ai tooling
AI Financial Model Generator for Business Sales
Prepare cleaner sale models with AI support for revenue, EBITDA, buyer questions, and private matching workflows.
By MergerMatch Editorial TeamPublished Updated Editorial method
An AI financial model generator can help owners and brokers organize historical financials, surface possible adjustments, and prepare buyer questions before a sale process starts. It should preserve links to source records and assumptions so a qualified accountant or adviser can verify every material figure before use.
For acquirers, AI can also help review model logic and compare an opportunity against a mandate. The point is not to replace accounting diligence. It is to make the first financial conversation cleaner.
Preparation workflow
Working through a defined sequence reduces the risk of errors compounding across the model.
| Stage | What happens |
|---|---|
| Source gathering | Upload three to five years of P&L statements, balance sheets, tax returns, and revenue schedules to the AI tool. |
| Revenue bridge | AI identifies volume, price, churn, and customer mix patterns across periods. |
| EBITDA adjustments | Flag owner compensation above market, one-off items, and non-recurring costs for add-back review. |
| Working capital baseline | Calculate the net working capital the business needs to operate day-to-day. |
| Buyer scenario testing | Run sensitivity on the key revenue and cost assumptions a buyer is likely to challenge. |
| Data room handoff | Place the verified model with linked source files in Rooms before buyer access is opened. |
Model areas to prepare
| Area | Why it matters |
|---|---|
| Revenue bridge | Helps buyers see volume, price, churn, and customer mix. |
| EBITDA adjustments | Separates recurring earnings from owner-specific or one-off items. |
| Working capital | Frames how much cash the business needs to operate. |
| Forecast drivers | Connects future performance to measurable operating assumptions. |
| Buyer questions | Turns weak spots into a diligence checklist before review. |
What AI flags but cannot confirm
AI can surface mathematical inconsistencies, flag line items where accounting treatment differs across years, and identify where documentation does not support an add-back claim. It cannot confirm the correct accounting treatment for a given adjustment, assess the legal or tax consequences of owner compensation structure, or verify that filed returns agree with management accounts without human comparison.
Every EBITDA adjustment presented to a buyer should carry an accountant’s note, not just an AI flag.
From model to matching
MergerMatch can use financial ranges and deal-size fit without exposing the full model publicly. An owner can describe a business with revenue in a certain range and EBITDA in a corresponding range, and MergerMatch routes that anonymized profile to buyers whose mandates fit. Detailed files and normalized adjustments belong in Dataroom after disclosure is authorized.
Having a defensible, source-linked model before matching helps buyers and sellers reach a preliminary view on whether a deal is even in range, before expensive diligence begins.
Broker workflow for multiple seller clients
Brokers managing several seller engagements simultaneously face a specific risk: financial assumptions from one client can contaminate the model structure of another if source files are not clearly separated. AI tools must maintain per-engagement file sets, and any owner compensation normalization or revenue adjustment should be labeled by engagement before the broker’s accountant reviews it.
A practical broker financial model workflow looks like this:
| Step | Action | Separation required |
|---|---|---|
| Source upload | Upload P&L, balance sheet, and tax returns per client into separate sessions | Never mix client files in a shared upload |
| Model draft | Generate revenue bridge, EBITDA adjustments, and working capital baseline | Label all outputs with the engagement name and version |
| Normalization notes | Record adjustment rationale and the source line item for each add-back | One accountant review per engagement before any buyer sees the model |
| Scenario outputs | Prepare sensitivity runs on the key assumptions buyers typically challenge | Store outputs in the client’s Rooms workspace with versioned file names |
| Matching handoff | Convert the verified model to deal-size and EBITDA ranges for the anonymized MergerMatch profile | The full model stays in Rooms until both parties are ready for confirmatory diligence |
How acquirers use AI to review financial models
Acquirers receive financial models at different stages of a process. During early matching on MergerMatch, the seller shares only revenue and EBITDA ranges in an anonymized profile. A detailed model is not disclosed until the buyer signals interest and the seller authorizes confirmatory diligence access through Rooms.
When a model does arrive, AI can help an acquirer review it efficiently:
- Mandate comparison: Check whether the opportunity’s revenue, EBITDA, margin profile, and geographic concentration are within the buyer’s stated acquisition mandate. If a fund targets EBITDA between US$1M and US$5M, an AI tool can extract the seller’s normalized EBITDA from the model and flag whether it sits inside or outside that range.
- Revenue concentration check: Identify whether the top three to five customers account for an unusual share of total revenue, which is a common discount trigger in SME deals.
- Working capital assessment: Compare the trailing working capital peg to industry benchmarks and flag whether the seller has normalized for seasonal peaks or excluded specific items without adequate disclosure.
