Beyond the headline sale price

Research date: September 15, 2026. Market data, deal-term benchmarks, valuation multiples and legal status described here reflect what was verifiable on that date. Multiples move fast; legal and tax rules move slower but move. Re-check anything time-sensitive before relying on it.

This is not legal, tax, or financial advice

This chapter is general information assembled from public sources. It is not legal advice, tax advice, accounting advice, or investment advice, and reading it does not create any professional relationship with anyone.

Selling a company is one of the few founder activities where doing it without professional help is close to indefensible. The chapter flags the specific points where a lawyer and a tax professional are mandatory rather than optional. Those flags are not hedging. They mark places where the downside of getting it wrong is measured in six or seven figures and is frequently irreversible.

Two things in particular:

• A transaction lawyer with M&A experience is mandatory. Not your incorporation lawyer, not a generalist. Purchase agreements are 60–120 pages of allocated risk, and the allocation is not intuitive.
• A tax professional is mandatory, and must be engaged before the letter of intent, not after. The tax outcome of a sale is frequently worth more than the last full turn of negotiation on headline price. A structure decision made casually in an LOI can cost more than every price concession you fought for. This is covered in detail in Part J.

The organizing theme: asymmetry

A first-time seller does this once. The buyer across the table does it constantly — a private equity firm's corporate development team may run twenty processes a year; a serial strategic acquirer's M&A group runs a standing playbook; the buy-side lawyer has papered hundreds of these agreements and knows exactly which clauses are conventionally negotiated and which are conventionally conceded.

This asymmetry is not a metaphor. It shows up mechanically:

• The buyer knows that leverage collapses the moment exclusivity is signed, and prices its letter of intent accordingly — generous headline, vague everything else.
• The buyer knows which diligence findings reliably support a retrade and which do not.
• The buyer knows that a seller who has already told their spouse, their team, and themselves that the deal is happening has an emotional sunk cost that is worth real money at the margin.
• The buyer's advisors are paid to do this well. The seller's advisors, if the seller has any, are frequently paid on a success fee that makes closing their objective, which is not identical to the seller's objective.

Every structural recommendation in this chapter is an attempt to narrow that gap. The two highest-leverage moves, stated up front because most of the rest is detail:

1. Do the cleanup 12–24 months before you run a process, so diligence finds nothing.
2. Negotiate everything that matters before signing the LOI, because afterwards you are negotiating against a counterparty who knows you have no alternative.

How to read this chapter

It runs in roughly transaction order: preparation → buyers → intermediaries → valuation → process → diligence → structure → earn-outs → proceeds → tax → closing → failure. Sections can be read independently.

Labels: [Verified] — stated in a cited primary or near-primary source. [Estimate] — a practitioner or marketplace range with imperfect methodology or a self-selected sample. [Analysis] — the author's reasoning, not a sourced fact.

Medians, not means. Deal data is severely right-skewed; where a source gives both, the median is quoted.

Survivorship bias, applied throughout. Almost all published deal data covers deals that closed. SRS Acquiom's comes from transactions where it acted as payments or shareholder-representative agent; the ABA's from signed, publicly filed agreements; marketplace data from listings that sold. None includes companies that went to market and found no buyer, LOIs that died in diligence, or founders who quietly wound down. Read "median escrow was 10%" as "median escrow among companies good enough to get bought." Part L is the partial corrective.

PART A — Preparation and positioning

A1. What makes a company saleable, and what makes it unsaleable

A saleable company is not the same as a good company. Buyers are not paying for the thing you are proud of. They are paying for a bundle of future cash flows they believe will survive your departure, plus whatever strategic value the asset has inside their organization. Saleability is mostly about transferability and verifiability.

The saleability checklist, in rough order of how often each one bites:

[Analysis] The single question behind all of it: can the buyer own this, and can they prove what they are buying? A business that generates $2M of cash but where the buyer cannot establish clean title to the software, cannot verify the revenue, and cannot operate it without you is not a $2M-cash business from the buyer's point of view. It is a lawsuit with a revenue stream attached.

A2. The diligence problems that kill deals late

Late is the operative word. These do not surface at the teaser stage or the management meeting. They surface after the LOI, after exclusivity, when you have no leverage and have already spent money. In Axial's analysis of 75 broken lower-middle-market deals in 2025, diligence findings outside the quality-of-earnings work accounted for 25.3% of failures and quality-of-earnings discrepancies for a further 21.3% — combined, roughly 46–47% of all deal failures were diligence-driven (Axial, Dead Deal Report: Unpacking 2025's Broken LOIs). [Verified, with the caveat that the sample is 75 self-reported deals from one deal network.]

The recurring set

1. Unassigned or unassignable IP. The most damaging finding in startup diligence (Crowley Law, Startup Due Diligence: What Investors Actually Look For). Three variants, in descending order of frequency:

• Founder pre-incorporation work. Code, designs, or a brand created before the entity existed, never contributed in writing. Cheap and fast to fix if the founder is cooperative and still around — and a negotiation at the worst possible moment if a departed co-founder wrote part of the codebase and left badly.
• Contractor work with no assignment. In the US, a contractor owns what they create absent a written assignment. "Work made for hire" language alone is insufficient for most software, which is why assignment clauses say "hereby assigns" rather than "agrees to assign" (Chapter 11 §2).
• Prior-employer claims. Work done on a prior employer's time or equipment.

Time to fix: days if everyone signs; months to never if someone is hostile or unreachable.

2. Missing founder IP assignment specifically. Many founders never execute a Confidential Information and Invention Assignment Agreement with their own company. [Analysis] Every buyer's counsel checks this in the first week. Execute it at incorporation.

3. Worker misclassification. Contractors performing employee-like work create exposure for unpaid payroll taxes, unemployment insurance, workers' compensation premiums, benefits, and in some states penalties and interest. Buyers price this as a hard dollar liability plus a reserve. It also compounds with (1): a misclassified worker is likelier to have signed nothing about IP. Chapter 11 §8 covers the classification tests.

Time to fix: reclassification going forward is immediate; the historical exposure does not go away and generally has to be quantified, disclosed, and either escrowed against or resolved through a voluntary disclosure programme.

4. Unpaid sales and use tax. Post-Wayfair economic nexus means a SaaS company with customers in many states may have registration and collection obligations in states where it has no physical presence, and roughly half the states tax SaaS. Companies that never collected it owe the tax plus penalties and interest, and the liability generally does not expire in states where no return was ever filed — the statute of limitations typically never starts running. [Verified — this is settled state-tax mechanics; see Chapter 11 §10 for the nexus thresholds.]

Time to fix: voluntary disclosure agreements (VDAs) with each state typically take 3–9 months per state and usually limit the look-back to 3–4 years and abate penalties. Doing this while a deal is live is not realistic. This is the clearest argument for the 12–24 month runway.

5. Open-source licence violations. Copyleft licences (GPL, AGPL) impose obligations when software is distributed or, for AGPL, made available over a network. A buyer whose product strategy involves shipping or embedding your code will have its counsel run an automated scan. Finding AGPL-licensed code in the core of a proprietary SaaS product is, depending on the buyer, either a remediation cost or a walk-away. [Analysis] The remediation is usually rewriting or replacing the offending component — a real engineering project, not a paperwork fix. Chapter 11 §6 covers licence categories.

6. Customer concentration. Not a "problem" you can fix in diligence; it is a structural fact that changes price and structure. The 10% level is where it becomes a disclosure-grade issue; private equity buyers commonly draw a hard line around 15%; above 30% many institutional buyers decline entirely (Beancount.io, Customer Concentration Risk, May 2026). [Estimate] Practitioner sources put the resulting valuation impact at roughly 20–35% for concentrated versus diversified peers, and — more importantly — concentration shifts consideration out of cash-at-close into earn-out and escrow.

7. No clean financials. A company on cash-basis books with no reconciliation between the billing system and the general ledger cannot pass a quality-of-earnings review. Buy-side QoE providers commonly disallow 10–30% of the add-backs a seller proposes (Papermark, Quality of Earnings in 2026). [Analysis] At a 6x EBITDA multiple, each $100,000 of disallowed add-backs removes $600,000 of headline price. This is the highest-return preparation work available per dollar spent.

8. Undocumented key-person dependency. If the founder holds every enterprise relationship, knows the deployment process, and is the only person who can close a sale, the buyer is buying a job that the buyer cannot do. The structural consequence is not usually a lower headline price. It is a longer earn-out, a longer required employment term, and a larger share of consideration made contingent on you staying.

9. Cap table defects. Missing board consents for option grants, unexercised promises made verbally, stray SAFEs never entered into the ledger, option grants with no 409A support, missed 83(b) elections. Each one is small; collectively they slow diligence and create indemnity items.

10. Contracts that are unsigned, non-assignable, or have change-of-control terms. In an asset sale, most customer contracts require consent to assign — which means telling customers a sale is happening, which is exactly what a seller does not want to do before closing. Anti-assignment and change-of-control clauses are one of the strongest practical arguments for a stock purchase or merger structure.

A3. How long cleanup takes, and why you should audit 12–24 months ahead

[Estimate] These are practitioner-consensus ranges, not measured data. The point of the table is the shape, not the precision: the expensive problems are the ones that cannot be fixed inside a deal timeline.

[Analysis] The argument for a 12–24 month lead time is not that cleanup takes 12–24 months of work. It is that it takes 12–24 months of calendar — VDAs, accounting conversions, concentration reduction and management hiring all run on clocks you do not control. A founder who decides in January and goes to market in March carries every one of these into diligence.

The practical form of this is a mock diligence exercise: hire a transaction lawyer and an accountant to run the standard buy-side checklist against your own company, 18 months out, and produce a findings list. It costs a fraction of what the findings will cost you in a retrade.

PART B — Who the buyers are, and how they differ

Buyer type determines price, structure, diligence intensity, probability of closing, and probability of retrade — often more than anything about your company. Understanding the buyer's economics tells you what they will do.

B1. Strategic acquirers

What they are. Operating companies buying for a product, a customer base, a team, a technology, or the removal of a competitor.

What they pay for. Synergy — either revenue synergy (sell your product to their customers) or cost synergy (eliminate your G&A, your infrastructure, sometimes your team). A strategic can rationally pay more than a financial buyer because its cash flow from the asset is different from yours. This is why strategics set the ceiling in most competitive processes.

How they behave in diligence. Thorough, slow, and dangerous in a specific way: the diligence team includes people who compete with you internally. Product, engineering and sales leaders at the acquirer see your roadmap, pricing, churn and customer list, and if the deal dies that information does not come back. [Analysis] Stage the data room; hold customer names and detailed pricing until late.

Strategic processes also run through internal approval chains — corporate development recommends; a business-unit head, a CFO and often a board approve. A deal can die because the sponsoring executive changed roles. Always ask who the internal sponsor is and what approvals remain.

Retrade risk. Moderate. Strategics retrade less often than sponsors on price, because the approval they obtained internally was for a number and reopening it is awkward. But they are more likely to walk entirely if diligence surprises them, and more likely to restructure consideration (more earn-out, more retention) rather than cut the headline.

Speed. Slowest to first offer, moderate to close. Regulatory review is a real consideration only at larger sizes or in concentrated markets.

B2. Financial buyers — private equity

What they are. Funds buying control with a mix of equity and debt, targeting a 3–7 year hold and a return on exit. In the lower middle market, frequently buying a platform and then bolting on acquisitions.

What they pay for. Cash flow that supports leverage, plus a credible path to multiple expansion (usually: make the company bigger, so it sells at a larger-company multiple). They care enormously about EBITDA quality, margin durability, and whether the business can service debt.

How they behave in diligence. The most rigorous, the most professionalised, and the most numerate. A PE buyer will commission a buy-side quality of earnings, a market study, and often technology and insurance diligence. They know exactly which add-backs to disallow.

Retrade risk. Highest. This is not a moral observation; it is structural. A sponsor's return model is sensitive to entry multiple, their diligence is designed to find EBITDA discrepancies, and their institutional culture treats a post-LOI price adjustment as normal practice rather than bad faith. The Axial data — QoE EBITDA discrepancies at 21.3% of failures and renegotiation conflict at a further 14.7% (Axial, 2025) — is substantially a story about sponsor-led processes.

What they want from you afterwards. Usually: you stay, you roll a meaningful portion of your equity into the new entity (10–30% is common), and you run the business under a board they control. Rollover equity is not cash. It is a second bet on the same asset with someone else driving. Treat it as such.

B3. Search funds and ETA buyers

What they are. An individual or pair — often recent MBAs — who raise a small search budget from investors, spend 18–30 months looking for one company to buy, then run it themselves. "Entrepreneurship through acquisition."

The 2026 benchmarks from Stanford's periodic study, as summarised by practitioners: median first-acquisition purchase price $16.0 million, median EBITDA multiple 6.2x (down from around 7x in prior cohorts), target median revenue ~$8.1M, median EBITDA ~$2.5M at a 25% margin, around 30 employees, median 20 months from search start to close, and an average of 2.5 signed LOIs per searcher — meaning the typical searcher breaks roughly one and a half deals before closing one (Five Experts, Anatomy of a Search Fund Acquisition: Benchmarks from Stanford's 2026 Study). Notably, 63% of first acquisitions included seller paper — seller notes, earn-outs, or rollover equity. [Verified as reported; the underlying Stanford study samples self-reporting search funds, which is a survivorship-biased population.]

