Founders ask me some version of the same question: “How much can I take out?” It sounds like one question. It is actually two, and conflating them is how businesses either starve their own growth or create a tax problem they didn’t see coming.
The first question is about capacity: how much cash can this business sustainably release without breaking something that matters? The second is about characterization: once you decide to take cash out, how should that payment be labeled and taxed? The first is an operating decision. The second is a compliance decision. You need an answer to both, but they are not the same answer, and the order matters.
Most owner distribution problems I see are not tax problems. They are cash-flow problems dressed up as tax problems. A founder takes a draw that looks fine on the annual profit-and-loss statement, then discovers in week nine of a slow collections quarter that payroll is tight. The draw wasn’t wrong because of how it was taxed. It was wrong because nobody tested it against the cash the business actually had.
Start With Capacity, Not With a Percentage
There is no defensible rule of thumb for owner distributions as a percentage of revenue or profit. Anyone who gives you one is guessing. The right number is specific to your business, and it comes out of the instrument you should already be running: the 13-week cash flow forecast.
If you are not running a weekly cash forecast, that is the first thing to fix, and it is more important than the distribution question. A 13-week forecast projects cash in and cash out week by week, including payroll, payroll taxes, debt service, rent, supplier payments, and the timing of customer collections. It is not a budget. It is a liquidity map.
Here is the test I use. Set your proposed owner draw, then run it through all 13 weeks. The draw is sustainable only if the lowest projected weekly cash balance across the entire horizon stays above a floor you have defined in advance. That floor should cover three things: the largest single customer or receivable concentration you carry, one full payroll cycle plus its tax deposits, and whatever debt service is contractually due in the window.
If the lowest week breaches the floor, the draw is too high, regardless of what the annual statement says. A draw that survives the annual P&L but fails in week nine is not a distribution policy. It is a liquidity bet, and you are the one holding the other side of it.
Why the Annual P&L Misleads You
Annual financials smooth out timing. Cash does not. A service business with strong gross margins can look profitable all year and still run out of cash in a seasonal trough because receivables aged out and the owner took distributions on the strength of booked revenue rather than collected revenue.
This is where unit economics do their work. Before you set a draw, you need to know three numbers: gross margin per job or per account, contribution after direct labor, and the revenue level at which fixed overhead is covered. Those numbers tell you how much of each new dollar of revenue actually becomes cash you can use. A draw that looks affordable on total revenue can be unaffordable if the revenue is coming from low-margin work that consumes working capital faster than it generates it.
For a light manufacturer, the same logic runs through inventory. A growing order book can absorb cash for months before it returns any. If you are taking distributions while inventory and receivables are expanding, you are financing growth out of the owner’s pocket whether you intended to or not. The forecast will show you this. The annual statement will not.
Separate the Draw Decision From the Tax Label
Once you know how much cash the business can release, you still have to decide how that release is characterized. This is where the rules get specific, and where founders get into trouble by treating the label as a formality.
If you operate as an S corporation, the IRS is direct about the requirement: the corporation must pay reasonable compensation to a shareholder-employee for services before non-wage distributions may be made. The agency also states that it has the authority to reclassify payments made to shareholders from non-wage distributions, which are not subject to employment taxes, to wages, which are. That is not a theoretical risk. Courts have consistently held that S corporation officers and shareholders who provide more than minor services and receive or are entitled to compensation are subject to federal employment taxes.
What counts as reasonable compensation is not spelled out in the Code or the regulations. The IRS notes that there are no specific guidelines, and that courts decide on the facts and circumstances of each case. The factors courts have weighed include training and experience, duties and responsibilities, time and effort devoted to the business, dividend history, payments to non-shareholder employees, what comparable businesses pay for similar services, compensation agreements, and whether a formula was used to set compensation.
One case worth knowing: in David E. Watson, PC v. U.S., the shareholder received wages of $24,000 a year alongside large distributions. The taxpayer argued that the corporation only intended to pay wages of $24,000 and that its intent controlled. The Eighth Circuit disagreed, holding that the test is whether the payments received by the shareholder were truly remuneration for services performed, and that the intent to limit wages is not a controlling factor.