- Preliminary question list: Generate structured diligence questions from model anomalies — for example, a spike in cost of goods sold in one year that does not correspond to a revenue change.
- Comparison across opportunities: Where an acquirer is reviewing multiple opportunities simultaneously, AI can produce a side-by-side summary of key metrics, allowing the team to triage which opportunities warrant deeper attention.
None of this replaces a qualified accountant or financial adviser reviewing the underlying statements. It reduces the time between receiving a model and forming a preliminary view on whether the opportunity is worth pursuing.
Modeling different deal structure scenarios
Sellers benefit from building scenario models before buyers arrive. An acquirer evaluating a business through MergerMatch may apply any of several structures depending on their fund type, capital availability, and risk tolerance. A seller whose model already covers debt coverage, earn-out thresholds, and seller-note repayment is able to respond to early structure questions before a formal data room opens.
AI can draft initial scenario outputs from historical financials. A qualified accountant or financial adviser should verify any structure before it is shared with a buyer.
| Deal structure | What AI can help model | Key buyer questions it anticipates |
|---|---|---|
| Full cash acquisition | Revenue bridge, normalized EBITDA, working capital peg, and target price range based on comparable multiples | What is the normalized earnings base, what working capital is included in the price, and is the stated EBITDA defensible? |
| Leveraged acquisition | Debt service coverage from historical EBITDA, interest coverage ratio, and loan-to-value parameters | Can the business service acquisition debt from operating cash flow, and what covenant headroom remains? |
| Earn-out arrangement | Milestone revenue or EBITDA thresholds, earn-out period projections, and risk-adjusted baseline scenarios | What is the probability of hitting the earn-out trigger, and how will revenue be measured during the earn-out period? |
| Seller financing | Implied total price with deferred consideration, seller note terms, and repayment capacity from projected free cash flow | Does the business generate enough free cash flow to service the seller note alongside operational needs? |
| Minority or partial recap | Implied enterprise value, retained equity percentage, and projected owner liquidity at different hold periods | What equity stake does the minority acquirer receive, and what exit path or liquidity event does the structure assume? |
Running these scenarios before matching begins means the seller has already worked through the questions a buyer is most likely to ask. Gaps in the model become a preparation checklist rather than a discovery during negotiation.
Modeling a partial exit, equity recapitalization, or management buyout
A full cash acquisition is one outcome. Many SME owners prefer to sell a controlling stake while retaining equity, bring in a growth partner through a recapitalization, or support a management team in buying the business. Each of these structures requires a different financial model, a different buyer category, and different matching criteria in an anonymized seller profile.
IBBA Market Pulse data shows that non-full-sale structures — partial exits, equity recapitalizations, and management-led transactions — represent a consistent share of completed SME transactions, particularly in the US$1M to US$10M EBITDA range where owners retain operational capacity alongside capital constraints.
| Structure | What the model must show | Buyer mandate criteria that apply | Common AI preparation task |
|---|---|---|---|
| Minority stake sale to a financial sponsor | Retained equity value at the implied enterprise multiple, projected hold-period return at various exit multiples, debt capacity if leverage is introduced at closing | Minority control rights, sector and geography mandate, return horizon of three to seven years | Model retained equity stake value at the agreed implied multiple and at three exit-multiple scenarios, showing the range of outcomes over the expected hold period |
| Controlling stake with seller rollover | Enterprise value at closing, seller rollover percentage and resulting equity stake, growth assumptions and target exit multiple over the rollover period | Sector experience and operational capacity, rollover comfort, financial profile within mandate | Model the rollover equity stake at various exit multiples and hold periods, showing total seller proceeds as a current cash component plus a deferred equity component |
| Equity recapitalization | Recapitalization amount as a proportion of enterprise value, remaining equity stake and implied value, debt service capacity from operating EBITDA | Leverage appetite, control or co-control preference, sector and geographic mandate | Model the dividend amount against EBITDA coverage ratios and show remaining equity value alongside projected total proceeds at different exit multiples |
| Management buyout | Acquisition price funded by management equity contribution and senior or subordinated debt, debt service coverage from historical EBITDA, vendor financing terms from the selling owner if applicable | Management team track record, leverage capacity, specialist MBO lender or subordinated debt appetite, typically not a traditional buyer mandate on a matching platform | Model debt coverage at different leverage levels, calculate management equity required at various price points, show vendor loan repayment capacity from projected free cash flow |
For each of these structures, the AI model draft serves the same role as in a full sale: identifying gaps, testing assumptions under stress, and preparing the questions a financial partner or adviser will ask before any binding commitment. A qualified accountant or financial adviser should review any structure model before it is shared externally.
