--- title: "The Capital Expenditure Plan: The Gap Between the Number on Paper and the Asset on the Floor" description: "A capital expenditure plan is a multi-year decision record in which items are separated into maintenance, compliance, capacity and discretionary categories, each tied to an asset-level condition assessment, with plan-versus-actual variance measured on a regular cycle. Absent that separation, an investor will typically assume maintenance capex at or above depreciation, and that single assumption lowers the valuation base directly." url: https://www.beirek.com/en/blog/capital-expenditure-plan-due-diligence canonical: https://www.beirek.com/en/blog/capital-expenditure-plan-due-diligence published: 2026-05-18 modified: 2026-05-18 category: "Cash, Working Capital & Funding" category_url: https://www.beirek.com/en/blog/category/cash-working-capital-funding language: en-US reading_time_minutes: 8 publisher: BEIREK LLC publisher_url: https://www.beirek.com license: "© BEIREK LLC — citation with attribution and link permitted" keywords: ["capital expenditure plan","maintenance capex versus growth capex","deferred maintenance valuation impact","investment readiness diligence","free cash flow normalisation"] topics: ["Capital expenditure planning and governance","Investment readiness and valuation diligence","Asset condition assessment and remaining useful life","Covenant calibration and drawdown scheduling"] alternate_language_url: https://www.beirek.com/tr/blog/capital-expenditure-plan-due-diligence --- # The Capital Expenditure Plan: The Gap Between the Number on Paper and the Asset on the Floor > **In short:** A capital expenditure plan is a multi-year decision record in which items are separated into maintenance, compliance, capacity and discretionary categories, each tied to an asset-level condition assessment, with plan-versus-actual variance measured on a regular cycle. Absent that separation, an investor will typically assume maintenance capex at or above depreciation, and that single assumption lowers the valuation base directly. *In most companies the capital expenditure plan is not a document but a sequence carried in the minds of two or three people. When the diligence table asks to see that sequence, the resulting gap is priced not as a modelling assumption but as a permanent reduction in the base of free cash flow.* --- There is a moment that recurs at the diligence table. In a five-year projection the capital expenditure line sits either as a percentage of revenue or as a round figure repeated identically across each forecast year, and when the question is asked which assets that line actually corresponds to, the answer arrives not from a document but from a person in the room. Usually the founder or the technical director explains without hesitation which line will come up for overhaul and when, which equipment has how many years left in it, which investment would be pulled forward if a particular customer commitment materialised. The account is coherent, and in most cases it is also accurate. What it is not, anywhere, is written down, and therefore it is not capable of verification. The question the company has never put to itself is a narrow one: whether this sequence remains the same sequence when the person carrying it is not in the room. The second observation is quieter than the first. Fixed asset records are maintained properly, invoices are in order, the capitalisation threshold is defined and applied consistently; what does not exist is a single schedule placing last year’s planned investment alongside last year’s actual investment. Because variance is never computed, nothing enters institutional memory about which item was deferred, who decided to defer it and on what technical grounds, or what the deferral produced downstream in the operation. The company executes capex without managing capex, and the distance between those two verbs becomes visible in the first week of an investment review. The mechanism beneath this gap is not negligence; it is a shortcut that remains entirely functional under a specific set of conditions. In a growing company with a single decision centre, a formalised capital expenditure plan slows opportunistic procurement, and the ability to commit within the week that a suitable secondhand unit becomes available at an attractive price is a genuine advantage. So long as that advantage exceeds the cost of the plan, the shortcut is rational. The difficulty lies not in the shortcut itself but in its persistence once scale, credit structure and the number of stakeholders have changed. When a company begins drawing on a construction loan or carrying a covenant package for the first time, the predictability of the capex decision starts to be priced more heavily than its speed, while the internal reflex remains calibrated to speed. A second mechanism originates in a conceptual conflation: the capex plan is treated as though it were the budget. A budget is an annual authorisation instrument and by its nature anchors on the prior year’s figure; a capital expenditure plan is a multi-year commitment map and ought, by its nature, to anchor on the physical condition of the asset, its remaining life and the operational consequence of failure. Where the two substitute for one another, the plan becomes indexed, without anyone intending it, to the depreciation schedule — and depreciation is an accounting allocation convention, not a maintenance calendar. That indexation tends to underprovide on fast-wearing lines and overprovide on long-lived assets, and both errors accumulate in the same direction year after year. The third mechanism sits in ownership. The capital expenditure plan stands at the intersection of two calendars, the calendar of technical necessity and the calendar of cash, and those calendars typically belong to two different functions. The operating side knows what is required and when, but does not see the financing window; the finance side knows the cash timetable, but cannot form a view on how much service life a compressor has left. Absent an established rhythm of reconciliation between them, the plan becomes a document that neither function fully owns and therefore neither function updates. In unowned areas, implementation gaps and slippage appear on a predictable schedule. The institutional cost surfaces first in the base of free cash flow. Where maintenance capex cannot be distinguished from capacity capex, the investor is left with no route other than a conservative assumption, and judging by the typical construction of technical diligence reports, that assumption places maintenance capex at or above the depreciation level, precisely because no evidentiary chain has been offered to support anything lower. That single assumption depresses normalised cash flow in every forecast year and permanently narrows the base to which the multiple is applied. The loss here is not of the kind recoverable in a further negotiating round; the assumption has entered the architecture of the valuation model. The second channel is deferred maintenance, and it is the more expensive of the two. Postponing an overhaul by a year presents as a strong margin in that year’s income statement while resting on no balance sheet line whatsoever; it is a borrowed margin, and the debt comes due at the first technical inspection. The consequence is not confined to the quantum of the finding, since for the buy side the finding immediately generates a further question: what other undocumented commitments exist. From that point the scope of representations and warranties widens, the escrow proportion rises, an asset condition