When an organisation chart is requested during a diligence process, the distance between the document's creation date and the date of the request is frequently a matter of days. The chart exists, it is tidy, the boxes are aligned; but the fact that it was produced within the same week indicates that it is not a record reflecting how the company works, so much as a depiction assembled in response to the request. In the same session, when three executives shown in three separate boxes are asked who decides a routine matter such as supplier selection, all three name the same person. This does not mean the chart was drawn incorrectly. It means the chart occupies a plane that is independent of where decisions are actually made, and that plane has its own internal consistency, which is precisely why nobody inside the company experiences it as a contradiction.
This pattern repeats across almost every mid-market and growth-stage company, and the reason it repeats is not negligence. In the founding period the volume of decisions is low, the decision-maker is singular, and the cost of writing authority down exceeds the benefit it delivers; everyone knows who decides what, and where they do not know, there is a single door to knock on. That configuration is rational for as long as transaction volume stays low, since the definitional work, the internal debate and the revision burden required to build a written authority matrix are, at that stage, a genuine cost carried against a thin benefit. The difficulty lies not in the shortcut itself but in the shortcut outliving the condition that produced it — the company triples in size, headcount broadens, geography diversifies, and yet the decision architecture remains calibrated to the scale of its first year.
At this point the chart takes on a second function, and this is where the real confusion originates. The structure on paper is no longer used to describe the organisation but to present the organisation outward — to a bank, a customer, a tender committee, a prospective investor. Two separate organisations consequently come into existence: the first is the documented representational structure, which appears layered, delegated and balanced; the second is the operating structure, which is flat, centralised and paced by one person's approval rhythm. The coexistence of the two disturbs nobody internally, because everyone inside knows which one is real and navigates accordingly, whereas the reviewing party is present in the room for the express purpose of measuring the distance between them.
At the diligence table the organisation chart is never read in isolation. It is cross-checked against the signature circular and bank mandates, against approval threshold tables, against the payroll and title register, against board and executive committee minutes, and against the approval chains embedded in procurement and hiring workflows. Every inconsistency among these five sources generates a separate finding line: a unit shown as reporting to the operations director whose budget approvals are in fact granted by the founder, a position that appears on the chart without a corresponding entry on the payroll, or a single individual standing at the top of both the production and the procurement lines. Such findings are not reported as technical errors but as governance observations, and the governance section of a diligence report is the section that bears most directly on how the transaction itself is structured.
The measurement dimension is where this subject is most often left underdeveloped, because organisational structure is intuitively treated as something that cannot be quantified. What is measurable, however, is not the boxes but the decisions: how many levels a purchase request above a defined threshold passes through, how many days a hiring decision travels between requisition and offer, at which level a customer complaint is closed and what share of complaints escalate one level upward, and, when a management position falls vacant, by whom and within what period the delegated authority is assumed. These indicators demonstrate whether the distribution of authority asserted by the chart is real, and they do so far more reliably than any management assertion could. Where a company keeps none of this data, any assessment of management quality necessarily rests on representation, and every line item resting on representation is priced conservatively.
What is sought in the ownership dimension is not who the chart belongs to but who is accountable for keeping it current, and at what cadence that accountability operates. In most companies the organisation chart is effectively unowned; it sits in a human resources file, yet because structural change decisions are not taken there, updates are triggered only when an external request arrives. The concrete consequence of that absence of ownership is straightforward: when a new layer is inserted or a reporting line is split in two, the corresponding adjustment to authority thresholds, signature powers and reporting lines lags by months, and throughout that lag the gap between the operating structure and the documented structure widens. As the gap widens, decisions falling into undefined territory migrate automatically to the most senior level available, so that centralisation emerges not as a management preference but as the natural settlement of a vacuum.
The continuity dimension is where the reviewing party is, in substance, asking a single question: can this company's present performance be reproduced without some portion of its present management? That question is answered not through the chart but through how the chart behaves under stress. The existence of a period during which the founder or a key executive was out of circulation for a sustained interval, a record of the speed at which decisions were taken during that period, and evidence that the delegation mechanism has actually been exercised at least once — where those three converge, the claim of independence may be treated as demonstrated. Absent all three, the claim remains a statement of intent, and statements of intent do not enter a valuation model.
The channel through which this deficiency reaches the balance sheet is direct and predictable. Where founder dependency is identified, the transaction is typically restructured through one of three mechanisms: migration of a portion of the consideration into an earn-out structure, a key-person undertaking requiring the founder and named executives to remain with the company for a defined period, or an expansion of the representations and warranties package under governance and internal control headings, accompanied by a higher escrow percentage. None of these presents itself as a price negotiation; each presents itself as risk allocation, and yet each moves the present value of the cash reaching the seller in the same direction. The cost of an undocumented authority architecture is more often concealed in the maturity of the amount not paid at closing than in the discount rate applied to the forecast.
Remediation in this area does not begin with drawing a better chart; it begins with recording, in a single table, which decisions sit with whom at which threshold. The structure we build carries two layers: in the first, decision types — procurement, hiring, pricing, contract execution, capital expenditure, customer discounting — are separated by monetary and qualitative thresholds, with each threshold bound to a defined level; in the second, that table is mapped one-to-one against the signature circular, bank mandates and the approval workflows configured in the underlying systems, since an authority matrix that lives only on paper reproduces exactly the problem the chart already created. When the mapping is complete, the resulting list of inconsistencies is usually longer than the company itself expected, and that list sets the first remediation agenda without further deliberation.
The second intervention binds the structure to a rhythm that keeps it current. The mechanism that converts an organisational change from a human resources transaction into a governance decision is a requirement that every structural change produce three outputs in the same sitting: the updated chart, the updated authority threshold line, and a single-paragraph decision record carrying the rationale for the change. Keeping that record at the moment of proposal rather than at the moment of approval renders retrospective justification impossible, which is what makes institutional memory auditable rather than merely archived. To this is added a short quarterly review posing only two questions: who in fact decided during the last quarter, and was that the person named in the table? Where the two answers diverge, what usually requires correction is not the practice but the table.
The delegation architecture is the complementary component of this structure, and it is generally built last, notwithstanding that it is among the earliest headings raised in diligence. Having a pre-designated deputy for each critical position is not sufficient; the deputy must have exercised that authority at least once in practice, and the exercise must have been recorded, failing which the delegation amounts to a list of names. Planned handover periods — documented evidence that the decision flow continued uninterrupted during the weeks a manager was on leave — constitute the cheapest and most persuasive proof of the continuity claim available to a company. The cost of producing that proof is close to nil, but producing it requires elapsed time, and it cannot be manufactured in the week the diligence begins.
An organisation chart is, in the end, a record of what a company knows about itself. What is measured is not the symmetry of the boxes but the rate at which the boxes and the decisions coincide; and that rate functions less as a description of who produced this year's result than as an estimate of who could produce the same result next year. What determines the valuation is precisely how narrow a range that estimate can be stated in.