On MergerMatch, the control or minority structure preference in a buyer’s registered mandate determines which seller profiles are routed to them. A seller whose model supports a partial sale or management buyout should specify their preferred structure in the anonymized profile so matched buyers are already aligned on structure before a conversation begins. Structure mismatch — where a buyer mandate requires full control but the seller profile indicates a minority stake — is one of the most common reasons an early buyer conversation does not progress to a Rooms access request.
The IBBA Market Pulse includes quarterly data on partial and management-led transaction structures, helping sellers and brokers calibrate preparation depth against the realistic buyer pool in their sector and size range.
Industry-specific model preparation: what AI adjusts by sector
Financial models for SME sales are not one-size-fits-all. Buyers in different industries apply different scrutiny to different line items, and the EBITDA adjustments that are normal in one sector can trigger questions in another. AI can adapt the model’s structure and focus areas based on the business type, helping owners and brokers concentrate preparation on the metrics buyers actually challenge.
| Sector | Key buyer metrics | AI adjustment focus | Common buyer challenge |
|---|---|---|---|
| Professional services and consulting | Revenue per fee-earner, client concentration, average engagement length | Normalize partner or owner compensation to market rate, separate project fees from retainer income | Will revenue hold without the departing owner, and how many clients are relationship-dependent? |
| Retail and ecommerce | Gross margin, inventory turnover, customer return rate | Separate channel revenue streams, normalize marketing spend for owner-funded campaigns | What is the margin trend, and what is the inventory quality at time of sale? |
| Manufacturing and distribution | Gross margin, capacity utilization, order backlog | Normalize owner compensation, exclude one-off capital expenditure, build backlog schedule | What is the customer concentration risk, and what equipment requires replacement near-term? |
| SaaS and software | Monthly recurring revenue, churn rate, customer acquisition cost payback | Distinguish recognized recurring revenue from one-off license fees, calculate net revenue retention | Is churn trending up, and what is the customer renewal concentration? |
| Hospitality and food service | Revenue per seat or room, occupancy rate, labor cost ratio | Exclude pandemic-period anomalies, separate event revenue from base trading, normalize lease cost | What are the lease renewal terms, and does the trailing twelve-month trend represent a stable trading base? |
| Healthcare and allied health | Revenue per licensed professional, payer or referral mix, practitioner dependency | Identify which licenses and referral agreements transfer to a new owner, normalize practitioner compensation | Is the business dependent on one or two licensed practitioners, and are key referral relationships transferable? |
IBBA Market Pulse quarterly data shows that EBITDA multiples vary materially by sector and deal size for completed SME transactions, and that preparation informed by sector-specific buyer expectations reduces the most common buyer objections from four or five to two or three per engagement.
For matching on MergerMatch, the anonymized profile uses revenue and EBITDA ranges drawn from the verified model. The sector-specific preparation is used internally and in Rooms during buyer diligence, not in the initial anonymous profile that routes to matched buyers.
Pre-Rooms model handoff: what to verify before buyer access opens
The preparation workflow ends with “place the verified model in Rooms before buyer access is opened,” but that verification step deserves a structured approach. A buyer who finds a material discrepancy after the data room opens typically treats it as a credibility issue, not a drafting oversight, which affects their view of all other disclosed materials.
Five checks reduce the risk of a buyer finding a gap in the first review:
| Verification check | What to confirm | Common gap that reaches a buyer |
|---|---|---|
| Source reconciliation | Every revenue, EBITDA, and balance sheet figure in the model traces to the corresponding line in the uploaded P&L, balance sheet, or tax return | AI draft uses a rounded or estimated figure from a management summary rather than the source line item, creating a mismatch a buyer finds when cross-referencing statements |
| EBITDA adjustment documentation | Each add-back carries a labeled source line item, a one-line rationale, and an accountant confirmation note — not a number alone | Adjustment is present in the model without a traceable explanation, which a buyer’s adviser flags during confirmatory diligence as an unsupported claim |
| Revenue classification accuracy | Recurring, project, and one-off revenue correctly separated in the bridge, with seasonal peaks identified and normalized | A seasonal revenue month included in the recurring revenue figure without qualification inflates the recurring income percentage a buyer uses to set a preliminary valuation |
| Working capital baseline method | Net working capital calculated using a trailing twelve-month average and normalized for unusual items, with the methodology stated | A working capital figure calculated from the most recent month during a high-inventory period overstates the peg the buyer must fund at closing |
| Scenario output labeling | Every deal structure scenario (leveraged acquisition, earn-out, seller financing) labeled as an illustrative estimate with its key assumptions explicitly stated | A projected EBITDA figure in a scenario output read as a confirmed figure by a buyer who receives it without the assumption context, leading to a mismatch in expectations at LOI |
IBBA Market Pulse data indicates that sellers who pre-disclose identified model adjustments and their supporting rationale before confirmatory diligence receive fewer formal buyer inquiries on financial accuracy during the Q&A phase, which is associated with shorter time-to-LOI intervals.