What they pay for. Stable, boring, cash-generative businesses with a retiring owner. They are usually financing with SBA 7(a) debt or search-fund investor equity plus a bank facility.

How they behave in diligence. Earnest, slow, and dependent on third parties. The searcher personally cannot close; their investors and their lender can both kill the deal. The Stanford data on why searcher deals die — due diligence discoveries 79%, valuation gaps 45%, insufficient investor support 40% — shows the third-party dependency clearly.

Retrade risk. High, but usually not out of strategy — out of necessity, because a lender re-cut the advance or the investor group balked.

[Analysis] For a seller, a searcher is a legitimate buyer with two specific costs: a longer, less certain process, and a very high likelihood that a meaningful slice of the price is seller-financed. If you want all cash at close and certainty, a searcher is rarely the best counterparty.

B4. Competitors

A special case of strategic buyers, worth separating because the information risk is at its maximum and the probability of a bad-faith process is non-trivial.

[Analysis] A genuinely interested competitor is often your best-priced buyer — the synergy and the defensive value are both real. A competitor who is not genuinely interested has been handed a free education. Mitigations: a strong NDA with a non-solicit, staged disclosure, a clean team for the most sensitive data, and — most important — never run a process where a competitor is the only bidder.

B5. Marketplaces and small-deal platforms

For businesses roughly under $5M, and especially under $1M, the transaction often happens on a marketplace rather than through a banked process. Empire Flippers, Acquire.com, Flippa, Quiet Light, Website Closers and BizBuySell dominate different slices.

What the marketplace data shows. Empire Flippers' public scoreboard reports cumulative sales of roughly $606M across 2,674 listings sold, an average of 125 days to sale, and an average sale at 95% of list price (Empire Flippers Scoreboard, accessed September 2026). BizBuySell's Q1 2026 report shows 2,345 closed transactions, a median sale price of $350,000, median revenue of $713,404, median cash flow of $165,256, and an average cash flow multiple of 2.7x (BizBuySell Q1 2026 Insight Report). [Verified as published. Severe survivorship bias: both datasets count listings that sold. Neither publishes the denominator of listings that expired unsold, which practitioner estimates put well below a 50% sale rate for marketplace listings generally.]

How marketplace buyers behave. Faster, less formal, far less diligence, and far more likely to be an individual with SBA financing or personal capital than an institution. Documents are lighter; escrow is a platform escrow rather than a negotiated indemnity structure; the seller pays a percentage commission. Retrade risk is moderate and usually driven by verification — buyers check read-only analytics and payment-processor data, and a gap between claimed and verified numbers reliably kills the deal.

B6. A comparison

[Estimate] Ranges are practitioner consensus assembled from the sources cited in this part.

PART C — Intermediaries

C1. The four options

Business broker. Main Street. Sub-$5M enterprise value, often sub-$2M. Listing-driven: the business goes on a marketplace or a broker network, buyers self-select. Commission-style fees, 8–12%, with a minimum. Limited process management, limited negotiation, rarely any financial modelling.

M&A advisor / boutique bank. Roughly $5M–$300M. Runs a structured process: prepares materials, builds a curated buyer list of dozens to hundreds, manages the auction, manages diligence, negotiates alongside counsel. Retainer plus a success fee, usually on a modified Lehman scale.

Investment bank. $50M+ and up. Same as above but with broader institutional reach, sector coverage teams, and often financing capability. Fees negotiated as tiered percentages, typically lower in percentage terms because deal sizes are larger.

Doing it yourself. Viable in narrow circumstances and dangerous outside them.

(Peony, M&A Advisor Fees: Lehman Scale, Retainers & Hidden Clauses, 2026 for the category boundaries and fee structures in this part.)

C2. Fee structures, concretely

The Lehman formula and its descendants

Classic Lehman (5-4-3-2-1): 5% of the first $1M of transaction value, 4% of the second, 3% of the third, 2% of the fourth, 1% of everything above $4M. On a $10M deal that is $200,000 — a 2.0% effective rate.

Double Lehman (10-8-6-4-2): 10% / 8% / 6% / 4% / 2% on the same tiers. On a $10M deal, $400,000 — 4.0% effective. Double Lehman, or a close variant, is the modern default in the lower middle market. Classic Lehman was designed in an era of different deal sizes and is now rarely used unmodified. (Peony, 2026)

Effective all-in rates by deal size [Estimate, from the same source]:

Retainers and work fees

Monthly work fees of $5,000–$10,000, or a fixed upfront retainer of $25,000–$75,000, are standard for M&A advisors. Approximately 72% of advisors credit work fees dollar-for-dollar against the success fee (Peony, 2026). If the engagement letter does not say the work fee is creditable, that is a negotiation point, not a fact of life.

Minimum fees

Around 67% of engagement letters include a minimum success fee, ranging from $50,000–$250,000 on small deals to $200,000–$600,000 on $5M–$30M deals (Peony, 2026). [Analysis] On a small deal the minimum, not the percentage, is your actual cost — which means the percentage you negotiated may be irrelevant. Model the minimum against your realistic price range before signing.

The tail

The tail (or residual) clause entitles the advisor to a fee if you sell within a defined period after the engagement ends, to a buyer covered by the clause. Advisors typically ask for 18–24 months. Two defences [Analysis, consistent with practitioner guidance]:

1. Cap the duration — 12 months is a reasonable ask.
2. Limit it to named buyers the advisor actually contacted, with a written list required within a short window (10 days) of termination. A tail that covers "any buyer" for two years means you cannot sell to anyone, including someone who walked in the door unaided, without paying.

The most expensive definition in the letter

"Transaction value." If it includes assumed debt, the full earn-out at maximum, rollover equity, retained real estate, and retention payments, the fee base can substantially exceed the cash you actually receive. [Analysis] The seller's version: transaction value means cash actually received by shareholders, with earn-out fees paid if and when the earn-out is paid, and rollover equity excluded or discounted. This single clause is frequently worth more than the headline percentage.

C3. When each makes sense

[Analysis] Rough thresholds, acknowledging that the boundaries are soft:

• Under ~$1M and a single obvious buyer already in hand: doing it yourself plus a good transaction lawyer is often correct. An advisor's minimum fee would consume too much of the proceeds relative to the value added.
• Under ~$2M with no buyer in hand: a marketplace or broker. The value is buyer reach, not negotiation.
• $2M–$10M: genuinely ambiguous. An advisor costs 4–6% all-in but creates competition, and competition is worth far more than 6% — the FE International analysis of comparable SaaS businesses shows "one interested buyer only" as a discount factor and "buyer competition (multiple bids)" as a premium factor (FE International, SaaS Valuation Multiples 2026). [Analysis] If you can credibly generate three or more interested parties yourself, self-running is defensible. If you cannot, the advisor pays for themselves.
• $10M+: use an advisor. The process complexity, the diligence load, and the negotiation asymmetry against a professional buyer all justify it, and the fee is a smaller share of proceeds.
• $50M+: use a bank with sector coverage.

The under-appreciated function of an intermediary is not price. It is process and insulation. The advisor is the person who makes the aggressive ask, absorbs the "no," and manages the pressure of a timeline — while you keep running the business. Founders who negotiate directly with buyers frequently damage the working relationship they will need post-close.

C4. How to evaluate one

Questions that produce useful answers:

• How many transactions have you closed — not worked on — in the last 24 months, in my sector, at my size? Ask for the list.
• What is your close rate on engagements taken? An advisor who takes every mandate and closes a third is running a volume business.
• Who actually does the work? Meet the associate who will run the model and the data room.
• How many buyers will you contact, and can I approve the list first? Approval rights are non-negotiable — you will have hard exclusions.
• What did your last three failed processes fail on?
• Fee mechanics: effective rate on my realistic range, minimum, tail, definition of transaction value, expense cap, retainer creditability.
• Are you a registered broker-dealer or relying on the M&A broker exemption? State requirements vary, and an unregistered intermediary on a securities transaction can create rescission risk. Ask your lawyer.

C5. Conflicts of interest

These are not hypothetical, and they are structural rather than a matter of individual integrity:

1. The success fee makes closing the objective. An advisor earning 3% is nearly indifferent between $20M and $19M ($30,000 to them, $1M to you) but very much not indifferent between $19M and no deal. At the margin they will push you to accept.
2. The fee base may include consideration you never receive. Insist that earn-out fees are paid only when the earn-out is.
3. Buyer relationships are repeat business. [Analysis] This rarely produces disloyalty; it reliably produces a softer negotiating posture than a one-shot player would take.
4. Staple financing and buy-side fees. Any economic relationship with the buyer requires disclosure and is usually worth declining.
5. The broker affiliated with the buyer. Occurs at the small end. Walk away.

[Analysis] None of this means don't use an advisor. It means the advisor is not your fiduciary in the way your lawyer is, and the person whose incentives are cleanly aligned with yours is the transaction lawyer on an hourly fee who gets paid whether or not you close.

PART D — Valuation, concretely

D1. The three earnings bases, and when each applies

Buyers do not have one valuation method. They have three, and which one applies to you is mostly a function of your size, because size determines whether the business can support management without the owner.

SDE — Seller's Discretionary Earnings. Net profit, plus the owner's full salary and benefits, plus owner perquisites, plus interest, taxes, depreciation, amortisation, and non-recurring items. The logic: a buyer who will personally run the business gets the owner's entire economic benefit. SDE applies below roughly $5M enterprise value, and dominates below $2M.

EBITDA. Earnings before interest, taxes, depreciation and amortisation — with a market-rate management salary deducted, because the business is expected to pay someone to run it. EBITDA applies from roughly $5M enterprise value upward.

Revenue multiple (ARR or TTM revenue). Applies when growth or strategic value dominates current profitability: venture-scale software, high-growth SaaS, and any asset a strategic buyer wants for reasons other than its P&L.

The SDE–EBITDA gap is a trap. The same business produces two different numbers, and the difference is the owner's salary. FE International's framing: the same business can show a 33% valuation difference depending on which basis is used (FE International, SaaS Valuation Multiples 2026). [Analysis] Sellers quote SDE multiples; buyers of $5M+ businesses quote EBITDA multiples. A seller who compares their "4x SDE" offer to a published "6x EBITDA" benchmark is comparing incompatible numbers. Always confirm which base a quoted multiple uses, and whether it is applied to trailing twelve months, last fiscal year, or a forward run-rate.

D2. Why headline "10x ARR" figures mislead at small scale

The "10x ARR" reference point comes from two places: public software company valuations in the 2020–2021 window, and venture round pre-money valuations, which are not transaction prices for the whole company.

Two corrections:

Correction 1: public comps have repriced. Median public enterprise SaaS traded at roughly 4.6x TTM revenue in August 2026, having bottomed around 3.2x in June 2026 — against 6.0x in July 2025, 16–18.5x through most of 2021 (Aventis Advisors, SaaS Valuation Multiples 2015–2026). FE International's read of the same repricing puts median public enterprise SaaS at 3.3x TTM revenue at the end of Q1 2026, down from 4.9x at year-end 2025 and 6.2x at year-end 2024, and attributes it substantially to investor concern about agentic AI and per-seat pricing models (FE International, 2026). [Verified as published; the two sources use different universes and dates, which is why they differ — take the direction, not the decimal.]

Correction 2: small private companies trade below public comps, not above. Aventis's private M&A dataset (2015–2026) by deal size, EV/Revenue medians:

(Aventis Advisors, SaaS Valuation Multiples 2015–2026) — overall private median 4.5x EV/Revenue (Q1 2.4x, Q3 8.1x) and 23.0x EV/EBITDA (Q1 12.8x, Q3 47.1x). [Verified as published. Note the discontinuity at $50M: that is the size at which institutional buyers and competitive processes enter, and it is the single clearest illustration in the data that process quality and buyer type move multiples as much as company quality.]

[Analysis] The practical implication for a founder with $1.5M ARR: the relevant benchmark is not 10x and not the public median. It is somewhere in the low-to-mid single digits of ARR, or a mid-single-digit multiple of SDE, depending on quality. Anchoring on 10x is the most common way a first-time seller wastes a year.

D3. Current multiple ranges by business type

All figures below are [Estimate] unless marked otherwise — they are practitioner and marketplace benchmarks, drawn from populations of completed sales, and therefore systematically exclude businesses that could not find a buyer.

Small SaaS (private, sub-$20M)

FE International's 2026 private benchmarks, by ARR band:

Top-quartile businesses — Rule of 40 compliant, 110%+ net revenue retention — reach 8x and above (FE International, SaaS Valuation Multiples 2026).

Profitable software sold on earnings

For software businesses where profitability rather than growth is the story, EBITDA multiples in the high single digits to mid teens are the realistic private-market range at lower-middle-market size, converging toward the Aventis private median of 23.0x EV/EBITDA only at larger sizes and higher growth. [Estimate] The search-fund cohort — which buys profitable, often software-and-services businesses at around $8M revenue and $2.5M EBITDA — paid a median 6.2x EBITDA in the 2026 Stanford cohort (Five Experts on Stanford's 2026 study). That 6.2x is a useful floor reference for what a non-strategic, leverage-funded buyer pays for a good small business.

