When partners separate, everything funnels into one question: what is the business worth?
Every buyout mechanism from the exits article, every death arrangement from the last one—all of them eventually hand off to a valuation. And here's what you must understand about that moment: it is the worst possible time to be inventing a valuation method. The seller needs the number high; the buyer needs it low; each can hire an expert who will oblige. Between honest methods, a business can be "worth" figures that differ by multiples—and each partner will sincerely believe the method favoring him is the objective one. Which is why the agreement decides the method now, while neither of you knows which side of the table you'll one day sit on. That ignorance is precious: behind that veil, you'll pick what's fair rather than what's favorable, because fair is the only strategy when you don't know your position.
Our guide lays out seven ways. Let me walk through them the way I do at the table—what each one is, and whom it fits.
Option 1 — Wholesale value. Reckon it as if someone were opening a new company and buying up all the goods, equipment, or software at wholesale prices. That total is the business's worth; debts—owed by you and owed to you—stay out of the count. This is liquidation logic: what would the stuff fetch? It fits inventory-heavy businesses and separations where nobody claims the enterprise has value beyond its assets. Its virtue is that it's hard to argue with; its limit is that it values the body and ignores the soul—no goodwill, no customer base, no momentum.
Option 2 — Balance sheet, corrected. Take the balance-sheet number and add back everything that was written down only for tax purposes but still has real value. That correction is the whole point. Books are kept to minimize taxes: equipment depreciated to zero that runs perfectly, or amortized assets still earning daily. An uncorrected balance sheet systematically understates value and hands the buyer a discount at the seller's expense.
Option 3 — A multiple of profits. This is the method the business world runs on, and the guide walks through it step by step. Average the last three years of profits, but first add back what the raw books hide: the owners' salaries, interest on loans, personal expenses run through the business (cars, tzedakah, and the like), and amortization on assets that hold value. Divide by three for a true average annual profit. Multiply by an agreed factor—higher for stability and scale—and add the corrected net value from the balance sheet. The factor schedule is agreed upon in advance, stepping up through profit tiers: one multiple for profits under $500,000, rising by tiers to another for profits above $10 million. Why does the factor grow with profits? Because a business earning $8 million is usually more stable—with systems, teams, and diversified customers—than one earning $400,000 that may be a one-man show wearing a company's name. The buyer of the small business is buying more risk per dollar of profit, so each dollar is priced lower. Settling the schedule now spares you the two-expert theater of dueling multiples later.
Options 4 and 5 — Three experts. Each side appoints one expert; the two experts jointly pick a third; all three appraise the business independently. If two agree, that's the value. If all three differ, there are two variants: average the highest and lowest (e.g., 100, 90, 130 → 90 + 130 = 220 ÷ 2 = 110), or—the guide's recommended default—take the middle number (e.g., 100, 90, 130 → 100). The middle-number rule has a quiet elegance: an extreme appraisal, high or low, simply can't win—so neither side gains anything by hiring a gunslinger. The mechanism itself disciplines the experts toward honesty. That's what a good clause does: it makes the fair move and the smart move the same move.
Option 6 — Two experts and a decider. Each side appoints an expert; if they can't agree, each submits a final number, and the dispute-resolution path (detailed in the next article) must pick one of the two—no compromise in between. This is "baseball arbitration," and the no-splitting rule is the engine: when the decider may split the difference, each side exaggerates to drag the midpoint; when he must choose one number as a whole, the more reasonable number wins—so both sides race toward being reasonable. The rules are designed so that honesty is the winning strategy.
Option 7 — Market price. Value the business based on what comparable businesses in the same industry actually sell for. This is effective where a real market exists—such as for franchises, routes, or practices with brokers and published comparables. Where no such market exists, "market price" simply relocates the argument to which businesses count as comparable.
And the combination. Methods can be mixed across the business's life: one method for the first few years (for example, the corrected balance sheet, while profits are too young to be meaningful), another as profits stabilize (the multiple), and a third above a certain size (three experts, when the stakes justify the process). Write out which method applies when—a valuation clause that grows with the business.
The number is settled. Now—how does it actually get paid? Skip this, and the fight relocates: "I'll pay you out over ten years." "You'll pay me this month." Terms that cannot be met help no one.
Two axes decide everything. Does the partnership end immediately, or do you remain partners until the money is fully paid? Staying partners until full payment is the seller's best security—if the buyer misses payments, the seller is still an owner with an owner's rights. Ending immediately gives the buyer maximum clarity—the business is his, and the debt is personal, secured by a lien on the shares. And is payment one-time or in installments? There are four clean patterns: the full amount now, with the partnership ending only at full payment; the full amount within 90 days, with the partnership over and the debt secured by a lien; a 25% deposit with quarterly installments over two years, remaining partners as each payment buys out a growing slice; or the same 25%-and-quarterly schedule with the partnership ended at once and the balance as a lien-secured personal debt. Match the terms to reality: an operating business generating cash can support installments out of profits—often the only way a working partner can afford to buy out a money partner. A clean break with outside financing may be worth more to both parties than a higher number paid slowly.
And what if a payment is missed? The answers range in severity, and the guide makes you look at each squarely. Harshest: all payments made so far are deemed a gift, and the seller returns as a partner exactly as before. This powerfully motivates on-time payment—but I always point out its trap: if the business turns bad, the buyer can deliberately default, tell his former partner "congratulations, you're a partner again," and unload half the losses on him. A default clause that the defaulter wants to trigger is a defective clause. Gentler: the seller returns as a partner proportionally—his stake reflecting only what remains unpaid. Or, the unpaid balance converts to a loan under a heter iska at an agreed rate. Or, the whole matter goes to the dispute-resolution path. In every case, include a grace period—a written notice and a set number of days to cure—so one late wire during a bank holiday doesn't detonate the deal.
I'll close where I began. On the day of separation, the two of you will not agree on the business's worth—expecting otherwise is expecting the seller and buyer of the same object to share one mind. The kindness—and it is a kindness—is done today: choosing the method, the terms, and the default rules while it is all hypothetical and you would both sincerely answer "whatever's fair." Write that answer down while you mean it. The day one of you stops meaning it is exactly the day the page starts working.
The Partnership Guide (דער שותפות מדריך) walks you and your partner through every question in this series, step by step, in plain language—and turns your answers into a complete, ready-to-sign partnership agreement. Answering the questions is free; you only pay when your agreement is ready to print.
Start your partnership agreement →