For MergerMatch Rooms, the verification checklist should be completed before the first buyer access group is opened. AI can compare the current model figures against each uploaded source file and flag mismatches, but the accountant and broker confirm each flag before the model is marked as verified and Rooms access is authorized.
The NIST Generative AI Profile, published in 2024, recommends testing, documentation, and human oversight for generative AI risks. The European Commission’s AI Act overview also identifies documentation, accuracy, traceability, and human oversight as central controls for relevant higher-risk systems. The IBBA Market Pulse survey, published quarterly by the International Business Brokers Association and M&A Source, tracks EBITDA and SDE multiples, deal structures, and seller-financing patterns across SME transactions, providing market context for normalizing financial model assumptions.
For financial work, retain the uploaded statement, model version, transformation rule, assumption owner, and review date. A generated number is not verified merely because it reconciles mathematically. Accounting treatment and transaction adjustments still require qualified human review.
Responding to buyer financial challenges during active diligence
After the model enters Rooms and buyers begin their review, the seller and broker are likely to receive challenges to specific figures. The five most common challenge types in SME financial model reviews are predictable, and AI can help prepare structured responses before a challenge arrives. A response provided to one buyer that confirms, revises, or adds to a model figure is a material disclosure and must be issued to all active buyer groups with Rooms access at the same time.
| Buyer challenge | AI drafting task | Equal-disclosure requirement | Common gap |
|---|---|---|---|
| EBITDA add-back dispute | Draft a structured add-back support note with the source line item, rationale, and accountant confirmation for each contested adjustment | Add-back support note issued to all active buyer groups simultaneously through Rooms Q&A threads, not only the buyer who raised the question | Broker sends an informal clarification to one buyer without issuing the same note to all active groups, creating unequal information that becomes a negotiating point at LOI |
| Revenue concentration challenge | Draft a concentration bridge showing customer count, revenue share by tier, contract renewal dates, and any documented diversification actions | If customer-level detail is shared in response to the challenge, the same detail is disclosed to all buyers with current Rooms access | Seller reassures one buyer verbally about concentration risk without providing the same written confirmation to other groups, leading to inconsistent valuation expectations |
| Working capital peg dispute | Recalculate the working capital baseline under the buyer’s proposed method and present both the seller’s method and the buyer’s alternative, with source reconciliation | A revised working capital methodology disclosed to one buyer must be offered to all buyers through the same formal channel | Broker informally agrees to a revised working capital assumption with one buyer before presenting the same basis to remaining active groups, creating a sequencing problem in multi-buyer processes |
| Revenue classification challenge | Produce a revenue bridge separating recurring, project, and one-off revenue by customer and period, with source line items from the uploaded P&L statements | Revenue classification clarification is a material note and must be issued to all active buyer groups simultaneously | Seller separately confirms to one buyer that a category is recurring without updating all other groups, leading to different base assumptions across active buyers |
| Forecast assumption challenge | Draft a structured assumption note explaining the basis for each contested forecast driver, supported by historical operating metrics | A projection assumption note is material and must be issued to all active buyer groups at the same time | Seller informally agrees to revise a growth assumption with one buyer without issuing a formal revised note to all other groups, creating asymmetric deal assumptions ahead of LOI |
AI can help the seller and broker anticipate which challenges are most likely before Rooms access is opened. Reviewing the model against these five challenge types during the pre-Rooms verification pass means many responses are already prepared before a buyer raises the question.
For MergerMatch Rooms, each challenge response should be logged in the relevant buyer group’s Q&A thread so the process record is complete. If a clarification is material to all groups, a supplement note issued to every active group simultaneously is preferable to per-buyer Q&A responses that create a fragmented disclosure record.
IBBA Market Pulse data indicates that sellers whose financial model responses are documented, consistent across buyer groups, and available as formal Rooms records receive fewer repeat questions on the same point during confirmatory diligence, which is associated with shorter time-to-signed-LOI intervals for SME transactions in the US$1M to US$15M range.
