A trade buyer or private equity acquirer runs transactions as an ordinary business activity. They have a template process, a standing diligence panel, a house form of sale and purchase agreement, and a clear internal view of what they will and will not concede. The founder on the other side of the table is doing this once, using advisers they met eight weeks ago, while still running the company. Nothing about that is unfair. It is simply asymmetric, and the asymmetry has a price.
The asymmetry, and where it costs you
The gap does not usually show up as a lower headline multiple. It shows up in the mechanics: an adjustment to earnings that the seller had not anticipated and cannot rebut, a working capital target set by reference to a period the seller did not choose, an earn-out measured on a metric the buyer controls, a warranty schedule that survives longer than the seller assumed. Each of those is negotiable. None of them is negotiable well at three in the morning in week eleven of an exclusive process, which is exactly when they tend to be settled.
The counter is preparation and optionality. Preparation means the company can withstand diligence because the issues were found and fixed by the seller rather than discovered by the buyer. Optionality means more than one credible party at the table. A seller with both of those things negotiates from a different position entirely, and the terms move accordingly.
This article deals with whole-business transactions. If you are negotiating a venture financing rather than a sale, the economics run on different rails — preference, participation and anti-dilution — and we cover those separately in the clauses that cost more than valuation.
Sell side: normalised earnings, and finding the adjustments first
Most private company sales are priced off a multiple of maintainable earnings, usually described as normalised or adjusted EBITDA. The buyer is not buying last year's statutory profit. They are buying an estimate of what the business will earn under their ownership, and every difference between those two numbers is an adjustment somebody has to argue for.
The adjustments a buyer's accountants will look for are predictable. Owner remuneration set for tax reasons rather than market reasons, in either direction. Rent paid to a related party on terms an unrelated landlord would not offer. Genuinely non-recurring costs — a legal dispute, a failed system implementation, a restructure — and, just as often, costs presented as one-off that recur every year in a different guise. Development or fit-out spend capitalised where a buyer would expense it, or the reverse. Revenue recognised on a basis the buyer's policy would not support. Personal expenditure running through the company. Discontinued lines still carrying overhead.
Find these yourself, twelve to twenty-four months before you go to market, and each one becomes an add-back you have evidenced and can defend. Let the buyer find them and two things happen at once. The number moves against you, and — more damaging — your credibility moves against you. A buyer who finds one unflagged adjustment starts looking for the second with a different attitude, and the discount they apply is no longer arithmetic.
Every earnings adjustment discovered in diligence costs twice: once in the multiple, and again in trust. Commission your own quality-of-earnings analysis before the process starts, fix what can be fixed, and disclose the rest with the workings attached. Sellers almost never regret this. They regularly regret skipping it.
Working capital seasonality belongs in the same exercise. A business whose debtor and stock balances swing materially through the year needs that pattern documented and understood by the seller before a buyer proposes a target based on whichever months suit them. More on that below, because it is where the money is.
Sell side: what a buyer prices down
Beyond earnings, a buyer is pricing risk of loss after they own it. Five things drive most of the discount.
- Customer concentration. If a small number of customers carry a large share of revenue or gross profit, the buyer is underwriting your relationships, not the company's. Long-term contracts help; documented relationship depth across more than one person helps more. Genuine diversification, if you have the runway to build it, helps most.
- Key-person dependency. If the business runs on the founder's relationships, pricing judgement or technical knowledge, the buyer will structure to retain that person — through an earn-out, a restraint and a service agreement — or discount for its loss. Building a management layer that can operate without you is the highest-return pre-sale project there is, and it takes years, not months.
- Contract assignability and change of control. Read every material customer, supplier, lease and finance agreement for assignment restrictions and change-of-control clauses. A share sale usually preserves contracts but can still trigger a change-of-control consent; an asset sale requires assignment, and any counterparty whose consent is required has just acquired leverage over your transaction. Find these before the buyer's lawyers do.
- Intellectual property ownership. Confirm the company owns what it uses. Contractor-developed code without a written assignment, a brand used for years but never registered, a domain held personally by a founder, open-source components with licence terms nobody has read. These are cheap to fix in advance and awkward to fix under time pressure.
- The corporate record. Share register, share transfers, director appointments and resignations, minutes of the resolutions that authorised past issues and buybacks, the constitution, any shareholders' agreement and its amendments. In our experience the corporate section of a data room fails more often than the financial section, because nobody owns it. It is also the section that can delay completion outright, since a defective share register is not something a warranty fixes.