Content sites and online media

Empire Flippers' scoreboard reports realised multiples of roughly 2.2x for typical listings, 2.3x for premium listings, 3.1x for $1M+ premium listings, and 1.2x for distressed, against annualised earnings, with an average of 125 days to sale at 95% of list price (Empire Flippers Scoreboard). [Verified as published — but note that marketplace brokers in this segment have historically quoted monthly multiples (a "40x" listing meaning 40 months of net profit, i.e. ~3.3x annual), so always confirm the denominator before comparing.]

[Analysis] Content sites are the segment most exposed to platform risk in 2026. A business whose traffic depends on a single search or social algorithm is being underwritten by buyers who have watched several such businesses lose most of their traffic in a single update. Expect lower multiples, heavier earn-outs, and more traffic-verification diligence than in any other category.

Agencies and services

[Estimate — this is the least well-sourced category in this part; published agency benchmarks are almost entirely self-reported by agency-focused brokers, and no equivalent of the SaaS datasets exists. Treat the directional points as sound and any specific number you encounter as unverified.] Agencies price on SDE or EBITDA, not revenue, and sit at the lower end of every range because of three structural features buyers dislike: project revenue rather than contracted recurring revenue, client concentration, and owner-embedded relationships. The premium factors are the mirror image — retainer-based contracts with notice periods, no client above 10%, a named service lead other than the founder on each major account, and a defensible specialisation. A generalist agency with project revenue and founder-led sales is among the hardest categories of profitable business to sell at all.

Marketplaces

[Analysis] Valued on take-rate revenue or contribution margin, with buyers focused on liquidity (match rate, time to fill), supply/demand concentration, and disintermediation leakage — transactions that start on-platform and settle off it. Strong network effects and low leakage attract strategic revenue multiples; concentration on either side prices closer to an agency.

Venture-scale software

Valued on forward revenue, growth and strategic fit. [Analysis] The price is set by what the asset is worth inside the acquirer, which can be disconnected from standalone economics — and this is the category where founder proceeds diverge most from the headline (Part I).

Small businesses generally, for calibration

BizBuySell's Q1 2026 data: median sale price $350,000 on median revenue of $713,404 and median cash flow of $165,256, at an average cash flow multiple of 2.7x (BizBuySell Q1 2026 Insight Report). [Verified as published.] [Analysis] This is what the actual middle of the American small-business acquisition market looks like: low single-digit multiples of owner earnings on businesses worth a few hundred thousand dollars. It is a useful antidote to reading only startup-ecosystem valuation content.

D4. What moves a multiple, and by how much

Ranked roughly by impact, based on the drivers FE International identifies plus the concentration data:

Upward:

Downward:

FE International offers a worked comparison that is worth restating because it isolates the effect: two businesses at $6M ARR and 22% growth — one with 108% NRR, 74% gross margin (services bundled in) and a 19% top customer cleared at 3.4x ARR; the other with 121% NRR, 81% software margin and under 6% concentration cleared at 5.8x ARR. A $14M headline difference from fundamentals rather than market conditions (FE International, 2026). [Verified as published; it is a composite illustration from the advisor's own deal experience, not a controlled study.]

D5. The Rule of 40

Definition. Revenue growth rate (%) plus profit margin (%) should total at least 40. Which margin — EBITDA, free cash flow, or operating margin — is not standardised, which is the metric's first weakness.

Its role in valuation. FE International reports companies meeting the Rule of 40 trading at 6.6x revenue versus 2.3x for those below — close to a 3x spread (FE International, 2026). Aventis's Q2 2026 public-company data shows a median Rule of 40 score of 26%, with only 15% of 46 reporting companies above the 40 threshold; the median company showed 11.8% growth plus 14.6% profit margin (Aventis Advisors, SaaS Valuation Multiples). [Verified as published.]

What it is actually good for. [Analysis] The Rule of 40 is a screening heuristic for growth-stage software, not a valuation model. Its genuine insight is that growth and profitability are substitutable — that a buyer should be roughly indifferent between 40% growth at breakeven and 10% growth at 30% margins. Its genuine use to a seller is as a communication device: if you clear 40, say so early and loudly, because only about 15% of public comparables do.

Where it breaks, and sellers should not lean on it:

1. It says nothing about retention. A company growing 40% while churning 30% of its base scores the same as one growing 40% with negative churn. The second is worth multiples of the first.
2. It is unstable at small scale. Going from $500k to $800k ARR is 60% growth. The Rule of 40 rewards it identically to $50M to $80M, which is a categorically different achievement.
3. The margin definition is unstandardised. A seller quoting an EBITDA-based Rule of 40 to a buyer who computes it on free cash flow is having a different conversation.
4. AI-era cost structures distort it. Inference costs sit in cost of revenue for AI-native products, so gross margins — and therefore the margin half of the rule — mean something different than they did for classical SaaS.

[Analysis] For a sub-$10M-ARR seller: track it, mention it if it flatters you, and do not build your valuation expectation on it. Retention and concentration will move your price more.

PART E — The process

E1. The sequence

A banked sell-side process runs roughly like this. Timings are [Estimate] practitioner norms; a self-run process to a single buyer compresses stages 1–5 into a phone call, which is exactly the problem.

Total: 6–12 months from launch to closing for a typical lower-middle-market process, on top of the preparation period. [Estimate]

E2. Teaser and CIM

The teaser is a one- or two-page anonymous summary: sector, size, growth, why the business is interesting, why it is for sale. It goes out before an NDA, so it must not identify the company. Its job is to get an NDA signed.

The CIM (Confidential Information Memorandum) is the 30–60 page document that does the selling: business description, market, products, customers (usually anonymised as "Customer A, 14% of revenue"), financials with normalisations, growth opportunities, management team, and a summary of the transaction process.

[Analysis] Two rules about the CIM that first-time sellers break.

First: everything in the CIM will be tested in diligence. A growth projection you cannot substantiate becomes a credibility problem later, and credibility problems become price reductions. The CIM should be aggressive in framing and conservative in fact.

Second: the CIM's normalised EBITDA is the number the whole deal anchors to. If your CIM shows $2.4M of adjusted EBITDA and the buy-side QoE concludes $1.9M, you are not negotiating from $2.4M any more — you are defending a 21% gap while the buyer decides whether you were careless or dishonest. Getting a sell-side QoE done before the CIM is written ($25,000–$50,000 for lower-middle-market, 3–5 weeks — Papermark, Quality of Earnings in 2026) is the single most effective defence against the most common cause of deal failure.

E3. NDAs

A seller's NDA in an M&A context should do more than protect confidentiality:

• Non-solicitation of employees for a defined period, covering both hiring and soliciting. Without it, a failed process is a recruiting exercise for the buyer.
• Non-solicitation of customers where the counterparty is a competitor.
• A standstill, in a process where a public company might use the information to make an unsolicited approach.
• Restrictions on who inside the buyer sees what — a "clean team" provision where a competitor is involved.
• Return or destruction of materials, with certification.
• No representation as to accuracy, protecting the seller from claims based on preliminary information.

[Analysis] Frequent buyers will seek a carve-out for general advertisements and unsolicited applications. That is reasonable. A carve-out for anyone who "responds to a recruiter" is not.

E4. Indications of interest

An IOI is a non-binding, usually one-to-three-page letter giving a valuation range, the proposed structure at a high level, the source of funds, the expected diligence and timeline, and the required approvals. Nothing in it binds anyone.

[Analysis] The IOI stage is where the seller's leverage peaks. With multiple IOIs in hand, you can push on structure — cash at close, earn-out size, escrow, exclusivity length — before you have chosen anyone. Every seller protection is cheaper to obtain at the IOI-to-LOI transition than at any later point. The temptation is to select on headline number. The right selection criteria are headline number, cash at close, certainty of funds, and speed — weighted roughly equally.

A useful discipline: ask every IOI-stage bidder to state in writing (a) cash at closing, (b) total escrow and holdback, (c) the earn-out and its metric, (d) the employment and non-compete terms they will require of founders, and (e) their conditions to closing including financing. Bidders who will not answer are bidding on a number they intend to revise.

E5. Management presentations

A half- or full-day session where the buyer meets the team. The seller's job is to demonstrate that the business is not the founder — the head of sales presents the sales section; the CTO presents technology. A founder who answers every question has just priced their own key-person discount. This is also where the employee-disclosure problem begins: the common approach is a very small group under retention agreements and confidentiality obligations, expanded only after signing.

E6. The letter of intent

This is the most important document in the transaction that nobody treats as important.

What is binding and what is not

A typical LOI is mostly non-binding — the price, the structure, the closing conditions, the employment terms are all expressly stated to be non-binding and subject to definitive documentation.

A small set of provisions is binding, and they are the ones that matter:

[Analysis] So the seller gives up something binding and real — the right to run a competitive process — in exchange for something non-binding and indicative. That is the trade. It is not unreasonable; a buyer will not spend $150,000 on diligence while you shop them. But it is the moment your leverage transfers, and it should be priced accordingly.

SRS Acquiom's data shows no-shop provisions in 91% of the private-target deals it studied (SRS Acquiom, 2026 M&A Deal Terms Study) — exclusivity is effectively universal at institutional deal sizes. [Verified.]

Why exclusivity favours the buyer

Once you sign:

1. Your alternatives evaporate. The other bidders move on, staff up other deals, and are not waiting for you in 90 days.
2. The buyer's costs are sunk but recoverable through price. They have spent money on diligence; a retrade converts that spend into value.
3. Your costs are sunk and unrecoverable. Legal fees, QoE fees, advisor retainers, and several months of your attention diverted from running the business.
4. The business often underperforms during the process — Axial attributes 8.0% of broken deals to exactly this (Axial, 2025) — which hands the buyer a legitimate reason to re-cut.
5. You have told people. Your co-founder, your spouse, sometimes your key employees. The psychological cost of walking away rises every week.

What to negotiate BEFORE signing the LOI

This is the highest-value list in the chapter. Every item below is negotiable at the LOI stage and effectively non-negotiable afterwards.

1. Exclusivity length, and an outside date. Ask for 45–60 days, not 90–120. Tie any extension to the buyer having completed specified workstreams. Include an automatic termination if the buyer proposes a price reduction outside defined circumstances.

2. Cash at closing, stated as a dollar number. Not "approximately $X million, subject to adjustment." The single most useful sentence a seller can get into an LOI is a specific cash-at-close figure.

3. Escrow and holdback: percentage, duration, and whether it is the exclusive remedy. Get all three in the LOI. "Escrow to be agreed" means "escrow to be dictated."

4. The indemnity architecture at a high level. Cap, basket type and size, survival period, and whether RWI will be used and who pays for it. A buyer who says "standard indemnity" in the LOI and then produces an uncapped fundamental-reps regime with a 10% general cap is not doing anything unusual, and you will have no leverage to change it.

5. Working capital: the target, and the methodology. See Part G. This is the most reliable way a seller loses money they thought they had. The LOI should state the target (or the formula for computing it) and say that it will be computed in accordance with the historical accounting practices used to prepare the financial statements the buyer relied on.

6. Earn-out metric, measurement period, and the protective covenants. If there is an earn-out, the metric and the operating covenants belong in the LOI. See Part H.

7. Your employment terms, comp, and title. Buyers routinely defer this and then present it as a fait accompli two weeks before closing, when your alternative is to blow up the deal.

8. Non-compete and non-solicit scope and duration. See Part G9.

9. Treatment of the option pool and unvested equity. Whether options accelerate, are assumed, or are cashed out net of strike — and whether the cost of that comes out of the purchase price.

10. Expense allocation on a broken deal, and whether there is any reverse break fee.

11. Conditions to closing, listed exhaustively. In particular: is there a financing condition? A buyer with a financing condition is a buyer whose lender can kill your deal. Financing constraints accounted for 10.7% of broken LOIs in Axial's 2025 sample.

12. Who the seller representative will be, and that their fees come out of the proceeds rather than the founder's pocket.

[Analysis] A seller who gets items 1–6 into the LOI has converted the single most one-sided moment in the transaction into something close to fair. A seller who signs a two-page LOI with a price and a 120-day exclusivity has given away the deal and does not yet know it.

PART F — Diligence

F1. What buyers examine

Confirmatory diligence after the LOI typically runs in parallel workstreams:

Financial / quality of earnings. Three to five years of monthly financials plus a trailing-twelve-month view, revenue recognition testing, customer cohort and churn analysis, normalisation of EBITDA, net working capital analysis, and a debt-like-items schedule (Papermark, Quality of Earnings in 2026).

Legal. Corporate records and board consents; capitalisation and every security ever issued; IP chain of title; material contracts including change-of-control and assignment provisions; employment agreements and classification; litigation and threatened claims; regulatory and licensing; insurance; real estate and leases.

Tax. Federal, state and local filings; sales and use tax nexus and exposure; payroll tax; R&D credits claimed; transfer pricing if international; the historical treatment of equity compensation.

Technology. Architecture review, code quality, security posture, open-source composition scan, technical debt, infrastructure cost, key-person dependency in engineering, incident history.

Commercial. Customer reference calls (usually late and with consent), market sizing, competitive position, pipeline verification.