Preparing for closing price adjustments and completion accounts
After the LOI is signed and the exclusivity period is running, the financial model’s next role is in the closing price mechanics. Most SME acquisition agreements include either a completion accounts mechanism, which adjusts the final price based on actual working capital, net debt, and cash at close, or a locked-box structure, which fixes the price at a historical balance sheet date and restricts any value removal after that date. AI can help build and maintain the schedules that support these adjustments before close and help prepare a structured response to a buyer’s completion accounts challenge after close.
| Closing task | What the financial model supports | AI preparation task | Common gap |
|---|---|---|---|
| Reference working capital schedule | The agreed working capital amount used to calculate the closing price adjustment | Build the trailing twelve-month working capital average from P&L and balance sheet uploads and map it to the balance sheet categories included in the agreed definition | Seller uses a single-month snapshot rather than a methodology-stated trailing average, leaving the reference figure open to challenge when actual working capital at close differs |
| Net debt schedule at the reference date | All debt instruments, debt-like items (deferred tax liabilities, pension obligations, accrued liabilities exceeding normal operating levels), and cash counted in the enterprise-to-equity bridge | Extract all debt-like items from the balance sheet and map them against the buyer’s debt definition in the LOI | Seller omits a debt-like item that the buyer’s adviser identifies at close, requiring a last-day price revision and creating negotiation friction after conditions are otherwise satisfied |
| Locked-box leakage monitoring during exclusivity | Any value removed from the business between the locked-box date and closing (dividends, unusual distributions, related-party payments above agreed thresholds) | Flag balance sheet movements during exclusivity that may qualify as leakage against the locked-box restriction clauses in the LOI | Owner does not recognize a related-party payment as a leakage event and it is discovered by the buyer’s adviser in the closing balance sheet |
| Completion accounts statement preparation | The formal financial statement submitted at or after close showing actual working capital, net debt, and cash against reference figures | Draft the completion accounts format from the LOI definitions, pre-populate from the most recent balance sheet, and generate the adjustment calculation | Seller submits completion accounts in a different format than the LOI specifies, triggering a dispute on methodology rather than figures |
| Completion accounts challenge response | The seller’s formal response when the buyer disputes a specific line item in the completion accounts | Draft a line-item response note with the source balance sheet entry, the agreed definition, and the calculation method for each disputed item | Seller responds informally to a buyer challenge without a structured source-linked response, which the buyer’s adviser treats as ambiguous and escalates to the independent accountant dispute mechanism |
For locked-box transactions, the locked-box date is typically the most recent year-end or half-year balance sheet date before the LOI is signed. The seller’s accountant prepares the locked-box accounts and confirms the leakage definition. AI can generate the monitoring template from the LOI’s restriction clauses, but every payment reviewed against that template requires the accountant’s confirmation that it qualifies or does not qualify as leakage.
Completion accounts disputes are governed by the provisions of the purchase agreement rather than the LOI. The seller’s legal adviser should review all completion accounts responses before submission, and any amount in dispute above the agreement’s threshold is typically resolved by an independent accountant appointed under the dispute resolution mechanism in the purchase agreement.
For MergerMatch Rooms, placing the reference working capital schedule and net debt schedule in the data room during confirmatory diligence gives the buyer’s advisers the source material to prepare their own closing statement in parallel. Uploading both schedules at the same time as the formal financial statements reduces the risk of a methodology dispute at close.
The IBBA Market Pulse quarterly data on working capital pegs, seller financing, and closing adjustments provides useful benchmarks for SME transactions in the US$1M to US$15M range.
Designing the earn-out financial model when contingent consideration is part of the structure
When a buyer offers contingent consideration in the form of an earn-out tied to post-close revenue, EBITDA, or gross profit, the financial model serves a different purpose from the full cash acquisition model. The seller must evaluate whether the earn-out thresholds are realistic, how the buyer’s post-close decisions could affect whether the metrics are achieved, and what the present value of the contingent component actually is once risk and timing are incorporated. IBBA Market Pulse data shows earn-out provisions appear in a consistent share of completed SME transactions in the US$1M to US$10M EBITDA range, and earn-out calculation disputes are among the most common post-close litigation areas for completed SME transactions.