The composition and sequencing of the data room itself is a discipline of its own; we set out the full index in what a data room contains.
Share sale or asset sale
This is the first structural fork and it usually divides the parties, because the features that make a share sale attractive to a seller are the same features that make it unattractive to a buyer.
| Dimension | Share sale | Asset sale |
|---|---|---|
| What transfers | The shares in the company; the company keeps everything it owns and owes | Named assets and specified liabilities only; the vendor company remains |
| Historic liabilities | Pass to the buyer with the company, known and unknown | Generally stay with the vendor unless expressly assumed |
| Contracts | Usually continue undisturbed, subject to change-of-control clauses | Require assignment or novation, counterparty by counterparty |
| Employees | Continue with the same employer; terms unchanged | Technically terminate and are re-engaged by the buyer; entitlements must be settled or transferred |
| Warranties | Extensive — the buyer is inheriting history, so the schedule is long | Narrower, focused on title to the assets and their condition |
| Usually preferred by | The seller — clean exit, single transaction, no residual company to wind up | The buyer — chooses what to take, leaves unknown history behind |
The tax consequences differ materially in both directions — for the seller on the character and timing of the gain, and for the buyer on the treatment of the price paid, the availability of deductions and depreciation, GST treatment, and whether the target's accumulated tax attributes survive. Those outcomes depend on the specific facts and on current legislation, and they can be the largest single variable in the deal. They must be modelled with a tax adviser before the structure is agreed, not after the heads of terms are signed, because structure is very difficult to reopen once both sides have anchored on it.
In practice the fork is often resolved by price. A buyer taking a share sale is accepting unknown historic risk, and will want that reflected in warranty cover, an escrow or the number itself. That is a legitimate trade, and it is a better conversation than a stalemate on principle.
Enterprise value to equity value: the bridge, and the peg
The headline price is nearly always an enterprise value — the value of the operating business, independent of how it is financed. What the seller receives is equity value, and the two are connected by a bridge that gets negotiated far less carefully than the multiple.
Enterprise value less net debt, plus or minus the working capital adjustment, equals equity value. Both of those terms are contested. Net debt sounds mechanical and is not. Bank facilities and finance leases are obvious. The argument is about the items in between: shareholder loans, unpaid dividends, accrued bonuses and holiday pay, deferred consideration from an earlier acquisition, dilapidations, unfunded capital commitments, tax owing, related-party balances. Each of these can be characterised as debt-like and deducted from the price. A seller who has not defined the net debt items in the heads of terms will find them defined later by the buyer's accountants, in a report the seller is paying to receive.
The working capital adjustment exists because the buyer is paying for a business delivered with a normal level of working capital in it. Too little and they must fund the shortfall on day one; too much and the seller has effectively left cash behind. So the parties set a target — the peg — and the price adjusts dollar for dollar against actual working capital at completion.
The working capital peg transfers more value in most deals than the last turn of the multiple, and it is almost never negotiated as hard. Argue about three things: the averaging period used to set the target (a full twelve months, not a buyer-selected quarter), the line items included and excluded, and the accounting policies applied — which must be the same ones used to set the target and to measure the outcome. A target set on one basis and measured on another is a one-way price reduction.
The mechanism is symmetrical in principle and asymmetrical in practice, because the buyer's advisers usually draft it. Insist on a worked example appended to the agreement, using real historic figures, showing exactly how the calculation runs. It costs an hour and removes most of the ambiguity that later becomes a dispute.
Earn-outs, deferred consideration and escrow
An earn-out bridges a genuine disagreement about the future: the seller believes the growth is real and imminent, the buyer will pay for it only once it appears. Splitting the difference in structure rather than in price lets both sides sign. It also generates more post-completion disputes than any other clause in the agreement, for a simple reason — the seller no longer controls the business whose performance determines what they are paid.
A workable earn-out has four characteristics. The metric is one the seller can still influence and the buyer cannot readily manipulate; revenue and gross profit are harder to distort than EBITDA, which is exposed to allocated group overhead, changed accounting policy and reinvestment decisions. The metric is defined in the agreement itself, with the accounting policies specified, not left to be determined later in good faith. The period is short — long earn-outs measure a business that has already been integrated beyond recognition. And the seller has express rights: access to the underlying records, a stated frequency of reporting, protective covenants on how the business will be run during the period, and an agreed expert determination route if the calculation is disputed.