HR. Org chart, compensation, benefit plans, key-person retention risk, union or works-council issues outside the US.

Insurance and environmental, where relevant.

SRS Acquiom's 2026 study identifies a general shift in diligence emphasis as one of its six headline findings (SRS Acquiom, 2026 M&A Deal Terms Study), and the general direction in recent years has been toward more intensive financial and technology diligence, partly because RWI underwriters require it.

F2. The data room

A virtual data room, indexed to the buyer's request list. Practical points:

• Stage it. Tier 1 (corporate, financials, contracts with names redacted) opens immediately; Tier 2 (customer names, detailed pricing, compensation, source code) later, and to a clean team only where the buyer is a competitor.
• Log everything. Access logs are evidence of what was disclosed — which matters for the disclosure schedules and for defending an indemnity claim.
• Disclose problems deliberately and early. [Analysis] A problem you disclose is a diligence item; a problem the buyer finds is a credibility event and a retrade lever. If you have a sales tax exposure, put the quantification and remediation plan in the data room in week one. You pay for it either way; concealing it costs you the number plus your negotiating position.
• Watch what "no such document exists" means. A buyer asking for something you do not have is telling you what they expect to exist.

F3. Quality of earnings

The QoE is the most consequential diligence workstream, because it produces the number the price is computed from.

Cost and timing [Verified, from Papermark, 2026]:

The add-back trap. Buy-side QoE providers commonly disallow 10–30% of the add-backs a seller proposes. At a 6x multiple, each $100,000 disallowed removes $600,000 of headline price (Papermark, 2026). [Verified as published.]

The eight adjustment categories buyers scrutinise: owner compensation normalisation, personal and discretionary expenses, non-recurring items, revenue recognition timing, related-party transactions, non-cash charges, pro-forma run-rate adjustments, and accounting policy changes.

[Analysis] The defensible add-back has three properties: it is documented with a source record, it is genuinely non-recurring or genuinely discretionary, and the buyer can see why the expense will not exist post-close. "One-time legal fees" that appear in each of the last four years are not non-recurring. A pro-forma run-rate adjustment for a price increase implemented two months ago is defensible if you can show the invoices; it is not if you can only show the pricing page.

The sell-side QoE is the counter-move. Commissioned before going to market, it lets you find and fix the discrepancies on your own time, and it gives the buyer's team a reconciled starting point rather than a blank sheet. It does not prevent a buy-side QoE, but it substantially reduces the gap between them — which is the gap that kills deals.

F4. How long diligence takes

[Estimate] 45–90 days of confirmatory diligence post-LOI is the norm for lower-middle-market deals, running in parallel with definitive-agreement drafting. Axial's broken-deal data shows average days under exclusivity before a deal broke ranging from 37 days (family offices) to 159 days (where a senior lender was the deciding party) (Axial, 2025) — which is a useful proxy for how long different buyer types take before reaching a conclusion, good or bad.

F5. Running a business while in diligence

This is the operational risk that founders most consistently underestimate, and it has a direct financial cost: business underperformance during the process accounted for 8.0% of broken LOIs in Axial's sample (Axial, 2025), and underperformance that does not break the deal still hands the buyer a retrade argument.

[Analysis] Practical countermeasures:

1. Designate a deal quarterback who is not the CEO if you possibly can — a CFO, a head of finance, or the advisor's associate. The founder should be in management presentations and negotiations, not assembling documents.
2. Front-load the document assembly. Most of the data room can be built during preparation, before a buyer exists. A seller assembling five years of contracts under a 60-day clock will do both jobs badly.
3. Budget the CEO's time honestly. [Estimate] Practitioner consensus is 30–50% of the founder's working hours during active diligence. Plan the quarter's operating goals around that, not despite it.
4. Protect the quarter. Make the forecast you gave the buyer conservative enough that you will beat it. A miss during diligence is worth real money to the buyer.
5. Decide the employee-communication plan before you need it. People notice. A rumour that runs ahead of you is worse than a controlled disclosure to a small group under retention agreements.
6. Do not stop selling or stop hiring. A business visibly in stasis is a business the buyer can re-underwrite downward.

PART G — Deal structure and the purchase agreement

This entire part is the territory where a transaction lawyer is mandatory. The paragraphs below describe what the terms are and which way they cut. They are not a substitute for counsel drafting and negotiating them, and nothing here should be used to decide anything.

G1. Asset purchase vs stock purchase vs merger

The three forms

Asset purchase. The buyer purchases specified assets and assumes specified liabilities. The selling entity survives, holds the cash, and is then wound up or distributed. The buyer chooses what it takes — which is precisely the point.

Stock (or equity) purchase. The buyer purchases the shares from the shareholders. The entity continues unchanged with all of its assets and all of its liabilities, known and unknown.

Merger. The target merges into the buyer or a subsidiary of the buyer, by operation of law. Liabilities transfer as in a stock purchase, but — the critical practical advantage — it does not require every shareholder to sign, only the statutory vote. For a company with fifty shareholders and a dozen inactive early investors, a merger is often the only workable structure. Dissenting shareholders get appraisal rights; SRS Acquiom's data shows 58% of mergers include an appraisal-rights-related closing condition (SRS Acquiom, 2026 M&A Deal Terms Study).

SRS Acquiom's 2026 dataset is dominated by mergers and equity purchases; the ABA's 2025 study found 21% asset purchases among its 139 deals (K&L Gates on the 2025 ABA Private Target Deal Points Study, December 2025). [Verified.] [Analysis] Asset purchases are disproportionately common at the small end, below the size range either study covers.

Why buyers and sellers want different structures

[Analysis] The tension is almost entirely tax and liability versus consent mechanics. The buyer wants the step-up and the liability shield; the seller wants one layer of tax and no consents. The result in practice is that the structure usually follows the consent and liability analysis, and the tax consequence is then engineered on top of it — through a Section 338(h)(10) election or an F-reorganisation (Part J).

The tax consequences, at a high level

Warning: this is the paragraph where getting it wrong is most expensive, and where a tax professional is not optional.

• Asset sale by a C corporation: two layers of tax. The corporation pays tax on the gain from selling the assets, then shareholders pay tax again on the distribution of the proceeds. This is why a C-corp seller resists an asset sale hard.
• Asset sale by an S corporation or LLC: one layer, but the character of the gain is split by asset class, and a meaningful share can be ordinary income rather than capital gain — depreciation recapture, inventory, receivables, and the portion allocated to a personal non-compete or consulting arrangement. The allocation of purchase price across asset classes is negotiated and reported on IRS Form 8594 by both parties, and it directly determines how much of your proceeds are taxed at ordinary rates.
• Stock sale: one layer, capital gain, and — critically — the only structure under which QSBS treatment is available (Part J2).
• Section 338(h)(10) election: legally a stock sale, taxed as an asset sale. Requires an S corporation (or a member of a consolidated group) target, a corporate buyer acquiring 80%+, and a joint election by buyer and seller. The seller is taxed on 100% of the gain even if selling less than 100% (PCE Companies, F Reorganization or 338(h)(10) Election).
• F-reorganisation: a "mere change in identity, form, or place of organization" under §368(a)(1)(F), commonly used to convert an S corporation into a structure where the buyer acquires LLC interests and gets a basis step-up, while the seller keeps stock-sale-like treatment. Fewer restrictions than 338(h)(10) (PCE Companies).

[Analysis] The negotiating point that first-time sellers miss: if the buyer wants a structure that costs you tax, that is a price term. A buyer asking for a 338(h)(10) election is asking you to pay more tax so they can get a depreciation stream. The conventional response is a gross-up — the buyer pays you enough extra to leave you tax-neutral. Your tax adviser computes the number. It is frequently six figures on a mid-size deal, and it is routinely left on the table by sellers who did not know to ask.

G2. Cash vs stock consideration

Cash is certain, immediately taxable (subject to installment treatment for deferred pieces), and liquid.

Acquirer stock is none of those things. Before accepting it, the questions that matter:

1. Is it public or private? Public stock has a price and, subject to lock-ups and securities-law restrictions, a market. Private acquirer stock has neither. It is valued at whatever the acquirer's last preferred round implied, which is not what common stock is worth, and it may be illiquid for a decade or forever.
2. What class? Common stock in a venture-backed acquirer sits behind that acquirer's entire preference stack. You are taking on somebody else's liquidation waterfall (Part I).
3. Lock-up and registration. How long until you can sell, and do you have registration rights or are you relying on Rule 144?
4. What is the exchange ratio mechanism? Fixed number of shares (you bear the price risk between signing and closing) or fixed value (the buyer bears it)? Is there a collar?
5. Tax. Some stock-for-stock exchanges qualify for tax-deferred reorganisation treatment; many do not. Getting stock you cannot sell and a current tax bill is the worst case, and it happens.
6. Diligence runs both ways. If you are taking meaningful stock, you are now an investor in the acquirer and should diligence them accordingly — financials, runway, cap table, preference stack.

[Analysis] A serviceable heuristic: treat private acquirer stock as worth somewhere between zero and 50% of its stated value unless you have done real diligence that justifies more, and negotiate the cash component on that basis. Founders who accepted paper at headline value in 2021 learned this expensively.

G3. Working capital adjustments and the true-up

This is the mechanism that most reliably surprises first-time sellers, and it is in almost every deal above marketplace scale. SRS Acquiom found 72% of deals maintain a separate purchase-price-adjustment escrow (SRS Acquiom, 2026 M&A Deal Terms Study). [Verified.]

How it works

Most deals are priced on a "cash-free, debt-free" basis with a normalised level of working capital delivered at closing. The logic: the buyer is paying for the business's earnings power, and it needs the receivables, inventory and payables balance that generates those earnings. If you strip the receivables out before closing, you have sold the same business with less of the machinery it runs on.

The mechanics:

1. The parties agree a target (the "peg") — usually an average of monthly net working capital over the prior 12 months.
2. At closing, an estimate of actual working capital is prepared, and the price is adjusted up or down against the peg.
3. Within 60–120 days after closing, the buyer prepares a final calculation. If actual working capital was below the peg, the seller pays the difference — typically out of a dedicated escrow. If above, the buyer pays the seller.
4. Disputes go to an independent accounting firm as expert, not arbitrator, with the scope limited to the disputed items.

Where sellers lose money

[Analysis] Four recurring failure modes:

The peg is set on a period that flatters the buyer. A seasonal business whose working capital is high in Q4 and low in Q2 will have a very different peg depending on the averaging window. A trailing-12-month average is the conventional answer; a "most recent quarter" peg on a seasonally high quarter is a quiet price cut.

The methodology is undefined, so the buyer's accountants define it. This is the big one. If the agreement does not say the calculation will be made in accordance with the same accounting principles, practices, methodologies and policies used in preparing the historical financial statements — with that language taking precedence over generic "in accordance with GAAP" — the buyer's team can apply a stricter reserve policy for bad debts, a different revenue cut-off, or a new accrual, and produce a shortfall out of nothing. SRS Acquiom reports that 39% of deals use a "worksheet approach" — an agreed sample calculation attached as an exhibit (SRS Acquiom, 2026). [Analysis] Insist on the worksheet. A worked example of the calculation as of a recent date, attached to the agreement, eliminates most of the ambiguity that produces disputes.

Items are double-counted as both working capital and debt. Deferred revenue is the classic fight in SaaS: is it a working capital liability, a debt-like item, or neither? If it lands in both the working capital calculation and the debt schedule, the seller pays twice. Accrued bonuses, deferred rent, and customer deposits create the same problem. SRS Acquiom notes 89% of deals exclude tax-related items from the working capital calculation (SRS Acquiom, 2026) — which is a specific instance of the general principle that what is in and out must be enumerated.

The adjustment is one-way or asymmetrically capped. A deal where the seller pays a shortfall without limit but only receives a surplus up to a cap is not a true-up; it is a discount.

What to negotiate

• A collar — no adjustment unless the variance exceeds some threshold (commonly 1–2% of the purchase price).
• Symmetry — the adjustment works in both directions on the same terms.
• The worksheet exhibit.
• The precedence clause: historical practices over generic GAAP.
• A short review period (60 days, not 120) and a clear dispute mechanic with the cost split.
• Explicit treatment of deferred revenue and every other item that could be classified twice.
• Cash and debt definitions that are exhaustive, not illustrative.

G4. Net debt and debt-like items

"Cash-free, debt-free" means enterprise value is adjusted to equity value by adding cash and subtracting debt. The fight is over what counts as debt. Buyers push to include as "debt-like": accrued but unpaid taxes, deferred compensation and bonuses, capital lease obligations, unfunded pension or PTO liabilities, deferred revenue, customer prepayments, accrued but unpaid transaction expenses, severance obligations, earn-out obligations owed to prior sellers, and any related-party payable.

[Analysis] Every item moved from "working capital" to "debt-like" is a dollar-for-dollar reduction in your proceeds, and unlike the working capital adjustment there is no peg to measure it against. This list should be negotiated as an exhaustive schedule in the LOI or early in drafting, not discovered in a markup two weeks before closing.

G5. Representations and warranties

Statements of fact about the company, made as of signing and usually again at closing. They allocate risk: if a representation turns out to be untrue, the buyer has a claim.

