The argument that surfaces on a shipping day in a manufacturing plant takes very nearly the same form in every company: a measurement within the batch sits at the edge of tolerance, the customer's delivery window closes that week, and the question on the table is not technical but jurisdictional — who has the authority to hold this batch. The objection raised by the quality function is generally recorded as an opinion, while the delivery pressure raised by production planning is recorded as a commitment; the two do not carry equal weight, and the decision defaults to the instinct of the most senior person in the room. On the wall of that same plant hangs a valid quality management system certificate, the procedure file is current, and the internal audit calendar is running. Even so, when asked who decided on that particular batch, against what threshold, and on what record, the answer resolves to a person rather than to a role.
The same pattern repeats in service businesses, differing only in surface. Who sets the acceptance criterion for a deliverable report, a software release or a design package, and who performs the final check before it leaves, varies from engagement to engagement. On one job the senior specialist signs off on their own output; on another the project manager scans it before delivery; on a third the client's first round of feedback functions as the de facto quality control. Companies describe this as flexibility, and the justification is frequently sound, since acceptance thresholds genuinely differ across engagements. The structural consequence, however, is constant: quality settles into the one domain for which everyone is held responsible and, precisely for that reason, no one is accountable.
The mechanism underneath this settlement is the dilution of accountability through distributed responsibility. Assigning a task to several roles at once appears at first to increase assurance; in practice each role, assuming that another is also watching, lowers its own control intensity, and the aggregate control ends up weaker than it would have been under a single owner. Layered on top of this is the asymmetric feedback structure of the quality decision itself: the cost of a stop decision materialises immediately, visibly and attributably to one individual — a missed delivery date, an idle line, an irritated customer — whereas the cost of a release decision surfaces months later, dispersed across functions, and usually inside a different budget line altogether. Producing a bias toward release under that asymmetry is not a weakness of the decision maker; it is the predictable output of the decision architecture.
The absence of quality ownership should therefore be read as a problem of authority design rather than one of discipline. Ownership here does not mean the right to write the standard; writing standards, updating procedures and issuing audit reports are activities most companies already perform competently. The distinguishing component of ownership is which role holds the final word at the moment production or delivery pressure collides with the quality threshold, and by whom that word can be overridden. A quality manager without stop authority, however high the position sits on the organisation chart, functions as a reporter rather than an owner; the documents produced are not the record of a decision but the justification of one taken elsewhere.
Documentation is routinely mispositioned within this picture. At the diligence table a valid system certificate constitutes weak presumptive evidence of quality infrastructure; the certificate demonstrates that a system was designed, not that it operates. The record with genuine verification power is not the certificate but the closed chain in which nonconformance entries connect to root cause analysis, corrective action, and verification of that action's effectiveness. Where the ends of that chain remain open — a nonconformance raised but no root cause recorded, a corrective action defined but never verified at closure — the distance between the document file and actual practice becomes measurable, and the reviewing party looks at precisely that distance.
The first surface on which the institutional cost appears is usually the margin line rather than the quality line. Rework hours land in production labour, scrap and yield loss in material cost, expedited freight used to recover a slipped date in logistics expense, and the free-of-charge revisions granted after a customer complaint in after-sales service or directly in project cost. This dispersion does not erase the cost of quality from the accounts; it renders it invisible, with the result that a persistent erosion of several points in gross margin is interpreted by management as pricing pressure or raw material inflation. When these items are separated out during diligence, the resulting figure is frequently of an order the company has never seen in its own internal reporting.
The second surface is transaction structure itself. Where the quality decision is found to rest with a single individual — most often the founder or a founding partner — the finding is recorded not as technical risk but as founder dependency, and founder dependency reshapes deal architecture in addition to flowing through as a valuation discount. Transfer of quality management as a condition precedent to closing, expanded representations in the warranty package under product conformity and recall headings, an increased escrow percentage, earn-out triggers conditioned on the founder's retention through a transition period — each of these is the contractual conversion of uncertainty arising from an ownership gap that could not be evidenced. The buyer in this position tends to prefer leaving the risk with the seller over reducing the headline price.
The third surface is measurement, and it is generally the weakest link. A substantial share of companies track the number of customer complaints, but that count carries no managerial information on its own; what matters is at which process step the complaint originated, how the first-pass yield distributes across lines or teams, and whether the stage at which nonconformance is detected has been drifting over time toward the customer or back toward production. When quality indicators are not wired into the operational decision rhythm — that is, when they do not appear on the weekly production meeting agenda with the same weight as schedule and shipment figures — measurement ceases to be a management instrument and becomes a reporting obligation, and that conversion is readily detected during review.
Structural intervention begins not with individual awareness but with the relocation of authority, and it separates into four components. The first is the definition of stop authority in a single role, in writing and together with its threshold values. The second is explicit regulation of by which body and on what record that authority may be overridden — that is, of deviation approval. The third is the accumulation of deviation approvals in a dedicated register brought to the management agenda at a defined frequency. The fourth is the anchoring of quality indicators to the operational meeting rhythm, at the same table where resource and schedule decisions are taken, rather than in a separate quality meeting. Established together, these four components move the quality decision out of the domain of personal courage and into that of repeatable institutional behaviour.
BEIREK's intervention in this area begins not with procedure drafting but with the positioning of the decision record. The structure we install captures the quality decision at the moment of proposal rather than only at the moment of approval: when a nonconformance is opened, the technical rationale, the commercial pressure and the proposed action stand side by side in the same entry, and the final decision is written on top of that entry, with the effect that the reasoning behind the decision cannot be reconstructed afterwards. To this is added a separate register for deviation approvals together with a rhythm that carries that register to the management table at fixed intervals; the accumulation itself makes visible a pattern that no individual decision reveals on its own.
The second line of intervention addresses the continuity dimension of ownership. A backup and handover protocol is defined for the role carrying quality authority — recording not merely who holds it, but who holds it in that person's absence and on what record the transfer occurs; separately, the owner of the quality indicators is deliberately decoupled from the owner of the production or delivery indicators, since two indicators concentrated in one role resolve predictably in favour of the schedule when they conflict. The equivalent of this arrangement in the diligence process is concrete: the investor encounters a chain of records demonstrating that quality is produced by the mechanics of the company rather than by the attentiveness of one individual.
A company's quality performance is rarely assessed on its own terms in diligence; what is actually examined is the set of conditions under which that performance was produced and whether it remains reproducible once those conditions change — when volume doubles, when the founding partner steps back, when a new customer segment imposes a different acceptance threshold. The weight of the quality ownership question comes from exactly this: what is being asked is not whether the product is good, but whether it can be shown where and by whom the decision that keeps it good is taken. The distance between those two questions is, in most transactions, the terrain on which the valuation negotiation is actually conducted.