Write down three plausible ways the buyer could, acting entirely lawfully, run the business so that the earn-out target is missed — moving a contract to a group entity, allocating head office cost, deferring a launch, changing a revenue policy. If the drafting does not already prevent all three, it is not finished. Assume good faith and draft for its absence.
Deferred consideration without a performance condition is different and simpler: part of the price, paid later. It is still credit risk on the buyer, so it should be documented as a debt, secured or guaranteed where the buyer's balance sheet warrants it, and the seller should understand that the tax point may not wait for the cash. Escrow and retention amounts sit alongside — a portion of the price held by a third party or by the buyer to satisfy warranty claims or a specific identified risk. Negotiate the size, the period, the release mechanics and, critically, whether the escrow is the buyer's sole recourse for warranty claims. Sole recourse is a substantial protection for the seller and is worth conceding elsewhere to obtain.
Buy side: the thesis comes before the targets
Acquirers who start by looking at what is for sale end up buying what was for sale. Write the thesis first, in plain language, and be specific about which of four things you are actually buying: capability you would otherwise have to build, customers or a market position you would otherwise have to win, capacity you would otherwise have to fund, or simply earnings at a price that beats your alternative uses of capital. Those four justify very different prices and very different integration approaches, and a deal that confuses them tends to disappoint on all four counts.
Then model the cost that never appears in the seller's information memorandum: integration. Systems migration and the parallel running that precedes it. Retention arrangements for the people you need to keep. Rebranding, re-contracting, re-licensing. Management attention diverted from the existing business for a year. Advisory fees on both the transaction and the integration. Duplicate cost that persists longer than the synergy case assumes, because redundancy is slower and more expensive than a spreadsheet suggests. Model synergies separately from base performance, phase them realistically, and make sure the deal still works if they arrive a year late.
Fund the acquisition against a proper post-completion cash forecast rather than an annual model. Transaction costs, completion adjustments, retention payments and integration spend all land in the same few months, and the combined entity's working capital cycle is not the sum of the two previous ones. A weekly view through that period is the difference between a tight quarter and a covenant breach — see the 13-week cash forecast.
Diligence, warranties and W&I
Diligence is not verification that the seller told the truth. It is the buyer forming their own view of what they are buying, and each workstream is testing a different question.
- Financial. Is the earnings base maintainable, and does the working capital and capital expenditure profile support the model? This is where the quality-of-earnings report sits. The buyer commissions it — even where the seller has done their own — because they need adjustments tested against their own accounting policies and their own view of what recurs.
- Tax. Are there historic exposures that survive the transaction, and does the proposed structure work for both sides? Payroll classification, GST treatment, related-party pricing and unfiled or amended positions are the usual sources.
- Legal. Title to the shares or assets, the corporate record, material contracts, litigation actual and threatened, leases, consents, regulatory permissions.
- Commercial. Is the market and the company's position in it what the model assumes? Customer interviews, where the seller permits them, are the most informative and most sensitive part of any process.
- Operational. Can the business deliver at the assumed volumes without capital the model does not contain?
- IT. Systems, licences, data, cyber posture and the real cost and timeline of migration or separation.
- People. Who matters, what they are paid, what they are owed, what their agreements say about restraint and change of control, and who will still be there in a year.
Warranties are the seller's statements of fact about the business; indemnities are promises to reimburse a specific identified exposure dollar for dollar. Disclosure is the seller's defence — a properly compiled disclosure letter and bundle qualifies the warranties against what has actually been revealed, which is why the disclosure exercise deserves the same care as the data room. Negotiation then focuses on the caps, the de minimis and basket thresholds below which no claim can be brought, the survival periods (typically shorter for commercial warranties than for tax and title), and the knowledge qualifiers.
Warranty and indemnity insurance is now common in mid-market transactions. A policy takes the buyer's warranty claims to an insurer rather than to the seller, which lets the seller exit cleanly with minimal residual liability and gives the buyer a solvent counterparty. It costs a premium, requires its own underwriting process running alongside diligence, and never covers everything — known issues, and matters the diligence did not adequately cover, are typically excluded. Whether it suits your transaction is a question for your corporate counsel and broker early in the process, not late.
Completion mechanics: completion accounts or locked box
There are two ways to fix the price against the balance sheet, with quite different cash-flow and risk consequences. Under completion accounts, a set of accounts is prepared as at the completion date, and the price adjusts afterwards for actual net debt and working capital. The economic risk and reward of the business pass at completion. The drawback is that the final price is unknown on the day, the accounts take weeks to prepare and are frequently disputed, and both parties carry an open position until they are agreed.