Fundamental representations — organisation and good standing, authority, capitalisation, title to shares, sometimes tax — survive longest, are capped highest (frequently at the full purchase price), and are non-negotiable in substance.

General or operational representations — financial statements, contracts, compliance with law, IP, employees, litigation, customers and suppliers — carry shorter survival and lower caps.

The negotiation levers on each representation:

• Knowledge qualifiers ("to the Company's knowledge") and how knowledge is defined — actual knowledge of named individuals, or constructive knowledge after reasonable inquiry. The difference is substantial.
• Materiality qualifiers, and then the materiality scrape, which reads materiality qualifiers out of the representations for the purpose of calculating damages (and sometimes for determining breach). The ABA's 2025 study found double materiality scrapes in 82% of deals, up from 69% (K&L Gates on the 2025 ABA Study). [Verified.] [Analysis] The scrape is strongly buyer-favourable and is now close to market standard — a seller resisting it entirely is unlikely to win, but the scope (damages only, versus breach and damages) is negotiable.
• Disclosure schedules. The exceptions to the representations. Building these carefully is the single most protective drafting exercise the seller does, and it is where the data room logs earn their keep.
• Sandbagging. Whether the buyer can claim for a breach it already knew about. A pro-sandbagging clause favours the buyer; an anti-sandbagging clause favours the seller; silence leaves it to the governing law, and Delaware's default has historically leaned pro-buyer. [Analysis] Ask your counsel; the answer is jurisdiction-specific.

G6. Covenants

Pre-closing covenants (only relevant where signing and closing are separated — the ABA's 2025 study found 97 of 139 deals had deferred closings): operate in the ordinary course, do not take specified actions without consent, cooperate on regulatory filings, give access for diligence, and the no-shop.

Post-closing covenants: non-compete, non-solicit, confidentiality, cooperation on tax matters, further assurances, and — where there is an earn-out — the operating covenants governing how the business is run during the earn-out period (Part H).

The MAC/MAE condition. A material adverse change or effect gives the buyer the right not to close. SRS Acquiom reports 54% of deals structured with both stand-alone and "back-door" MAC provisions (SRS Acquiom, 2026), the back-door version operating through the bring-down of representations rather than as a separate condition. The ABA's 2025 study added tracking for MAE definitions that include pre-signing facts or conditions (K&L Gates) — a notable buyer-favourable expansion. [Verified.]

[Analysis] MAE clauses are heavily carved out (general economic conditions, industry conditions, changes in law, the announcement of the transaction itself) and are historically very hard for buyers to invoke successfully in Delaware. Their real function is as leverage: a buyer who asserts a MAE is usually seeking a price reduction, not a termination.

G7. Indemnification: caps, baskets, survival

This is where the purchase price you agreed becomes the purchase price you keep.

The market data

SRS Acquiom (2,300+ private-target deals, 2020–2025, $569bn aggregate value) (SRS Acquiom, 2026 M&A Deal Terms Study):

ABA (139 deals, $25M–$900M, majority below $200M) (K&L Gates; ABA Business Law Today, December 2025):

[Verified. The divergence between SRS's ~46% RWI and the ABA's 63% is explained by sample: the ABA studies larger deals with publicly filed agreements; SRS covers a broader and smaller-skewed population. Read them as two different slices, not as contradicting each other.]

What the terms mean

Cap. The maximum the seller can be required to pay for breaches of general representations. Median 10% of deal value. Fundamental representations are usually capped much higher, often at the full purchase price. Fraud is typically uncapped, and the ABA now tracks fraud as a standalone indemnity as a new data point.

Basket. A threshold below which no claim can be made.

• A deductible basket: the seller pays only the amount above the threshold. Seller-favourable.
• A first-dollar (or "tipping") basket: once the threshold is crossed, the seller pays from dollar one. Buyer-favourable, and materially worse for the seller than it sounds.
• With 61% of baskets at 0.5% or less of transaction value, the threshold is low in either case — on a $20M deal, $100,000.

Survival period. How long the buyer can bring a claim. Median 12 months for general representations; fundamental representations typically 3–6 years or the statute of limitations. Note that a 12-month survival period is shorter than one full audit cycle, which is why buyers press for longer and why RWI (with 3-year and 6-year survival under the policy) is attractive to them.

Exclusive remedy. Whether indemnification is the buyer's only recourse, or whether they retain common-law claims. Sellers want exclusive remedy with a fraud carve-out; buyers want more.

Seller representative. In a deal with many shareholders, one party (often a professional firm such as SRS Acquiom or Fortis Advisors) is appointed to act for all sellers in escrow claims, working capital disputes, and earn-out administration. Their fees and an expense fund come out of the proceeds.

G8. Escrow, holdbacks, and representation & warranty insurance

Escrow norms are in the table above: present in 88% of deals, median 10% of transaction value, held typically 12–18 months, released in tranches.

RWI

An insurance policy — normally bought by the buyer, often paid for by the seller or split — that covers losses from breaches of the seller's representations. It substitutes for most of the escrow and most of the indemnity.

Current market terms (2026) (Montague Law, RWI in Middle-Market M&A, 2026):

Common exclusions: known matters identified in diligence, purchase price adjustments, earn-out disputes, covenant breaches, pension underfunding, certain environmental liabilities, transfer taxes, and forward-looking projections.

[Verified as published by a practitioner source; premiums and retentions vary by carrier, sector and deal.]

Why it has spread into the mid-market. [Analysis] Three reasons. Premium compression made it affordable at smaller sizes. Sellers — especially PE funds distributing proceeds — want a clean exit with no post-closing tail. And buyers get longer survival and a creditworthy counterparty instead of chasing individual sellers. The SRS data quantifies the effect directly: median escrow falls from 11.1% without RWI to 2.8% with it.

What RWI does for a founder seller. It converts a 10% escrow held for 12–18 months into roughly a 2.8% escrow, which for a $20M deal is about $1.6M more cash at closing. That is a large number. [Analysis] The trade-offs: the seller usually bears some or all of the premium and underwriting fee; known issues are excluded, so anything diligence surfaces remains the seller's problem and typically still requires a special indemnity or a separate escrow; and diligence becomes more intensive because the underwriter requires it.

Where RWI does not help. It is not available or not economic below roughly $20M EV in most cases, it does not cover the working capital true-up, and it does not cover earn-out disputes — which are the two mechanisms most likely to cost a lower-middle-market seller money.

G9. Non-competes and non-solicits after the FTC rule

Where the law actually stands, September 2026

The FTC's 2024 Non-Compete Rule is dead. A federal district court held that the FTC had exceeded its statutory authority and that the rule was arbitrary and capricious; the FTC voted on September 5, 2025 to dismiss its appeals and accept the vacatur (FTC press release, September 2025); and the Commission formally removed 16 CFR Part 910 from the Code of Federal Regulations effective February 12, 2026 (Federal Register, 91 FR, February 12, 2026). [Verified — primary sources.]

What this means practically: non-competes are governed by state law, as they were before 2024, and the FTC has shifted to case-by-case enforcement rather than a blanket rule. State law varies enormously — California, Minnesota, Oklahoma and North Dakota broadly prohibit employment non-competes; many states impose income thresholds, notice requirements, or garden-leave obligations. Chapter 11 §7 covers the employment side.

Why the sale-of-business context is different

This is the important point for a seller, and it was true even when the FTC rule existed: sale-of-business non-competes have always been treated far more permissively than employment non-competes, including in states that otherwise ban them. California — the strictest state — permits them by statute where a person sells the goodwill of a business or all of their ownership interest (Bus. & Prof. Code §16601 and related provisions). The FTC's 2024 rule likewise contained a bona-fide-sale exception. [Verified as a matter of black-letter law; the scope of §16601 has been litigated and the details matter enormously — consult California counsel if this applies to you.]

[Analysis] So a founder selling a company should expect to sign a non-compete, and should expect it to be enforceable. What is negotiable is scope, duration, and definition:

• Duration. Three to five years is common in a sale-of-business context and is materially longer than what would be enforceable in employment. Push for the shorter end, and for the clock to start at closing rather than at the end of employment.
• Scope of restricted business. The definition frequently drafts to "any business in which the Company or any of Buyer's affiliates is engaged." If the buyer is a large diversified acquirer, that can foreclose most of an industry. Negotiate it to the business as actually conducted by the target at closing.
• Geography. Should match where the business actually operated.
• The non-solicit. Employees and customers. A no-hire (as opposed to no-solicit) of employees is broader and increasingly scrutinised under antitrust principles in some jurisdictions.
• Tax consequence. Purchase price allocated to a personal non-compete covenant is generally ordinary income to the seller, not capital gain, and is amortisable by the buyer over 15 years. A buyer proposing to allocate a large amount to the non-compete is proposing a tax transfer from you to them. Flag it to your tax adviser.
• Carve-outs for passive investments and for activities you are already engaged in.

G10. Retention packages, and how they dilute the headline number

Buyers need the team to stay. The mechanisms:

• Retention bonuses paid at 12 and 24 months post-close, conditioned on continued employment.
• New equity grants in the acquirer, vesting over 3–4 years.
• Earn-outs conditioned on the founder's employment — which, note, create the tax problem described in Part J5.
• Deferred purchase price conditioned on employment, which is functionally compensation.

[Analysis] The mechanism that dilutes the headline number: the retention pool is frequently funded out of the purchase price rather than on top of it. A $30M deal with a $4M retention pool carved out of the price is a $26M deal for the shareholders, plus $4M of compensation contingent on people staying — and the founder's share of that $4M is taxed as ordinary income, not capital gain. The headline is $30M. The shareholder economics are not.

The questions to ask at LOI stage:

1. Is the retention pool inside or outside the purchase price?
2. Who decides the allocation — you or the buyer?
3. What happens to a participant terminated without cause? (Insist on acceleration on termination without cause or resignation for good reason.)
4. Does it replace, or sit on top of, the existing option pool's treatment?

G11. Purchase price versus what a founder nets

A summary of the deductions, in the order they hit:

Headline purchase price (enterprise value)
  − net debt repaid at closing
  − transaction expenses (advisory fees, legal, accounting, QoE, RWI premium)
  − working capital shortfall (or + surplus)
  − escrow and holdback (deferred 12–24 months, and at risk)
  − earn-out (deferred, contingent, often unpaid — see Part H)
  − retention pool (if funded out of the price)
  = Equity value distributed to the cap table
      − liquidation preferences (and participation)
      − option exercise mechanics
      = Common shareholders' proceeds
          × the founder's fully diluted percentage
          − taxes (federal, state, ordinary vs capital)
          = What the founder actually keeps

[Analysis] Part I works two of these through numerically. The gap between the top line and the bottom line routinely exceeds 50% for a venture-backed company and 25–35% even for a clean bootstrapped one.

PART H — Earn-outs, mechanically

An earn-out is deferred contingent consideration: part of the price is paid only if the business hits defined targets after closing. It exists because buyer and seller disagree about the future, and it resolves that disagreement by making the seller bear the risk.

H1. How common, how big, how long

SRS Acquiom (private targets, 2020–2025) (2026 M&A Deal Terms Study):

ABA (139 deals, $25M–$900M): earn-outs fell to 18%, from 26% in the prior study, and displayed "buyer-friendly features" (K&L Gates on the 2025 ABA Study). [Verified.]

[Analysis] Earn-outs are common but not universal, and they are much more common where the seller and buyer disagree about growth, where customer concentration is high, or where the founder is essential to the business. An earn-out proposal is diagnostic: it tells you which of your claims the buyer does not believe.

H2. What fraction actually pays out

This is the number sellers should internalise before agreeing to anything.

SRS Acquiom's own framing of its data: "closer to one out of five dollars gets paid across all deals with an earnout" (SRS Acquiom, Earnout and Milestone Trends in Private-Target M&A Deals). [Verified as published by the party that administers these payments.]

[Analysis] Roughly 20 cents on the dollar of contingent consideration, aggregate. That figure mixes life-sciences milestone earn-outs (which are genuinely binary and often fail) with revenue and EBITDA earn-outs in operating businesses (which pay partially more often), so it is not a clean estimate of any individual deal's odds. But it is the best published aggregate from the administrator with the largest dataset, and the direction is unambiguous.

The practical rule: an earn-out is not deferred purchase price. It is a lottery ticket with a negotiated face value. Evaluate the deal on cash at closing, and treat the earn-out as upside. If the deal is not acceptable without the earn-out, it is probably not an acceptable deal.

H3. Structures, and how each fails

Revenue-based (69% of earn-outs).

• Advantage to the seller: revenue is harder to manipulate than earnings. No allocation of the buyer's overhead, no discretionary accruals.
• Failure mode: the buyer changes the go-to-market. If the acquirer folds your product into a bundle, sells it through their channel at a different price, or reassigns your sales team, "your" revenue becomes unmeasurable or collapses. Revenue also says nothing about margin, so a buyer may resist it precisely because you can hit it by selling unprofitably.

EBITDA-based (65%).

• Advantage to the buyer: aligns with what they are actually buying.
• Failure mode, and it is severe: EBITDA is an accounting output, and after closing the buyer controls the accounting. Allocated corporate overhead, management fees, shared services charges, the cost of integrating onto the buyer's systems, purchase accounting effects, new hires the buyer insists on — each legitimately reduces EBITDA and each reduces your payment. An EBITDA earn-out without a fully specified calculation methodology and an exhaustive list of permitted and prohibited charges is close to unenforceable in practice.

