile, product mix, and competitive conditions have not shifted in any measurable way between the two periods. Asked to explain the quotes priced beneath list, the answers converge on a familiar set — competitor pressure, volume commitment, long-term relationship — yet none of these rationales is recorded in the same place as the quote itself. The pricing decision lives in a system; the reason for it lives in an e-mail thread, or nowhere at all. This separation is the first thing a sales organisation review surfaces, and it typically sits precisely where the company has never looked.
The second and sharper observation appears in the shape of the discount distribution. In an organisation whose approval threshold is set at fifteen per cent, granted discounts do not follow the smooth distribution arithmetic would predict; an unusual concentration forms between thirteen and fifteen per cent, thinning abruptly above the line. That shape indicates a sales force pricing against the condition of avoiding an approval request rather than against the condition presented by the customer. Instead of protecting margin, the threshold has become a centre of gravity that pins margin one notch beneath itself. Internally this is often read as evidence of discipline, since the volume of approval requests is low; a low request count, however, signals that the control is being routed around rather than exercised.
The mechanism operating underneath is the gap between authority and the cost of invoking it. Requesting a pricing approval is not, for a representative, the completion of a form; it is a social transaction carrying waiting time, the obligation to defend a rationale, and the possibility of refusal. So long as the cost of that transaction exceeds the personal cost of conceding a further two or three points, the representative predictably takes the second route. No failure of will is involved here — the conditions have been assembled such that protecting margin is individually expensive while surrendering it is individually cheap. Where the commission plan is calibrated on revenue rather than gross profit, the asymmetry sharpens further, since what the representative forgoes at the point of concession is not personal income but corporate margin.
A second layer of mechanism lies in the fact that pricing is less a negotiation than an anchoring exercise. List price, at the moment it is first shown, establishes the ceiling of the discussion, and every subsequent movement runs downward; the approval process governs the terminal point of that movement rather than its velocity. Designed as a gate that opens at the end of the quoting cycle, the mechanism no longer reviews a price — it formalises a commitment already given. The representative has stated the number, built the relationship on that number, and opened the approval request afterwards, with the consequence that a refusal now constitutes a customer relationship event rather than a pricing decision. The pressure the structure generates pushes the approver systematically toward assent, and the process, functioning on paper, filters nothing in practice.
The institutional cost of this surfaces first not in the income statement but in the explicability of the income statement. When gross margin moves several points between quarters and the movement cannot be decomposed into product mix, customer mix, and discount behaviour, margin ceases to be a management variable and becomes an outcome variable. In an investment review the inability to perform that decomposition is itself a finding, and one with an unforgiving implication: a company's capacity to forecast future margin cannot exceed its capacity to explain past margin. Forecast credibility fractures at exactly this point, and the fracture attaches not to the growth assumption in the business plan but to the assumption about the price at which that growth arrives.
The second cost channel is founder dependency, and it is the one that reaches valuation most directly. Where pricing decisions are rendered by the judgement of a single individual — usually the founder or the sales director — rather than by a written threshold structure, the knowledge that individual carries is recorded in memory rather than in a document: how far a given account can be flexed, where price is held deliberately low for strategic reasons, which promotional concession was granted as temporary and has since become permanent. To the extent that memory cannot be transferred, an acquirer cannot verify that what is being purchased will continue to operate at the same margin. In practice that unverifiability is priced in three places: a downward adjustment to the valuation multiple, the deferral of part of the consideration against post-closing margin performance, and a covenant requiring the key individual to remain for a defined period.
A third channel opens on the representation and warranty side. Price commitments embedded in long-term customer contracts, promises of tiered reductions, and most-favoured-customer style provisions, where they have not passed through a central approval process, sit scattered across the contract file and emerge one by one during diligence. Individually small, these undertakings aggregate into a disclosure that a portion of forward revenue is stripped of pricing flexibility. The acquirer responds either by requiring an escrow allocation or by demanding an indemnity covering undisclosed commitments, and both outcomes reduce the cash reaching the seller at closing. The absence of pricing approval is priced here not as a management weakness but as an obligation of uncertain perimeter.
Structural intervention proceeds through the redesign of decision architecture rather than through appeals to individual discipline, and it has four components. The first converts the approval threshold from a single percentage into a matrix of at least two dimensions: assessed jointly with contract term, or with payment terms, the discount percentage no longer permits the clustering behaviour that a one-dimensional line invites. The second is a sequencing rule requiring the approval request to be opened before a number is communicated to the customer; where approval becomes the precondition of the commitment rather than its ratification, the approver's capacity to refuse is restored. The third replaces free-text justification with selection from a closed category set, since narrative rationales do not aggregate into analysable data. The fourth distributes approval authority by transaction size and customer segment rather than by title.
The measurement layer is inseparable from this architecture, because an approval process that is not measured is quietly abandoned within a few quarters. The indicators that matter are behavioural rather than volumetric: the shape of the discount distribution around the threshold, the mean elapsed time from request to decision, the proportion of requests declined, and the category mix of stated deviation reasons. A refusal rate approaching zero is the earliest signal that the threshold is not functioning as a genuine control point, while a lengthening decision cycle forecasts that the process will be circumvented in the following quarter. Tracked together, these two indicators create the opportunity to intervene before the structure degrades.
Where BEIREK enters this area, the first thing constructed is not a new policy document but a decision record, in which the rationale stated at the moment of quotation, the identity of the approver, the elapsed decision time, and the final outcome are consolidated in a single place alongside the quote itself. Reconstructing that record six months backward produces, in most organisations, the first finding: a structural gap between the realised average discount and the level management had assumed. The threshold matrix is then recalibrated together with the commission plan, since an approval threshold cannot hold where a revenue-based incentive scheme has not been given a gross margin component — a structure that pushes the representative toward the threshold while requiring the approver to resist it is not sustainable.
The second line of intervention is rhythm. A monthly pricing review is established in which the subject is not individual transactions but the shape of the distribution and the weighting of reason categories, and whose output is not a list of decisions but a set of adjustments to the threshold matrix. On the ownership dimension, pricing approval is not removed from the sales function and transferred to finance — a transfer that lengthens cycle time and thereby kills the process in practice — but is instead defined as joint authority: sales fully empowered inside the threshold, deciding with finance outside it. Continuity is tested by whether an incoming sales manager, reading the threshold matrix and two quarters of deviation records during the first week, can understand the pricing logic without asking the founder; failure of that test means the process has not been institutionalised.
What a reviewer examining pricing approval is actually looking for is not the price itself but the company's capacity to explain its own margin. Every company has a price list, most have a threshold, and few can retrieve from a single record, six months later, why a particular discount was granted. Once that legibility exists, realised margin stops being a surprise and becomes a design outcome, and the buyer loses the ground on which a discount for unpredictability is claimed. The question worth putting is narrow: can the rationale for the largest discount granted last quarter be located today, without asking the person who granted it?