survey is added to the conditions precedent list, and the timetable of the process can extend by something approaching a full budget cycle. The third channel lies on the financing side. Given that credit committees assess the capex profile through debt service coverage and the drawdown schedule, an investment plan whose distribution across years remains indeterminate is met with additional contingency and tighter covenant calibration, part of the cost landing in margin and part of it in flexibility. The fourth channel emerges at the intersection with working capital: unplanned capex is, more often than not, funded out of supplier payment terms, and while that choice operates initially as an invisible solution, it degrades payment terms and purchasing conditions over time. In diligence that trace is read not in the capex line but in the supplier ageing schedule. The mechanism that neutralises this tendency is not individual discipline but plan architecture, and it separates into four components. The first is classification: every item is written into one of four classes — maintenance, mandatory compliance, capacity and discretionary — and each class carries a different approval threshold. The second is the evidentiary chain: the item is tied to a specific asset in the fixed asset register, to a condition assessment of that asset, to a remaining-life estimate, and to the operational consequence that would follow a one-year deferral. The third is that the decision record is kept at the moment of proposal rather than at the moment of approval — which alternatives were considered, why this amount, what the cost of deferral was estimated to be. The fourth is variance measurement, and here timing variance matters considerably more than amount variance, since slippage in timing is the earliest available indication that the technical calendar and the cash calendar have not been bound to one another. BEIREK’s intervention in this area typically begins not with drafting a document but with establishing a record and a rhythm. An asset-level capex register is constructed in which each item carries the proposer, the approver, the technical rationale, the cash window and the deferral scenario as separate fields, with the effect that proposal authority and approval authority can no longer accumulate in the same individual. A monthly reconciliation rhythm is then operated: committed, planned and expended amounts are compared as three distinct columns, and the rationale for any variance is recorded within the same session. For material items, a pre-mortem is run before the decision is taken — the investment is assumed to have been misspecified eighteen months on, the reasons are written out in advance, and that text remains on file. The second line of intervention aims at making the plan reproducible independently of any individual. The multi-year horizon is refreshed on a fixed rolling cycle, so that the plan does not reset when a person departs; because classification rules, thresholds and evidentiary requirements are written, an incoming technical director can reproduce comparable decisions on comparable grounds. The practical test in application is unadorned and is grasped quickly at the negotiating table: whether the capex plan the company uses internally and the capex plan presented to the investor are the same document. If they are, the plan constitutes an institutional capability; if two separate documents are being produced, the presented document is a narrative and will be treated as such in review. The value of a capital expenditure plan does not derive from the accuracy of its forecast; no multi-year capex projection closes without variance, and no investor expects one to. Value derives from the demonstrable fact that the same decision can be reproduced, on the same reasoning and in the same order, without the founder in the room. Where a company’s capex register, variance schedule and decision record evidence that reproducibility, future investment requirement ceases to be a source of uncertainty and becomes a parameter capable of being priced — and any parameter capable of being priced is, by definition, cheaper than a conservative assumption. ## Key Points - Where maintenance capex cannot be separated from discretionary capex, the investor builds free cash flow on the most conservative available assumption, and that assumption hardens into a permanent valuation adjustment rather than a negotiable line item. - Deferred maintenance presents as a strong margin in the income statement while appearing nowhere on the balance sheet; once surfaced in technical diligence it affects not only price but the perceived reliability of every other representation. - Timing variance in a capex plan is considerably more informative than amount variance, because slippage in timing indicates that the technical calendar and the cash calendar have never been reconciled to one another. - When the authority to propose and the authority to approve sit with the same individual, the capex plan becomes the most visible evidence of founder dependence and enlarges post-closing control negotiations. - The practical test of whether a capex plan is an institutional capability is simple: whether the document used internally and the document presented to the investor are the same document. ## Questions ### What distinguishes a capital expenditure plan from an annual budget? A budget is an annual authorisation instrument and typically anchors on the prior year’s figure; a capital expenditure plan is a multi-year commitment map and is expected to anchor on asset condition, remaining useful life and the consequence of deferral. Where the two are used interchangeably, the plan becomes indexed to the depreciation schedule without anyone intending it, and that indexation systematically underprovides on fast-wearing lines. ### What exactly is being looked for in a capex plan during investment review? Not forecast accuracy, but the structure of the decision: whether items are separated into maintenance, mandatory compliance, capacity and discretionary classes; whether each item ties to a specific asset in the fixed asset register and to a condition assessment; whether proposal and approval authority are held separately; and whether plan-versus-actual variance is measured regularly on timing as well as on amount. Where that structure exists, future capex becomes a priceable parameter. ### How is maintenance capex separated from growth capex in practice? The separation is established by evidence rather than intent. An item required to preserve existing capacity and the existing safety and compliance position is written to maintenance; an item creating measurable additional capacity or a new revenue line is written to growth. The basis for that classification is asset-level condition assessment and remaining-life estimation. Absent such a basis, an investor will typically assume maintenance capex at or above the depreciation level. ### Through which channel does deferred maintenance affect company valuation? A postponed overhaul raises that year’s margin while resting on no balance sheet line; it is a borrowed margin, and it surfaces in technical diligence. The effect is not confined to the amount involved, since the finding raises the question of what other undocumented commitments exist. In practice this translates into wider representations and warranties, a higher escrow proportion, and the addition of an asset condition survey to the conditions precedent list. --- Source: https://www.beirek.com/en/blog/capital-expenditure-plan-due-diligence Publisher: BEIREK LLC — https://www.beirek.com