Milestone-based (the "other" 23%).

• Regulatory approvals, product launches, named customer signings, technical integration.
• Failure mode: binary, and the achievement often depends on third parties — a regulator, a large customer — over whom neither party has control. The Delaware case law in H5 is dominated by this category.

Multiple structures (60% of earn-out deals). Tiered or staged targets — for example, a revenue gate plus an EBITDA gate, or annual tranches. [Analysis] Tiering with linear interpolation between thresholds is significantly better for a seller than a cliff. A cliff structure that pays $0 at 99% of target and $5M at 100% is an invitation to a dispute and, more often, simply to a miss.

H4. The two problems that determine outcomes

The measurement-period problem

The median earn-out runs 21 months. Two things go wrong in a 21-month window.

First, the business is no longer the business. From day one after closing it is being integrated — new systems, new pricing, new reporting lines, new compensation plans, new compliance requirements, sometimes a new name. Measuring "the acquired business's revenue" 18 months into that is an accounting fiction requiring careful definition.

Second, the period is long enough for the buyer's strategy to change and short enough that a single bad quarter is fatal. [Analysis] A three-year earn-out with annual tranches and catch-up provisions is far better for a seller than a single cliff at 21 months, because it diversifies across periods. Ask for annual measurement with a cumulative catch-up: if you miss year one but the cumulative two-year total hits the cumulative target, you earn the year-one payment.

The control problem

This is the whole game. After closing, the buyer controls the business. Every variable that determines whether you hit the target — pricing, headcount, marketing spend, product roadmap, sales compensation, which customers get prioritised, how overhead is allocated — is now a decision made by someone whose economic interest is served by your missing.

The interest is not always conscious or malicious. A buyer who genuinely believes the right move is to cut your sales team and bundle your product into theirs may be right about the strategy and still destroy your earn-out. The law generally does not require a buyer to run the business to maximise your earn-out unless the contract says so. Delaware courts have repeatedly held that the implied covenant of good faith and fair dealing does not fill gaps the parties could have addressed and did not.

H5. Disputes, and how they are litigated

Earn-out disputes are among the most-litigated provisions in M&A, and Delaware is where the case law lives.

The landmark recent decision: Johnson & Johnson v. Fortis Advisors, Delaware Supreme Court, January 2026. J&J acquired Auris Health for $5.75bn plus up to $2.35bn in earn-outs tied to FDA regulatory milestones. When the FDA required a different clearance pathway, no milestones were achieved. The shareholder representative sued.

Holdings (Arnold & Porter, Efforts and Earnouts: Lessons From the Delaware Supreme Court in Johnson & Johnson v. Fortis, March 2026):

• Commercially reasonable efforts: breach affirmed. The contract defined CRE by reference to efforts "commensurate with [J&J's] own priority medical devices." The court found J&J had improperly prioritised commercialisation and an internal competing product over achieving the milestones. The express, benchmarked efforts standard is what made this claim work.
• Implied covenant of good faith: reversed. The risk that the FDA might require a different regulatory pathway was foreseeable and contractually allocated to the shareholders. The implied covenant did not rescue them from a risk the contract had assigned.
• Fraud in the inducement: affirmed. J&J had represented "high certainty" of achieving a $100M milestone while concealing a patient death, an FDA investigation, and expected delays.
• Damages: post-trial judgment exceeded $1bn, remanded for recalculation.

[Verified.] [Analysis] The lesson pair is sharp and generalisable: an express, benchmarked efforts covenant produced a recovery; the implied covenant did not. Sellers who rely on general principles of good faith will lose. Sellers who negotiate specific covenants can win. Delaware also continues to allow extracontractual fraud claims to survive anti-reliance provisions in some circumstances, which is why asymmetric anti-reliance drafting matters.

Typical dispute mechanics. Most earn-out agreements send disagreements about the calculation to an independent accountant, and disagreements about conduct (breach of operating covenants, breach of efforts obligations) to litigation or arbitration. [Analysis] The distinction matters enormously: the accountant route is fast and cheap but cannot address the buyer's decisions, only the arithmetic. Make sure the agreement does not route conduct claims to the accountant.

H6. The protective provisions a seller should insist on

A practical list. Each of these must be in the LOI, not discovered in the draft agreement.

Metric and calculation

1. A fully specified calculation methodology with a worked example attached as an exhibit, using the same accounting principles and practices as the historical financials.
2. An exhaustive schedule of excluded costs: allocated corporate overhead, management fees, buyer-mandated shared services, integration costs, purchase accounting effects, the buyer's own transaction expenses, costs of buyer-directed initiatives, and any charge not incurred by the business pre-closing.
3. Revenue definition that survives bundling, repricing and channel changes — including an agreed allocation methodology if your product is sold in a bundle.

Operating covenants 4. An obligation to operate the business in the ordinary course consistent with past practice during the earn-out period. 5. An express efforts standard, benchmarked to something concrete — Fortis is the authority for why this works and general "good faith" does not. For example: efforts "no less than those the Buyer applies to its own products of comparable revenue and strategic priority," or a hard floor ("maintain sales headcount at no fewer than X and marketing spend at no less than $Y per year"). 6. Maintain separate books and records for the acquired business through the earn-out period. Without this, the calculation becomes unauditable. 7. An express covenant not to take actions with the purpose or effect of reducing the earn-out, and — following Fortis — an explicit prohibition on prioritising a competing product line. 8. No relocation, discontinuation, divestiture or material restructuring of the business without consent or an automatic acceleration.

Acceleration and protection 9. Acceleration on defined events: sale of the acquired business, sale of the buyer, termination of the founder without cause, resignation for good reason, or material breach of the operating covenants. This single clause is worth more than most of the others. 10. The founder's continued employment must not be a condition of the earn-out — both because the buyer can then simply terminate you, and for the tax reason in Part J5.

Process and enforcement 11. Quarterly reporting of the earn-out metric during the period, with the right to object contemporaneously rather than only at the end. 12. Audit and information rights — access to books, records and personnel. 13. Dispute mechanics that separate arithmetic from conduct, with conduct claims going to a forum that can award damages. 14. A fraud carve-out and asymmetric anti-reliance provisions preserving the seller's extracontractual fraud claims. 15. Interest on late payment, and security for the obligation where the buyer's creditworthiness is uncertain. 16. Linear interpolation between thresholds, not cliffs; annual measurement with cumulative catch-up, not a single terminal test.

PART I — The waterfall: how the price is actually distributed

I1. The order of payment

Money moves in a strict sequence. Each layer is paid in full before the next receives anything.

1. Transaction expenses — advisory or banking fees, both sides' escrow agent fees, seller's legal and accounting fees, QoE, RWI premium and underwriting fee, D&O tail policy, seller-representative fee and expense fund.
2. Debt repayment — bank debt, venture debt, SBA loans, capital leases, related-party loans. Usually paid directly at closing from the wire.
3. Purchase price adjustments — the working capital estimate at closing; the true-up later.
4. Escrow and holdbacks — withheld from the closing wire, released after the survival period.
5. Carve-out plan, if there is one — paid off the top, before the preference stack (see I4).
6. Liquidation preferences, in order of seniority, including accrued dividends.
7. Participation, where preferred stock is participating.
8. Common stock and option holders, pro rata, net of exercise price.
9. Earn-out, later, if earned, distributed through the same waterfall.
10. Taxes, paid by each recipient individually.

I2. Worked waterfall 1 — a clean bootstrapped sale

A two-founder bootstrapped SaaS business, incorporated as an S corporation, $3.1M ARR, $2.1M of adjusted EBITDA. Sold to a private equity platform as a stock purchase at a $12M enterprise value, cash-free and debt-free. Cap table: 10,000,000 fully diluted shares — Founder A 6,000,000 (60%), Founder B 3,000,000 (30%), employee options 1,000,000 (10%) with a weighted-average strike of $0.22.

From enterprise value to shareholder proceeds:

Distribution at closing. No preferred stock, so it is pro rata. Adding the $220,000 of aggregate option exercise price into the pot gives a per-share closing price of $0.9675:

Escrow release at 18 months. A sales tax exposure surfaces and the buyer claims $180,000 against the escrow. $1,020,000 is released:

Founder A's outcome.

What this example teaches. [Analysis]

• Even a clean deal loses about 11% between the headline percentage and gross proceeds — here $783,000, to debt, working capital, and fees, offset partly by the balance-sheet cash.
• A 60% owner of a $12M sale nets about 38% of the headline number. That is the good case.
• The S corporation forecloses QSBS entirely. Section 1202 requires stock in a domestic C corporation. Had this business been a C corp from inception, with stock held over five years and assets under the gross-assets ceiling, a substantial portion of the gain could potentially have been excluded from federal tax — a difference here of well over $1M. That is an entity-choice decision made at incorporation, years before anyone thought about selling. Part J2.
• The escrow claim is real money, and it came from a cleanup item. $180,000 of sales tax exposure — the exact category Part A said to resolve 12–24 months out.

I3. Worked waterfall 2 — a venture-backed sale where the headline lies

A Series C company that raised into a difficult market. Announced acquisition: "**$75 million*," comprising a $65M upfront enterprise value plus a $10M earn-out over 24 months. Structured as a merger.*

The preference stack, built over four rounds:

From enterprise value to distributable proceeds:

The waterfall, assuming escrow fully releases and the earn-out pays nothing:

Series C is participating, so it shares the residual with common. Series C holds 25% on an as-converted basis; common plus options hold 30%; Seed, A and B (45% as-converted — Seed 3%, A 12%, B 30%) take their preferences and do not convert, because conversion would pay them far less. Check this per class, not in aggregate. The conversion test runs separately for each round, and a single early round choosing to convert can change the common residual by an order of magnitude — had Seed held 5% rather than 3% here, it would convert, and Founder A's equity proceeds would rise from roughly $28,000 to roughly $426,000.

Founders hold 18% of the fully diluted company, employees 12%. The founders' share of the common pool is $42,545. Founder A, with 12% of the company, receives $28,364.

The headline is $75 million. The founder who owns 12% of the company receives about twenty-eight thousand dollars.

Why this happens. It is not fraud and it is not unusual. Four compounding effects:

1. $10M of the "$75M" is an earn-out that, on the aggregate evidence, is more likely than not to pay only partially or not at all (Part H2).
2. $5M of retention was carved out of the purchase price, not added to it — so the shareholders sold for $60M of distributable value while the press release said $75M.
3. $6M of venture debt was repaid off the top, before anyone else.
4. The Series C 1.5x participating preference with a cumulative dividend turned $15M of investment into a $24.9M claim — and because it participates, it takes a second bite of the residual.

The carve-out plan. This is why they exist. A board facing this waterfall knows that management has no economic reason to support the deal and every reason to resist it. So the board adopts a management carve-out plan — a fixed share of consideration paid before the preference stack:

Common receives nothing from the waterfall proper. Founder A's allocation from the carve-out pool, at 30% of the pool, is $1,320,720.

And here is the sting: carve-out payments to an employee are compensation, reported on a W-2, taxed at ordinary rates and subject to payroll taxes. At 37% federal, 5% state and ~2.35% Medicare, Founder A nets roughly $735,000 — plus whatever portion of the $5M retention pool they earn by staying two years, also ordinary income.

What this example teaches. [Analysis]

• Every dollar of liquidation preference is senior to every dollar the founder owns, and structured terms — participation, multiples, cumulative dividends — compound in the direction the founder does not want. See Chapter 7 §B13 on structured rounds.
• QSBS is worthless when there is no capital gain. Founder A may have held qualifying stock for six years; there is no gain to exclude.
• The founder's realised outcome came from a compensation plan, not from ownership — which means it was taxed at nearly double the rate, was contingent on the board adopting the plan, and gave the founder no negotiating leverage of their own.
• Carve-outs are discretionary and negotiated at the board level. A founder with a board seat and an informed lawyer can shape one. A founder who first learns about the waterfall three weeks before closing cannot.

I4. Carve-out plans, in general

What they are. A plan adopted by the board — usually before or at the start of a sale process — allocating a fixed percentage of transaction consideration to named employees, paid off the top, ahead of the preference stack.

Why they exist. Two reasons. First, an incentive problem: management whose equity is worthless has no reason to work hard on a sale, and in a deal requiring a shareholder vote, common holders may have the practical ability to obstruct. Second, a retention problem: the buyer needs the team, and a team facing a zero payout will leave.

Typical sizing. [Estimate] Practitioner ranges are commonly 5–15% of total consideration, with the plan documented as a bonus plan rather than an equity instrument. Existing preferred holders must consent, because the money comes out of their recovery.

What founders should know. [Analysis]

• Negotiate it early, before the deal is far enough along that the investors' consent becomes leverage over you.
• It is compensation for tax purposes, not capital gain. The after-tax value is roughly 60–65 cents on the dollar, not 75–80.
• Double-trigger it: payable on closing, and protected if you are terminated without cause before the payment date.
• Understand that it is discretionary until adopted. There is no legal entitlement to one.