Under a locked box, the price is fixed by reference to a historic balance sheet — the locked box date — and the buyer takes the economic risk and reward from that date, even though they do not own or control the business until completion. Between the two dates the seller is prohibited from extracting value ("leakage") other than agreed items ("permitted leakage"), with an indemnity for breach. The buyer is often compensated for the intervening period through an interest-like charge on the equity price. The price is certain on day one and there is nothing to argue about afterwards, which is why sellers and auction processes favour it. The buyer's protection depends entirely on the quality of diligence over the locked box accounts, so those accounts need to be robust and recent.
| Phase | Indicative elapsed time | What the CFO owns |
|---|---|---|
| Preparation | 12–24 months before launch | Normalised earnings and evidence, corporate record cleanup, contract review, management depth, reporting that stands up monthly |
| Materials and approach | 4–8 weeks | Information memorandum numbers, the financial model, the data room build, the buyer list with advisers |
| Indicative offers | 3–6 weeks | Management presentations, first-round questions, comparing offers on structure rather than headline |
| Heads of terms | 2–4 weeks | Price mechanism, net debt definition, working capital basis, earn-out metric, exclusivity length and conditions |
| Due diligence | 6–12 weeks | Running the data room, the question log, the disclosure exercise, defending adjustments with workings |
| Documentation | Overlaps diligence, 4–8 weeks | Sale and purchase agreement schedules, the completion statement mechanics and worked example, disclosure letter |
| Signing to completion | Days to months, depending on conditions | Conditions precedent, consents, funds flow, completion checklist |
| Post-completion | 3–6 months | Completion accounts or leakage review, escrow administration, earn-out reporting, integration of the finance function |
Timings vary widely with size, complexity and how prepared the seller was. The one column that reliably compresses is diligence, and it compresses in direct proportion to how much preparation happened in the first row.
| Red flag | What the buyer concludes | Usual consequence |
|---|---|---|
| Management accounts that do not reconcile to the statutory accounts | The numbers presented cannot be relied on | Deeper diligence, delay, and a discount for uncertainty |
| Unflagged earnings adjustments found in diligence | There will be others | Price reduction plus a harder line on every subsequent point |
| Defective share register or missing corporate resolutions | Title to what is being sold is uncertain | Completion delayed until rectified; occasionally fatal |
| IP not owned by the company | The core asset may not transfer | Specific indemnity, retention, or the deal stops |
| Revenue concentrated in a few customers on short or informal terms | Post-completion revenue is not secure | Earn-out, retention, or a lower multiple |
| Contractors who function as employees | Historic payroll and leave exposure | Specific indemnity and a holdback |
| Change-of-control clauses in material contracts | Third parties hold a veto | Condition precedent, consent risk, timetable slip |
| Working capital trending down through the sale period | The balance sheet is being managed for the exit | A tighter peg and closer scrutiny of the completion statement |
| Slow, incomplete or inconsistent responses to the question log | Either disorganised or concealing something | Loss of momentum, which is how deals die |
The first 100 days, and why finance decides the verdict
Whether an acquisition is later described as a success is usually settled in the first quarter, and disproportionately by the finance function. The reason is unglamorous: until the acquired business reports on the buyer's chart of accounts, on the buyer's timetable, with intercompany balances that reconcile, nobody can tell what is actually happening. Synergies cannot be tracked, the earn-out cannot be measured credibly, and the board is forming a view of the deal from anecdote.
Have the integration plan written before completion, with named owners and dates. Decide early which systems survive and commit to it — a hedge here costs more than either option. Map the acquired chart of accounts to the group's in the first weeks. Get the combined cash view running immediately, because the combined working capital cycle is the thing most likely to surprise you. Set the earn-out reporting pack up front, in the format the agreement requires, and run it from month one rather than reconstructing it in month eleven under dispute. And be explicit about what is genuinely not changing, because in the absence of information people assume everything is, and the people you most wanted to retain are the ones with the most alternatives.
None of this argues against transacting. Buying and selling businesses is how most private company value is ultimately realised, and a well-run process on a well-prepared business is a good outcome for both sides. The difference between that and the other kind is almost entirely decided before the process starts.
XLCFOWe prepare businesses for sale, run the finance workstream through diligence and completion, and act for acquirers on modelling, diligence coordination and post-completion integration — alongside your corporate counsel and tax advisers.
