This post uses hypothetical scenarios for illustrative purposes only. It does not describe any actual client, transaction, or representation, and is not legal advice.
A common 2026 lower-middle-market pattern looks like this: a founder has built a profitable services or software company with recurring revenue, a small team, and customer relationships that mostly run through her. A larger platform, often private equity backed, wants to buy it, and it wants her to come along. The letter of intent has a purchase price on page one. Somewhere behind it sits a second document that gets a fraction of the attention: the employment agreement she will sign at closing. It has a base salary, a bonus tied to performance metrics, at-will employment, and a severance provision measured in months of base pay. She and her counsel spend the negotiation arguing about the number of months. That is the wrong negotiation, and this post is about why.
The severance number is the wrong negotiation
Start with what actually moves at closing. The customers move. The contracts get assigned. The employees become the buyer’s employees. The founder’s systems, vendors, insurance, and banking relationships get shut down or absorbed. None of that is designed to be reversed, and none of it reverses cheaply. If the relationship ends at month fourteen, the buyer’s exposure is whatever severance it agreed to pay: a defined, budgeted, insurable number. The founder’s exposure is different in kind. She has to stand the infrastructure back up, re-paper her vendor and customer relationships, and ask customers who already moved once to move again. Customers do not like moving twice, and some will not.
That asymmetry is the risk, and it is measured in the enterprise value of the business, not in payroll. Severance priced against base salary cannot cover it, however many months get added, because the two numbers are not in the same unit. Picture a founder whose company produces high six figures of owner earnings a year. A few months of base salary does not begin to cover a rebuild, and it does nothing at all for the customers who do not come back. So the useful question at the LOI is not how many months, but what happens to the business, the customers, and the founder’s freedom to operate if the buyer ends the relationship early. Those are the terms that allocate the transition risk, and they belong on the same page as the price. Like the cash-free, debt-free definitions, they get harder to move with every week that passes after the LOI is signed.
Cause and Good Reason do the real work
Whatever severance exists is usually payable only on a termination without Cause or a resignation for Good Reason, so the definitions decide whether the protection is real. Buyer-side drafts tend to define Cause expansively: any material breach of policy, any failure to perform duties satisfactorily, and, in the version founders should refuse, failure to meet performance expectations. If missing a metric is Cause, the severance evaporates at precisely the moment it was supposed to matter, and the buyer holds a walk-away right it can exercise for free.
The founder-side asks are familiar from executive agreements, and they matter more here, because this executive just sold the buyer her company. Cause should require willful misconduct or a material breach that survives written notice and a cure period. Performance should not appear in the definition at all. Good Reason should let her resign with full severance if the buyer materially reduces her compensation or duties, changes her reporting line, relocates her, or, in a services business, reassigns the customers she brought. Good Reason is what keeps the buyer from constructively ending the relationship while calling it something else. In most cases the definitions are worth more than the months.
Metrics nobody can evaluate before day one
The bonus and any rollover or earnout are typically tied to numbers: revenue retention, new business, utilization, EBITDA of the acquired unit. The founder is asked to commit to those targets in a business she has never seen from the inside. She does not yet know the buyer’s customer base, its typical budgets, its pricing discipline, its mix of work, or how much of the platform’s overhead will be allocated against her unit. The buyer knows all of it. Committing to a number under that information gap is not a business judgment. It is a guess, and the counterparty already knows the answer.
Delaware’s earnout cases show where this ends up. In Johnson & Johnson v. Fortis Advisors (Del. Jan. 12, 2026), the Delaware Supreme Court reviewed a Court of Chancery award of roughly a billion dollars to former Auris Health stockholders whose milestone payments depended on the buyer’s efforts. The Supreme Court reversed part of the award but upheld the core of it: J&J was held to the “commercially reasonable efforts” standard the contract actually defined, measured against how J&J treated its own priority products. Just as important, the Court held that the implied covenant of good faith could not fill gaps the parties could have foreseen and chose not to address. The lesson for a founder-employee is that when the buyer controls the levers behind a contingent payment, only the written standard protects her. The implied covenant will not rescue a vague one.
Practically, that means a ramp period before metrics apply, targets set by mutual agreement after a defined onboarding window rather than in the LOI, metrics limited to things within the founder’s control, an efforts or resourcing commitment from the buyer, and a written rule for what happens to unearned amounts on a termination without Cause. The earnout playbook applies to the bonus plan almost line for line, because a bonus tied to buyer-controlled numbers is a small earnout with fewer protections.