Chapter 7 §B9 and §B13 cover cap tables and structured rounds; this chapter's contribution is only the mechanics of how the money moves at the end.

PART J — Taxes on exit

This part is general information, at a deliberately high level, and is not tax advice. A qualified tax professional is mandatory — not advisable, mandatory — and must be engaged before the letter of intent is signed. The tax outcome of a sale is frequently worth more than the last full turn of negotiation on headline price, and unlike price, most of it is determined by decisions made months or years earlier that cannot be undone at closing.

J1. Capital gain versus ordinary income

The core distinction. Long-term capital gain on assets held more than a year is taxed at preferential federal rates (0/15/20%), plus the 3.8% net investment income tax where applicable. Ordinary income is taxed at rates up to 37%, and compensation income also attracts payroll taxes.

The consideration that is capital gain: proceeds from selling stock; the capital-gain portion of an asset sale.

The consideration that is (or risks being) ordinary income:

• Retention bonuses and carve-out plan payments — compensation, W-2, payroll taxes. This is what made Founder A's $1.32M in Part I3 worth about $735,000.
• Purchase price allocated to a personal non-compete covenant.
• Consulting or employment payments post-close.
• Earn-out payments recharacterised as compensation — see J5.
• In an asset sale: depreciation recapture, inventory, and accounts receivable allocations.
• Option exercises on NSOs, and ISOs disqualified by a cashless exercise at closing.

[Analysis] The recurring negotiation is that buyers are often indifferent or mildly favourable to structures that convert your capital gain into ordinary income — an amortisable non-compete allocation, a retention package instead of price. This costs you roughly 15–20 points of tax and costs them nothing. Price it, and ask to be grossed up.

J2. QSBS / Section 1202, including the 2025 changes

Potentially the single largest tax benefit available to a US startup founder, and among the most commonly forfeited by accident.

The basic requirements (all must be satisfied; this is a summary, not a checklist to rely on):

• Stock in a domestic C corporation — not an S corporation, not an LLC.
• Acquired at original issuance from the company, for money, property, or services.
• The corporation's aggregate gross assets must not have exceeded the ceiling at any time before and immediately after issuance.
• The corporation must use at least 80% of its assets in a qualified trade or business — which excludes most professional services, financial services, hospitality, farming, and certain other categories.
• The seller must hold the stock for the required period.
• It applies to a sale of stock, not a sale of assets by the corporation.

The 2025 changes under the One Big Beautiful Bill Act. For stock acquired after July 4, 2025 ("Post-Enactment QSBS") (Perkins Coie, Significant Changes by the One Big Beautiful Bill Act to the Qualified Small Business Stock Provisions of Section 1202):

[Verified — law-firm client alert describing enacted statutory changes.]

Grandfathering matters. Stock issued on or before July 4, 2025 stays under the old rules — $10M cap, 100% only at five years. A company with stock issued in both periods must track the blocks separately. [Analysis] This creates a real planning asymmetry inside a single cap table: a 2023 founder and a 2026 employee holding stock in the same company face different rules.

Where founders lose QSBS by accident:

• Choosing an S corporation or LLC at incorporation, or converting to one later. Part I2's founder lost over a million dollars to this decision.
• Selling in an asset sale. §1202 applies to gain on the sale of stock. If the buyer insists on an asset purchase or a 338(h)(10) election, QSBS is generally unavailable. This is why the structure decision must be modelled for tax before the LOI.
• Redemptions. Certain redemptions of stock by the corporation, within defined windows around issuance, can disqualify the stock.
• Selling too early. Under the new tiered rules a 3-year sale gets 50%. And the partial exclusions are worth less than the headline suggests: the non-excluded portion of gain on stock held three or four years is taxed at the 28% §1202 rate, not the usual 15%/20% long-term capital gains rates (see Chapter 11 §10), not 100%. Under the old rules, at 4 years and 11 months, it gets nothing.
• Failing the active-business or gross-assets tests at some point in the history without anyone noticing.
• State non-conformity. [Analysis] A number of states — California most prominently — do not conform to §1202, so a founder may exclude gain federally and still owe full state tax. Several other states have partial or no conformity. Verify your state's treatment with a professional; the list changes and general articles about it go stale.

Documentation. [Analysis] QSBS status must be provable years after the fact — capitalisation records, gross-asset computations at each issuance, and evidence of the qualified trade or business. Many companies obtain a QSBS attestation from counsel. Assemble it while the records exist, not at closing.

J3. Stock sale versus asset sale treatment

Summarised from Part G1, restated in tax terms:

(PCE Companies, F Reorganization or 338(h)(10) Election.)

[Analysis] The purchase price allocation in an asset sale — negotiated between the parties and reported by both on IRS Form 8594 — determines the ordinary/capital split and is genuinely adversarial. The buyer wants allocation to short-lived depreciable assets and the non-compete; the seller wants allocation to goodwill. Do not treat the allocation schedule as an administrative afterthought.

J4. 83(b) and early exercise, as they surface at exit

Consequences of equity-compensation decisions made years earlier show up at closing. Chapter 11 §9 covers the mechanics; here is what breaks at exit:

• A missed 83(b) election on restricted stock. Without a timely 83(b) (filed within 30 days of the grant), the holder is taxed as the stock vests, at the then-fair market value, as ordinary income — so a founder whose company appreciated during vesting has been accruing ordinary income and has no long-term capital gain on that portion. There is no fix at closing. [Analysis] This is among the most expensive 30-day deadlines in startup life.
• The QSBS holding period runs from the acquisition date. Early exercise with an 83(b) generally starts that clock at exercise; exercising at closing starts it that day, which means no QSBS. This alone is a reason for employees to early-exercise where economically sensible — but it also means an unexercised option holder cashed out in the deal gets ordinary income, not capital gain, on the entire spread.
• ISO disqualifying dispositions. A cashless exercise-and-sell at closing does not satisfy the ISO holding periods, so the spread becomes ordinary income. Essentially every option holder cashed out in an acquisition is in this position. [Analysis] Founders regularly promise employees "capital gains treatment" on their options and are wrong.
• AMT from a prior-year ISO exercise that can no longer be recovered if the stock is now worth less than at exercise.
• Unvested equity treatment. Whether unvested shares or options accelerate, are assumed, or are cancelled — and the tax character of each — is a purchase agreement term, not a default.

J5. Installment sales and earn-out taxation

The threshold question is characterisation: is the earn-out purchase price or compensation? Three factors drive it (Frost Brown Todd, Making the Most on the Sale of Your Business: An Owner's Tax Considerations on Earnouts):

1. The closer the earn-out period aligns with an employment term, the more it looks like compensation — ordinary income.
2. The closer your post-closing salary is to market rate, the more the earn-out looks like purchase price — capital gain. If you are paid below market and given a large earn-out, the IRS has an argument.
3. If the buyer must pay the earn-out even if your employment terminates, that supports purchase-price treatment.

[Verified.] [Analysis] This is the second, tax-based reason for the Part H6 rule that the earn-out must not be conditioned on continued employment. The first reason was that the buyer could otherwise simply fire you. The difference in outcome is roughly 17 points of federal rate plus payroll taxes.

Once it is purchase price, earn-outs are generally contingent payment sales under §453, with gain recognised proportionally as payments are received, using a gross profit percentage of (total price − basis) ÷ total price. Where the maximum price is not determinable, the default approach recovers basis ratably over 15 years (Frost Brown Todd). Two complications:

• Imputed interest. Under §483 and the OID rules, a portion of each deferred payment is recharacterised as interest — ordinary income to you.
• §453A interest charge. If deferred obligations exceed $5 million, an additional interest-like charge applies to the deferred tax.

The open transaction doctrine — reporting no gain until basis is fully recovered — is available only where the value of contingent payments "cannot be reasonably ascertained," a standard the IRS treats as rare and extraordinary, with 20% accuracy-related penalty exposure if asserted without substantial authority (Frost Brown Todd). [Analysis] Do not plan around it without a formal opinion.

Electing out of the installment method. A seller can elect out and pay tax on the full value of the deal, including the estimated earn-out, in the year of sale. [Analysis] Occasionally sensible — for example, if you expect tax rates to rise, or if you are trying to fit gain into a QSBS exclusion year — and occasionally catastrophic, if you pay tax on an earn-out that never arrives. Given the ~20-cents-on-the-dollar aggregate payout rate from Part H2, electing out to pay tax upfront on an earn-out is usually a bad bet. Model it.

J6. State tax on exit, including residency

[Analysis, with strong caveats — this is the area where general articles are least reliable and professional advice is most necessary.]

• Residency determines the taxation of gain on intangibles. Gain from selling stock is generally sourced to the seller's state of residence, which makes residency the single biggest state-tax variable in a stock sale.
• Changing residency before a sale is heavily audited. High-tax states — California above all — audit departures aggressively, examine domicile rather than day-counting alone, and look at where your home, family, vehicles, professional licences, doctors, clubs and voter registration are. A move executed in the months before a signing, while the founder keeps a house and family in the old state, is a poor fact pattern.
• Sourcing differs for business assets. In an asset sale, a state can source gain to where the assets and business activity were located, regardless of residency. Moving does not help.
• Deferred consideration follows the year it is received, but not always the residency of that year. States apply different rules to installment payments earned while you were a resident. Earn-out payments received after a move may still be taxed by the former state.
• State QSBS conformity varies, and a number of states — again, California most notably — do not conform, so a fully excluded federal gain can be fully taxable at state level.
• Trailing nexus and composite filings. An entity with multi-state activity may create filing obligations for shareholders in states they never lived in.

The practical rule: [Analysis] If state residency planning is part of your thinking, engage a state and local tax specialist years before a transaction, not months. A move undertaken for genuine reasons well in advance is defensible. A move undertaken to avoid tax on a deal already in motion frequently is not, and the cost of losing that argument includes penalties and interest.

J7. The point worth repeating

[Analysis] Consider the founder in Part I2. Negotiating the headline price from $12M to $12.5M — a hard-fought 4% — would have added roughly $356,000 of pre-tax proceeds at her ownership level, or about $254,000 after tax. Incorporating as a C corporation rather than an S corporation at the outset, and qualifying for §1202, could have been worth more than a million dollars on the same transaction. The entity-choice decision made in a weekend at incorporation was worth several times the hardest negotiation of the sale. That asymmetry is the argument for engaging tax advice early, and for treating the structure sections of the LOI as price terms.

PART K — Closing, and after

K1. Signing versus closing

These are distinct events, and in the ABA's 2025 sample 97 of 139 deals had a deferred closing — only 42 signed and closed simultaneously (K&L Gates on the 2025 ABA Study). [Verified.]

At signing, the purchase agreement is executed, the representations are made, and the parties become bound — subject to the closing conditions.

Between signing and closing, the parties satisfy conditions: regulatory clearances (antitrust filings where thresholds are met, sector-specific approvals, foreign investment review), shareholder approval and the running of appraisal periods, third-party consents on contracts that require them, buyer financing, and the bring-down of representations.

The gap is the seller's period of maximum vulnerability. The seller runs the business under restrictive interim covenants, has announced nothing publicly but has told key people, and remains exposed to a buyer who can invoke a MAE or a failed bring-down. [Analysis] Minimise the gap where possible; where it is unavoidable, negotiate a tight outside date, a narrow MAE with generous carve-outs, and — if the buyer has a financing condition — a reverse break fee.

K2. Transition services

A transition services agreement (TSA) is most common in carve-outs, where the seller keeps shared functions the sold business still needs — payroll, IT, ERP access, finance — for a defined period post-closing, at a defined price.

[Analysis] For a founder selling a whole company, a formal TSA is less common; the equivalent is the founder's own employment or consulting agreement. The negotiating points are the same either way: define the scope precisely, define the duration, define what happens on extension, and define who bears the cost. An open-ended obligation to "provide reasonable transition assistance" is a commitment of unbounded personal time, and it is usually unpaid.

K3. Integration, and the founder's first year

What usually happens, in rough order. [Analysis, drawn from practitioner accounts rather than measured data]

Months 0–3. Announcement, systems migration, HR onboarding, the first budget cycle under new rules. Your team asks about compensation, titles and job security, not the product. Your authority is ambiguous: nominally you run things, but purchasing, hiring, pricing and legal now need approvals you never previously needed.

Months 3–9. Your product enters a portfolio roadmap, competing for engineering resources against products owned by people who have been there a decade. Decisions you made in an afternoon take six weeks. The people you hired for autonomy start leaving.

Months 9–18. You discover whether the acquirer's plan is the plan they described in diligence. Often it is not — not because anyone lied, but because the strategy or the sponsoring executive changed.

The specific asymmetries founders report:

• Loss of decision rights is faster and more complete than expected. The organisational chart says you run the business unit; the approval matrix says otherwise.
• Your economic incentive and your team's diverge. You have been paid. They have unvested acquirer equity and a retention bonus. They notice.
• The internal sponsor who bought you may leave. When they do, your advocacy inside the acquirer disappears.
• Earn-out and role conflict. Doing what the acquirer wants is often not what maximises your earn-out (Part H4).

K4. The retention-cliff pattern

The pattern is consistent enough to plan around: retention money vests in tranches, and departures cluster immediately after each tranche.