| Earn-out modeling task | What the model must show | Common gap |
|---|---|---|
| Metric selection and seller risk profile | Model the earn-out calculation under three metric options: revenue, EBITDA, and gross profit. Revenue earn-outs are less exposed to post-close cost allocation changes but harder for buyers who plan integration costs to accept. EBITDA earn-outs are the most common but expose the seller to buyer-controlled accounting and cost decisions after close. Gross profit earn-outs often work for businesses where post-close overhead changes are expected but revenue quality is the seller’s primary performance signal. | Seller accepts the buyer’s proposed metric without modelling how each metric responds to the buyer’s likely post-close decisions, including new overhead allocation, management fee charges, or integration costs that reduce EBITDA without operational underperformance. |
| Accounting treatment protection scenario | Model the earn-out outcome under two scenarios: accounting treatment consistent with the pre-close basis, and a scenario where the buyer introduces changes such as increased amortization, management fee charges, or overhead reallocation. The dollar difference between the two scenarios is the minimum value of an accounting-basis protection clause in the SPA. | Earn-out provisions drafted without a stated accounting basis, leaving the seller dependent on a phrase like “consistent with past practice” that is difficult to enforce when buyer-side costs are introduced as legitimate post-acquisition management decisions. |
| Post-close budget and business plan constraint modeling | Model the earn-out outcome under scenarios where the buyer restricts capital expenditure, marketing spend, or headcount below the seller’s pre-close plan. The model should show how much restriction closes the earn-out payment entirely and what percentage of the contingent amount is at risk under a conservative post-close budget. | Earn-out thresholds set from the seller’s historical growth trajectory without modelling the impact of the buyer’s integration plan, which often reduces investment in the earn-out period. |
| Non-interference clause value quantification | Model the difference in earn-out payments between a scenario where the buyer operates the business without material operational interference and a scenario where buyer-directed pricing changes, customer reassignments, or distribution decisions affect the metric. The dollar difference is the minimum value the non-interference clause represents and the minimum stake worth protecting in SPA negotiations. | Non-interference clauses drafted as general statements without the seller having quantified what operational interference is worth, which weakens enforcement when a specific buyer decision is challenged without a defined threshold. |
| Earn-out acceleration and business change scenario | Model the earn-out payment the seller would receive if the buyer sells or restructures the business before the earn-out period ends, under a structure with no acceleration clause versus one with a defined acceleration payment. The model should show whether the buyer could realize a profit on a subsequent transaction before the earn-out is fully paid. | Earn-out agreement drafted without an acceleration clause. The seller discovers mid-earn-out that the buyer is planning a further transaction that restructures the business before the earn-out measurement period is complete, with no contractual basis for compensation beyond the accrued partial amount. |
On MergerMatch, the matching profile should specify whether the seller accepts contingent consideration or requires a defined cash floor at close. Buyers whose mandate requires full cash will not advance with an earn-out-oriented seller regardless of financial fit. Specifying the structure preference in the matching profile before the teaser and IM are prepared avoids buyer conversations that stall on structure rather than commercial terms.
AI can draft each earn-out scenario from historical financials and the proposed earn-out terms in the buyer’s indicative offer. The metric definition, accounting basis protections, non-interference provisions, and acceleration clause all require qualified legal and financial adviser review before the seller accepts any earn-out proposal. The resulting models show the range of risk in the contingent component, not confirmed valuations.
IBBA Market Pulse data on earn-out structures in completed SME transactions shows that disputes over earn-out calculations, accounting treatment adjustments, and buyer interference with operations post-close are among the most common sources of post-closing litigation in the US$1M to US$10M EBITDA range.
Commissioning a sell-side quality of earnings report to support matching and diligence
A sell-side quality of earnings report is an independent accountant’s review of a seller’s financial model, EBITDA adjustments, revenue quality, and working capital methodology before buyers begin their own review. Sellers and brokers who commission a sell-side QoE before placing materials in Rooms give a buyer’s advisers a starting point that reduces the first round of diligence questions and often shortens the time between data room opening and LOI.
AI can help prepare the inputs that make a QoE engagement faster and less expensive. The accountant’s scope and conclusions remain theirs, but arriving at the engagement with organized source files, a labeled adjustment schedule, and a clean revenue bridge from AI-prepared inputs reduces the professional hours required to complete the review.