The non-compete turns transition risk into a lockout
Here is the piece founders miss most often. She will typically sign two restrictive covenants: one as seller, in the purchase agreement, and one as employee, in the employment agreement. Under Florida Statutes section 542.335, a covenant given by the seller of a business is presumed reasonable at up to three years and unreasonable beyond seven, and buyers routinely draft toward the long end. Since July 1, 2025, Florida’s CHOICE Act adds a second layer for a “covered employee” whose base salary exceeds twice the annual mean wage of the county, which describes nearly every founder employment agreement: a covered non-compete can run up to four years, and a court must enter a preliminary injunction unless the employee proves by clear and convincing evidence that an exception applies.
Now run the scenario forward. The buyer terminates without Cause at month fourteen. Severance runs for a handful of months. The sale-of-business covenant, and possibly a covered non-compete, run for years. The founder who needs to rebuild is barred from serving the market she built, and often from calling the customers who would follow her. Transition risk has become a lockout, and it was created by two documents that were each negotiated as if the other did not exist. The fix belongs in both: the covenants should shorten or fall away on a termination without Cause or a resignation for Good Reason, the founder should keep the right to serve customers she brought if the buyer ends the relationship early, and, in a services business, the agreements should spell out how those relationships transfer back. That is a term the buyer will resist. It is also the only term that puts the buyer’s incentives on the same side of the table as the founder’s.
Anything you build belongs to the buyer unless the paper says otherwise
Founders who plan to keep building, and most of them do, should read the invention assignment and work-for-hire language as carefully as the price. The default in nearly every buyer-drafted agreement is that anything conceived during employment that relates to the buyer’s business belongs to the buyer. A founder who plans to develop tools, software, or a product line during the relationship, and imagines taking it with her if things end, will usually find she cannot. The cure is a schedule of prior inventions, a carve-out for technology developed on her own time and resources, and, where the buyer wants to use what she builds, a license rather than an assignment. The CIIA agreement is where that fight lives, and it is far easier to win before closing than after.
The middle path: affiliate first, integrate later
None of this means the deal should not happen. It means the sequence deserves scrutiny. Sometimes the buyer’s real interest is the founder’s expertise and her ability to build a group inside the platform. Sometimes it is her customer relationships. Those interests support very different structures, and it is worth learning which one is driving before choosing.
If expertise is the draw, the founder can keep her company running and enter a consulting, affiliation, or joint-venture arrangement: a defined time commitment, a fixed retainer that supplies the predictability she was buying with the salary, an agreed split on business each side brings, a written confidentiality and non-solicitation protocol, and an express carve-out for her pre-existing and independently developed technology. Layer on an option, exercisable by either side after twelve to twenty-four months, to move to a full acquisition on a pre-agreed formula. Both parties learn what they cannot learn in diligence: whether the customers, the budgets, the mix of work, and the people actually fit. Neither side takes on the irreversible cost first. The platform buyer loses some lock-in and the founder gives up the guaranteed salary, so this is not a free option for either of them. But it converts an all-or-nothing bet into a staged one, and staged bets are usually what both sides should have wanted from the start.
What to ask for at the LOI
If the deal proceeds as a straight sale plus employment, the founder-side asks are these. First, Cause narrowed to willful misconduct or uncured material breach, with performance out of the definition. Second, Good Reason that covers diminution of duties, compensation, reporting line, and reassignment of the customers she brought. Third, metrics deferred to a post-onboarding reset, with a ramp and a buyer efforts commitment. Fourth, restrictive covenants that step down or terminate on a without-Cause termination or Good Reason resignation, plus a customer return mechanism. Fifth, an IP carve-out for prior and independently developed technology. Sixth, severance sized against the cost of reversal rather than months of base pay, or a make-whole tied to the purchase price if the relationship ends early without Cause. None of these can guarantee that the relationship works. What they can do is make sure that if it does not, the two sides land where they agreed to land, rather than where the buyer’s form put them.
If you are a Florida founder weighing an offer that combines a sale of your company with an employment agreement, feel free to reach out to our firm manager, Magda, at Magda@montague.law, or fill out our contact form. Mention you read this post.