• Retention bonuses at 12 and 24 months, and new acquirer equity on a standard 4-year schedule with a 1-year cliff, together create sharp incentive discontinuities. A founder whose meaningful money is fully paid at month 24 has no financial reason to stay at month 25.
• [Estimate] Practitioner and industry sources commonly cite attrition of roughly a third of acquired employees within the first year, with senior and technical staff over-represented — but the empirical literature on this is uneven, definitions of "acquired employee" vary, and the figure should be treated as an order of magnitude rather than a measurement.
• [Analysis] For founders specifically, the modal outcome is departure at or shortly after the end of the earn-out and retention period. This is not failure; it is what the incentive structure is designed to produce. The failure case is the founder who did not plan for it and finds themselves, at month 25, with no role, no company, and no next thing.

What to do about it, practically: negotiate the earn-out so it does not depend on your employment (Part H6, Part J5); negotiate severance and good-reason resignation triggers that protect your retention money if the acquirer changes your role; and decide in advance, before closing, what you intend to do at month 25.

PART L — When deals die

L1. Base rates

From LOI to closing: practitioner synthesis puts the figure at roughly 65–70% of signed LOIs reaching closing, with the lower middle market at the worse end of that range (Kadenwood Group, What Percentage of Signed LOIs Close?, citing Axial's 2025 dead-deal analysis and Capstone's middle-market index). The same synthesis puts 60–70% of launched sell-side processes producing a transaction at all, implying 30–40% of mandates die before ever reaching an LOI. [Estimate — this is an aggregation of practitioner sources, not a measured population study, and the definitional boundaries ("launched," "signed LOI") are soft.]

Compounding these: [Analysis] roughly 0.65 × 0.65 ≈ 40–45% of companies that engage an advisor and go to market end up closing a sale. That is a materially lower number than most first-time sellers assume, and it is the single most useful piece of expectation-setting in this chapter.

Corroboration from the buy side: the Stanford 2026 search-fund cohort averaged 2.5 signed LOIs per searcher before completing one acquisition (Five Experts on Stanford's 2026 study) — implying, from that buyer type, an LOI-to-close rate around 40%.

L2. Why they die

Axial's 2025 analysis of 75 broken lower-middle-market deals (Axial):

[Verified as published; sample is 75 self-reported deals from one deal network, so treat the precision with caution and the ranking as robust.]

Read it this way: [Analysis] roughly 47% of failures are diligence-driven, and both of those top two categories are preparation failures rather than market failures. A quality-of-earnings gap means the real number was always the real number; the seller simply did not know it before going to market. Adding the 14.7% retrade category — most of which follows from a diligence finding — puts something like 60% of deal failure in the category "the buyer found something the seller should have found first."

L3. Retrading

A retrade is a post-LOI reduction in price or worsening of terms, justified by a diligence finding.

Why it works. Every element of Part E6 — exclusivity has removed your alternatives, your costs are sunk, the buyer's costs are recoverable through price, and you have psychologically and often socially committed to the outcome.

What makes a retrade legitimate versus opportunistic. [Analysis] Legitimate: a genuine, material, previously undisclosed adverse finding — a real tax liability, a customer that has given notice, an EBITDA overstatement. Opportunistic: a small finding used as a pretext for a large adjustment, or an adjustment that arrives late with a short fuse, or one whose magnitude bears no relationship to the finding's quantified cost.

Defences, in order of effectiveness:

1. Sell-side QoE before going to market. Eliminates the most common trigger.
2. Proactive disclosure of known problems, quantified, in week one. A disclosed issue is a priced issue.
3. Anti-retrade language in the LOI — the buyer may not propose a price reduction except for matters not disclosed and exceeding a stated materiality threshold, with exclusivity terminating automatically if they do.
4. Short exclusivity. Sixty days, with extension conditioned on the buyer certifying no known basis for adjustment.
5. A real or credible alternative. The only genuine source of leverage. [Analysis] This is why running a competitive process matters even when you have a preferred buyer, and why some sellers keep a second bidder warm at low intensity through exclusivity, to the extent the no-shop permits.
6. Insist that any adjustment be quantified against the specific finding. "Your finding is $200,000 of exposure; the adjustment is $200,000, not $2 million."
7. Willingness to walk. Which requires having decided in advance, in writing, at what number and on what terms you would rather not sell.

L4. What a failed process costs

Direct costs. [Estimate] On a lower-middle-market deal: advisor retainers and reimbursable expenses of $50,000–$150,000 (the success fee is not owed), seller legal fees of $75,000–$250,000 depending on how far the definitive agreement progressed, sell-side QoE of $25,000–$50,000, plus the seller-side portion of any RWI underwriting fee already incurred. A realistic total is $150,000–$400,000 for a deal that died in confirmatory diligence.

Indirect costs, which are larger:

• Six to twelve months of founder attention, at 30–50% of working hours during the active phase. The opportunity cost is the growth that did not happen.
• Operational drift. The 8.0% of deals that die from underperformance during the process are a visible subset of a larger phenomenon; many surviving businesses come out of a failed process with a soft quarter.
• Employee damage. People who learned of the sale and then watched it fail draw conclusions. Some leave. Retention agreements signed in anticipation of a closing may now be liabilities without a corresponding benefit.
• Information transferred to a competitor, permanently, if the buyer was one.
• Market taint. [Analysis] Advisors talk. A business that went to market and failed within the last 12–18 months carries a question that the next set of buyers will ask. The conventional cooling-off period before re-launching is 12–24 months — long enough that the failure reason has demonstrably been fixed and a fresh set of financials exists.
• The tail clause. If you terminate the advisor, the tail may still capture a sale to anyone they contacted, for 12–24 months.

[Analysis] Netting all of this, a failed process is genuinely expensive — plausibly 1–3% of enterprise value in direct costs and considerably more in indirect ones. That cost is the strongest possible argument for the two recommendations this chapter opened with: do the cleanup years ahead, and negotiate the protections before signing the LOI. Almost everything that kills a deal is visible before the deal starts, to a seller who goes looking.

Closing note

The asymmetry between a first-time seller and a professional buyer is real, structural, and not going to be eliminated by reading a chapter. But it narrows substantially with three things, none of which requires expertise the founder does not have:

1. Time. Cleanup that takes 12–24 months of calendar cannot be compressed. Starting early is the only version of this that works.
2. Professionals engaged before the LOI. A transaction lawyer and a tax adviser, engaged at the indication-of-interest stage rather than after the LOI is signed, will change the economics of the deal by more than they cost. This is the point at which their advice is mandatory rather than optional, and it is also the point at which founders most often skip it because "nothing is binding yet."
3. Competition. One bidder is a negotiation. Three bidders is a market. Every protective term in this chapter is easier to obtain when the buyer knows you have somewhere else to go.

And one piece of arithmetic worth carrying: evaluate every offer on cash at closing, after fees, after the preference stack, after tax. The headline number is a marketing figure. The wire is the transaction.

Sources

All URLs accessed September 15, 2026 unless otherwise noted.

Deal-terms studies (primary — the authoritative sources on what actual agreements contain)

• SRS Acquiom, 2026 M&A Deal Terms Study (2,300+ private-target acquisitions closed 2020–2025, $569bn aggregate value) — https://learn.srsacquiom.com/rs/664-KLJ-520/images/SRSAcquiom-2026-Deal-Terms-Study.pdf?version=1
• SRS Acquiom, Six Key Findings from the 2026 M&A Deal Terms Study — https://www.srsacquiom.com/our-insights/deal-terms-study-key-findings/
• SRS Acquiom, Earnout and Milestone Trends in Private-Target M&A Deals — https://www.srsacquiom.com/our-insights/ma-earnout-milestone-trends/
• K&L Gates, 2025 ABA Private Target Mergers & Acquisitions Deal Points Study (139 deals, $25M–$900M), December 19, 2025 — https://www.klgates.com/2025-ABA-Private-Target-Mergers-Acquisitions-Deal-Points-Study-12-19-2025
• American Bar Association, Business Law Today, Announcing the ABA's 2025 Private Target Mergers & Acquisitions Deal Points Study, December 2025 — https://www.americanbar.org/groups/business_law/resources/business-law-today/2025-december/aba-2025-private-target-mergers-acquisitions-deal-points-study/

Legal and regulatory (primary)

• Federal Register, Revision of the Negative Option Rule, Withdrawal of the CARS Rule, Removal of the Non-Compete Rule To Conform These Rules to Federal Court Decisions, effective February 12, 2026 — https://www.federalregister.gov/documents/2026/02/12/2026-02866/revision-of-the-negative-option-rule-withdrawal-of-the-cars-rule-removal-of-the-non-compete-rule-to
• Federal Trade Commission, Federal Trade Commission Files to Accede to Vacatur of Non-Compete Clause Rule, September 5, 2025 — https://www.ftc.gov/news-events/news/press-releases/2025/09/federal-trade-commission-files-accede-vacatur-non-compete-clause-rule

Law-firm client alerts and case analysis

• Arnold & Porter, Efforts and Earnouts: Lessons From the Delaware Supreme Court in Johnson & Johnson v. Fortis, March 2026 — https://www.arnoldporter.com/en/perspectives/advisories/2026/03/lessons-from-johnson-and-johnson-v-fortis
• Perkins Coie, Significant Changes by the One Big Beautiful Bill Act to the Qualified Small Business Stock Provisions of Section 1202 — https://perkinscoie.com/insights/update/significant-changes-one-big-beautiful-bill-act-qualified-small-business-stock
• Frost Brown Todd (now FBT Gibbons), Making the Most on the Sale of Your Business — An Owner's Tax Considerations on Earnouts — https://fbtgibbons.com/making-the-most-on-the-sale-of-your-business-an-owners-tax-considerations-on-earnouts/
• Crowley Law LLC, Startup Due Diligence: What Investors Actually Look For — https://www.crowleylawllc.com/due-diligence-founder-guide/
• Montague Law, Reps and Warranties Insurance for Middle-Market M&A: 2026 Pricing, Retention, and When the Premium Earns Its Keep — https://montague.law/blog/rwi-florida-middle-market-ma-2026/

Valuation and market data

• FE International, SaaS Valuation Multiples 2026: Private Deal Benchmarks — https://www.feinternational.com/blog/saas-valuation-multiples
• Aventis Advisors, SaaS Valuation Multiples: 2015–2026 — https://aventis-advisors.com/saas-valuation-multiples/
• BizBuySell, 2026 First Quarter Insight Report — https://www.bizbuysell.com/news/bizbuysell-2026-first-quarter-insight-report/
• Empire Flippers, Scoreboard — https://empireflippers.com/scoreboard/
• Beancount.io, Customer Concentration Risk: The 10% Revenue Threshold, May 11, 2026 — https://beancount.io/blog/2026/05/11/customer-concentration-risk-10-percent-revenue-threshold-business-valuation-loan-capacity-negotiating-leverage-guide
• Five Experts, Anatomy of a Search Fund Acquisition: Deal Benchmarks from Stanford's 2026 Study — https://fiveexperts.com/resources/anatomy-of-a-search-fund-acquisition-deal-benchmarks-from-stanfords-2026-study

Process, diligence, fees and deal failure

• Axial, Dead Deal Report: Unpacking 2025's Broken LOIs (75 broken lower-middle-market transactions) — https://www.axial.net/forum/dead-deal-report-unpacking-2025s-broken-lois/
• Kadenwood Group, What Percentage of Signed LOIs Close? The Data on Broken Deals — https://kadenwoodgroup.com/perspectives/one-in-three-lois-never-closes
• Papermark, Quality of Earnings in 2026: 8 Adjustments and the Add-Back Trap — https://www.papermark.com/blog/quality-of-earnings
• Peony, M&A Advisor Fees: Lehman Scale, Retainers & Hidden Clauses (2026) — https://www.peony.ink/blog/ma-advisor-fees
• PCE Companies, F Reorganization or 338(h)(10) Election — https://www.pcecompanies.com/resources/f-reorganization-or-338h10-election

Cross-references within this library

• Chapter 7, Funding and Financing — §A16 (private equity), §A17 (acquisition as a financing and exit event), §B6 (term sheets), §B9 (valuation), §B13 (down rounds and structured rounds), §B15 (secondaries), §B16 (exit planning).
• Chapter 11, Legal, Operations and Founder Sustainability — §2 (IP assignment), §6 (open-source compliance), §7 (non-competes), §8 (contractor classification), §9 (equity compensation mechanics), §10 (tax, including QSBS and sales tax nexus), §22 (when a professional is required), §B9 (selling the company).
• Chapter 13, Choosing an Endgame — the strategic question of which exit to pursue. This chapter assumes that decision has been made.

Note on source quality

The two deal-terms studies (SRS Acquiom and the ABA) are the only sources here drawn from large samples of actual executed agreements, and they are the ones to trust where any source conflicts with them. Law-firm alerts on statutory changes and court decisions are reliable on what the law says and should still be verified against the primary text. Valuation and fee benchmarks from advisory firms and marketplaces are the best public data available for private-company transactions, but they are drawn from those firms' own completed deals — they are advertising as well as data, they are survivorship-biased, and their methodologies are generally not published. Every multiple in Part D should be read as a range with wide error bars, not a price.

Nothing in this chapter is legal, tax, or financial advice.