| Task | What it involves | Common gap |
|---|---|---|
| Deciding whether a sell-side QoE is appropriate | A QoE makes economic sense when the transaction is above approximately US$2M to US$5M EBITDA, the EBITDA adjustment schedule is complex with more than three owner-specific add-backs, the buyer universe includes financial sponsors who will commission their own QoE regardless, or the sale follows a period of financial restructuring where normalized figures diverge significantly from filed accounts. Below that threshold, a targeted accountant review of the EBITDA adjustments may provide equivalent buyer confidence at lower cost. For seller-financed or management buyout transactions, the QoE has additional value because the seller’s financial model directly sets the repayment capacity the buyer relies on. | Seller or broker commissions a full QoE scope for a transaction where a targeted adjustment review would provide the same buyer confidence at materially lower cost, reducing funds available to complete the deal preparation. |
| Preparing AI-assisted inputs for the QoE engagement | Upload the last three to five years of P&L statements, balance sheets, tax returns, and management accounts to the AI tool. AI generates a preliminary adjustment schedule with labeled source line items, a trailing twelve-month revenue bridge by customer and channel, and a working capital analysis flagging months with unusual movements. These outputs do not replace the accountant’s work, but they give the QoE team organized source material rather than raw files. AI can also identify which adjustment candidates require additional documentation before the accountant’s review begins, reducing back-and-forth during fieldwork. | Seller arrives at the QoE engagement with unorganized files and verbal adjustment explanations, extending fieldwork hours and increasing professional fees for document reconstruction work that AI could have prepared in advance. |
| Understanding what the QoE typically covers for SME transactions | A sell-side QoE for SME transactions covers five areas: revenue quality (recurring versus one-off classification and customer concentration), EBITDA adjustments (owner compensation normalization and non-recurring cost add-backs), working capital normalization (trailing methodology and seasonal adjustment), balance sheet liabilities (identifying debt-like items in the enterprise-to-equity bridge), and management accounts reconciliation to tax filings. Not all QoE engagements cover every area at the same depth. The seller and broker should confirm scope with the accountant before the engagement begins. | Seller assumes the QoE covers items outside its agreed scope. A buyer’s adviser finds an unstated assumption in the model that the QoE did not address, creating a targeted question the QoE findings cannot answer. |
| Placing the QoE in Rooms and managing buyer reliance | Place the QoE report in a restricted folder in Rooms that opens only when the buyer has signed a reliance letter or agreed reliance terms with the accountant. Without reliance, buyers can use the QoE as context but cannot formally rely on it for a claim. Record the QoE version and date in Rooms. If the business has a material financial change after the QoE is completed, note the change and the version gap clearly before reopening buyer access, or commission an updated scope. | Seller places the QoE in Rooms without reliance terms. A buyer relies informally on QoE figures during LOI negotiations, then disputes a post-close adjustment on the basis that the QoE confirmed the relevant figure. Without a reliance agreement, the seller has no clear protection against the claim. |
| Responding when a buyer’s accountant challenges a QoE finding | If a buyer’s accountant disagrees with a QoE adjustment, the first response should be a structured note from the seller’s accountant explaining the basis for the finding, the source line item, and the methodology. Where the challenge raises a new fact not available during QoE fieldwork, the seller should assess whether it warrants an updated scope before the LOI is finalized. The challenge response, when issued, is material disclosure and must go to all active buyer groups with Rooms access, not only the buyer whose accountant raised the challenge. | Seller’s broker informally concedes a QoE adjustment point to one buyer without issuing the same concession to all active groups, creating an asymmetric financial basis across LOI submissions. |
For MergerMatch matching, the existence of a sell-side QoE does not change the anonymized profile content. Financial ranges in the matching profile come from the verified model. The QoE is placed in Rooms after buyer interest is confirmed and NDA terms are in place, not in the initial matching profile.
The IBBA Market Pulse data on SME transactions shows that completed transactions in the US$5M enterprise value range and above more frequently involve a quality of earnings review initiated by the seller or buyer, and that the presence of organized pre-QoE financial material is associated with fewer re-negotiated adjustments during the confirmatory diligence phase.
FAQ
Can AI validate EBITDA adjustments?
AI can flag unusual items and help organize support. A qualified adviser or accountant should review final adjustments.
Should buyers receive the model before identity disclosure?
Usually no. Early matching should use ranges and anonymized fit. Detailed models can follow in a controlled data room.
What financial files does AI need to build a sale model?
Profit and loss statements for three to five years, monthly revenue detail, customer concentration breakdown, and any normalized EBITDA adjustments the owner or adviser has prepared. Tax returns help confirm filed figures.
When should an owner bring in an accountant?
Before sharing any model with a buyer. AI can draft the structure and flag anomalies, but EBITDA adjustments and normalized earnings should be reviewed by a qualified accountant before external disclosure.
What financial metrics do buyers most commonly challenge in SME transactions?
Revenue concentration is the most common challenge point — buyers discount when one or two customers account for a large share of revenue. Owner compensation normalization, working capital peg, and recurring versus one-off revenue classification are also frequent points of buyer scrutiny.
How does AI help a seller prepare for buyer questions about deal structure?
AI can use historical revenue and EBITDA figures to generate preliminary scenario outputs for common deal structures: full cash acquisition, earn-out, seller financing, and leveraged acquisition. The output is a draft that identifies gaps and probable buyer questions before any buyer arrives. An accountant or financial adviser should review any structure model before the seller uses it in a buyer conversation.
Does the financial model preparation process differ by industry?
Yes. Buyers in different sectors apply different scrutiny to different line items. Professional services models focus on revenue per fee-earner and client concentration. Manufacturing models focus on capacity utilization and equipment replacement schedules. SaaS models focus on churn rate and net revenue retention. AI can adapt its adjustment focus to the sector once the business type is identified from uploaded financials.
