How should I think about giving employees equity or profit share?

Answer the cash question before the ownership question

Almost every founder who asks this is really asking how to keep someone valuable. Ownership feels like the strongest answer available, and it is the hardest one to reverse.

The sequence that works is the opposite of the instinct. Get the salary right. Then get the bonus right. Only then, and only in specific cases, consider ownership. Firms that run that sequence in reverse end up with a legacy ownership structure that no longer reflects who is creating the value, and no clean way out of it.

Why sweat equity is the wrong currency

Do not award ownership on the basis of effort. Sweat equity is impossible to value when calculating ownership percentages. What percentage of the firm should go to a great rainmaker rather than a good one? The question cannot be answered, which means any number chosen is arbitrary and will later be resented by someone.

Effort has a market price and the market sets it objectively. If someone is responsible for project management, pay them what a project manager is paid. That is what sweat equity is worth, and salary is where it belongs.

The cleanest way to value ownership at the start of a firm is contributed capital. If a firm needs one million dollars to launch and one partner puts in five hundred thousand, another three hundred thousand and a third two hundred thousand, the split is fifty, thirty and twenty. It is clean and it is defensible. When a firm starts, it is almost impossible to value any other way, because there are no clients, no revenue, no intellectual property and no profit. One hundred per cent of zero is zero.

And contributions to wealth creation change over time, which is the deeper reason fixed equity arrangements are discouraged. Many firms have been ruined by disagreements over equity, and the disagreements are rarely about the original logic. They are about the fact that the original logic stopped being true.

Pay the role at the market midpoint

Salaries are the easy part and most firms overcomplicate them. Determine the role, find the going market rate for it, and pay at the midpoint. Benchmark salary data is widely available.

The midpoint is right because a boutique is neither a start-up nor a market leader. It sits in the middle. Adjust up or down as you see fit, but pay against the market rate, because labour is a commodity priced in the open market. If the person left, that is what they would fetch. If you were hiring their replacement, that is what you would pay. Anchoring on external benchmarks takes the subjectivity out of the conversation, which is most of what makes it unpleasant.

Then decide what the bonus is actually rewarding

Bonuses matter more in a boutique than in a large firm for a simple structural reason. Market leaders have hundreds of partners, so each share of the profit pool is small. Boutiques have few, so each share is large and each decision has a disproportionate effect on both the firm and the person.

There are three systems, and the choice between them is a statement about what the firm values.

Seniority. Increases are formulaic and tied to years of service. It is easy to administer, focused on the long term, and avoids arguments and year-to-year swings. Its costs are equally clear: past contributions may have stopped benefiting the firm, and young high performers will not wait around while the incumbents get rich.

Performance. Profits are divided by objective measures such as origination, yield and project profitability. Hit the goals and get paid. The problem is that it removes every incentive to build the firm. Referrals are not rewarded. Developing staff is not rewarded. Leverage suffers because partners hoard billings rather than hand work to juniors. Pay that prioritises short-term performance can destroy the ability of a firm to scale.

Reconciliation. Partners distribute profits after reconciling how wealth was created for one another, with a consistent methodology, collected evidence and written explanations. The partners are the jury. It balances short and long term, removes politics because everyone has a voice, and can evolve as the firm evolves. It pays contribution rather than title, and prevents coasting because it is recalculated annually. It is also time consuming, requires full disclosure so everyone knows what everyone else received, and gets unwieldy as the partnership grows.

Contribution to wealth creation is the criterion that holds

Owners exist to create wealth for themselves and for the other owners, and wealth is created in exactly two ways: increasing EBITDA, and increasing the multiple applied to EBITDA.

The arithmetic makes the conversation concrete. A firm with five million dollars of EBITDA that would fetch ten times earnings is worth fifty million. A partner whose work adds one million of EBITDA has created ten million of wealth. A partner who does something that moves the multiple from ten times to eleven has created five million. How much of that should they receive is a real negotiation, but it is a negotiation about a number that can be evidenced.

This is also the test to apply to any proposed equity grant. If the contribution can be attributed to wealth creation, it can be rewarded in cash against that attribution. Ownership is only the right instrument when the person needs to participate in the balance sheet rather than the income statement.

The mistake Greg made, and what it cost

Greg Alexander did not change his partner compensation system as SBI moved from start-up to boutique. By his own account it cost him millions of dollars and created lasting tension.

The system paid partners generous salaries that were not market based, at roughly twice what they could have fetched in the open market. Bonuses were paid as distributions, and the distributions were paid according to each partner's equity stake. If a partner owned twenty-five per cent, they received twenty-five per cent of the distributions.

That was the error, and the reason is the single most useful line on this subject: changing equity stakes is much harder than changing bonus payouts. The rigidity was the damage.

The cause was loyalty. The early partners had taken a real risk and had taken equity instead of cash, and Greg wanted to reward the sacrifice. What he did not do was decide when that debt was paid, so he kept paying it in perpetuity. Meanwhile the newer partners, whose contribution was large and growing, were undercompensated, and the gap compounded until it was extreme. The dollars should have shifted. They did not. Friendships were damaged and, in his own words, this was a failure of leadership by him.

The instructive part is that nothing in that story is about greed or bad partners. It is about a structure that could not move once contributions did.

If you do grant equity, do this first

Put a buy-sell agreement in place. It is a contract stipulating how a share of the business can be bought and sold, and it is what prevents expensive litigation later.

It should include a business valuation clause specifying that a valuation expert will determine the appropriate method. Do not predetermine a formula such as two times trailing twelve-month revenue, because that may bear no relation to what the firm is actually worth when the clause is triggered. Set ground rules for how a purchase can be funded and what can trigger a sale, and involve a tax adviser in the drafting.

Write it before it is needed. Doing so removes the emotion from decisions that are otherwise made at the worst possible moment. Very few boutiques have one, mostly because they do not think they need one. Every firm with more than one owner should have one.

What changes in an AI-native firm

The reconciliation system has always had one real weakness: attributing a person's activities to wealth creation is difficult, and people confuse correlation with causation. As more of the repeatable work is produced by software under supervision, that weakness gets sharper, because output stops being a clean proxy for individual contribution.

The practical consequence is that the evidence has to get better before the system does. Firms measuring profit at the firm level cannot attribute wealth creation to anyone. Firms measuring it per project can. That reporting change is the precondition for paying contribution rather than title, and it is worth making before the compensation design, not after.

This section is a reasoned extension of the framework rather than a position Greg has published.

When this answer flips

Two cases genuinely call for ownership. The first is a rising star whose retention actually requires it and whose contribution to wealth creation can be evidenced. The second is a partner-track structure where participating in the balance sheet is the explicit deal.

And one case makes the whole question moot. If you are underpaying relative to the market, equity will not fix it. Employees know what they are worth, almost everyone who leaves a firm that underpays lands a job with better pay, and offering a share of a firm while paying below market reads as an attempt to substitute upside for wages.

The short answer

Pay the role at the market midpoint, then reward contribution to wealth creation through the bonus, because a bonus can change every year and an equity stake cannot. Never award ownership for effort, because effort cannot be valued as a percentage and belongs in salary. Reserve equity for the few people who need to participate in the balance sheet, and put a buy-sell agreement with a valuation clause in place before you need it.