What should a broker or accountant check before placing a financial model in the data room?
Five checks reduce the risk of a buyer finding a material discrepancy after Rooms access opens: source reconciliation (every P&L figure traces to the uploaded source statement), EBITDA adjustment documentation (each add-back carries a labeled source line item and rationale, not just a number), revenue classification accuracy (recurring, project, and one-off revenue correctly separated in the bridge), working capital baseline method (trailing twelve-month average normalized for unusual items, not a single-month snapshot), and scenario output labeling (every deal structure scenario labeled as an illustrative estimate with stated assumptions). A buyer who finds a reconciliation gap in the data room typically interprets it as a credibility issue, not a drafting oversight.
How does a financial model for a partial sale or equity recapitalization differ from a full business sale?
A full sale model focuses on enterprise value, normalized EBITDA, and what the buyer pays for 100% of the business. A partial sale or equity recap model must also show the implied value of the equity being sold, the value the seller retains or rolls over, and the financial profile of the business after the transaction structure is applied. For a leveraged partial sale, that includes debt service capacity from EBITDA. For a rollover or minority stake sale, it includes the hold-period return model that a financial sponsor uses to evaluate the opportunity. For a management buyout, it includes the equity contribution required from the management team and the leverage the business can support. AI can draft scenario outputs for each from historical financial statements, but the retained equity valuation and deal structure should be reviewed by a qualified accountant or financial adviser before any external conversation.
How should a seller or broker respond when a buyer challenges a financial model figure during active diligence?
Treat any buyer challenge to a financial model figure as a potential disclosure event for all active buyer groups, not only the buyer who raised it. If the response confirms, revises, or adds to a model figure, that response should be issued through the Rooms Q&A system to every group currently in the process. AI can help draft a structured response note that includes the source line item, rationale, and accountant confirmation. For EBITDA add-back disputes, the response should carry a labeled source and rationale for each contested adjustment. For revenue concentration or classification challenges, the response should include a line-item bridge from the P&L. For working capital or forecast challenges, both the seller’s original methodology and the clarification should be documented so any revised baseline is clear. The equal-disclosure principle that applies to CIM Q&A applies equally to financial model challenges during active diligence.
How does the financial model feed into the closing price adjustment at completion?
In most SME transactions with a completion accounts mechanism, the final price the seller receives depends on whether actual working capital, net debt, and cash at close match the reference figures agreed in the LOI. AI can help prepare three inputs before close: the reference working capital schedule (establishing the agreed methodology and the trailing average figure), the net debt schedule (mapping all debt instruments and debt-like items to the LOI definition), and a locked-box leakage monitoring log (tracking balance sheet movements during exclusivity against the leakage restrictions). After close, AI can draft the completion accounts statement in the format required by the purchase agreement and prepare a structured source-linked response if the buyer challenges a specific line item. Any formal submission and all dispute responses require review and authorization from the seller’s legal adviser and accountant before they are sent.
How should a seller prepare a financial model when an earn-out is part of the proposed deal structure?
Five modeling tasks are distinct from the full cash model. First, select the metric (revenue, EBITDA, or gross profit) and understand how each responds to post-close buyer decisions, including overhead allocation or management fee charges introduced after close. Second, model the earn-out outcome under both consistent accounting treatment and a scenario where the buyer makes legitimate post-acquisition cost changes that reduce EBITDA without operational underperformance. Third, map how post-close budget restrictions reduce the metric below the earn-out threshold and affect total contingent payments. Fourth, quantify the value of a non-interference clause by comparing earn-out payments with and without operational interference by the buyer. Fifth, model an acceleration scenario showing the seller’s position if the buyer sells or restructures the business before the earn-out period ends. AI can draft each scenario from historical financials and the proposed earn-out terms, but the structure, metric definition, accounting basis protections, and non-interference and acceleration provisions all require qualified legal and financial adviser review before the seller accepts any earn-out proposal.
When should a seller or broker commission a sell-side quality of earnings report before matching begins?
A sell-side QoE makes economic sense when EBITDA is above approximately US$2M to US$5M, the adjustment schedule is complex with more than three owner-specific add-backs, the buyer universe includes financial sponsors who will commission their own QoE regardless, or the business has recently restructured. AI can prepare organized source inputs for the QoE engagement, including adjustment schedules, revenue bridges, and working capital analyses, reducing professional hours and fees. The QoE report is placed in Rooms under reliance terms after buyer interest is confirmed and NDA terms are in place, not in the initial matching profile. Any challenge from a buyer’s accountant to a QoE finding is material disclosure and must go to all active buyer groups simultaneously, not only the group whose accountant raised the question.