The practical takeaway is not to guess at a salary number. It is to recognize that the tax characterization of your draw is a separate decision from the amount of cash you can afford to release, and it needs its own documentation. Your accountant or tax counsel should be setting the reasonable compensation figure with reference to the factors above. Your 13-week forecast should be setting the total amount of cash that leaves the business. If those two numbers are set by the same person in the same conversation with no documentation, you have a policy in name only.
The Capital-Allocation Ladder
Here is where the growth tension actually lives. Every dollar above the sustainable draw is a choice, and most founders make that choice by default rather than by design. I use a five-rung ladder, and the ordering is the point.
First, the minimum operating reserve. This is the cash floor from your forecast, funded and left alone. It is not investable capital. It is the thing that lets you sleep and lets you say no to bad terms.
Second, debt service and covenant headroom. If you have a loan with financial covenants, the cash that keeps you comfortably inside them ranks above owner liquidity. Breaching a covenant is expensive in ways that are hard to reverse.
Third, funded growth initiatives with a defined payback. A hire, a piece of equipment, an inventory build, or a receivables float all compete for the same dollars as a distribution. Each one should show its cash payback window inside the forecast horizon before it outranks a distribution. If a growth investment cannot demonstrate when it returns the cash, it is a hope, not a plan.
Fourth, owner distribution above the baseline draw. This is the discretionary layer. It is real money and you are entitled to it, but it sits below the reserve, the debt, and the funded growth that has earned its place.
Fifth, tax-advantaged retirement contributions where eligible. For an owner with no employees, a one-participant 401(k) can absorb cash that would otherwise be distributed and taxed. The IRS describes this plan as covering a business owner with no employees, or that person and their spouse, and notes it has the same rules and requirements as any other 401(k). The owner can contribute as both employee and employer: elective deferrals up to 100% of compensation up to the annual deferral limit, plus employer nonelective contributions up to 25% of compensation as defined by the plan. The agency also notes that an owner with no common-law employees does not need to perform nondiscrimination testing, but that the advantage vanishes if the employer hires employees. That is a policy trigger, not a footnote: your distribution policy should specify when the retirement plan structure gets revisited as headcount changes.
I am not giving you percentages for each rung. The right split depends on your forecast, your debt, and your growth pipeline. What I am giving you is the ordering logic, because the ordering is what prevents a distribution from quietly outranking the reserve or the growth that would have paid for itself.
What a Written Policy Actually Contains
A distribution policy that lives in your head is not a policy. It is a habit. Here is what the written version needs.
A stated draw amount and a review cadence. Put a number on it and say when it gets reviewed, whether quarterly or at each forecast refresh. A draw that is never formally revisited drifts upward by default.
A defined reserve floor. State the floor in dollars or weeks of operating cost, and state what it is meant to cover. The floor is the constraint the draw must respect.
A distribution trigger tied to forecast thresholds. Discretionary distributions above the baseline should require the forecast to show the floor holding across all 13 weeks, not just the current week.
A decision log. Record each distribution decision, the forecast it was based on, and the reasoning. This is dull and it is the single most useful document you will have when you eventually sell or raise.
A breach rule. When the forecast breaches the floor, the response is to reduce the draw, not to drain the reserve. Write that down before you need it, because the moment you need it is the moment you will be least inclined to follow it.
The Sell-Side Angle for Sub-$30M Operators
If you are a bootstrapped or lightly funded operator in the $1M to $30M revenue range, a documented, stable distribution policy is worth more to you at exit than a higher trailing owner earnings figure. That is a qualitative judgment, not a measured valuation effect, and I want to be clear about the difference.
What I can tell you is where diligence friction comes from. Buyers and their advisors reconstruct normalized earnings, and owner compensation is one of the first lines they interrogate. Erratic draws, undocumented salary decisions, and distributions that move with the owner’s personal needs rather than the business’s cash cycle all create questions. A policy that shows a stated draw, a reserve floor, a review cadence, and a decision log answers those questions before they are asked. It makes the earnings story defensible because the pattern is visible and the reasoning is recorded.
This connects to the non-dilutive funding conversation as well. If you are considering an SBA 7(a) loan, the program provides a loan guarantee to lenders and can be used for short- and long-term working capital, with a maximum loan amount of $5 million. Under SBA SOP 50 10 8, effective June 1, 2025, 7(a) loans for startup ventures and changes in business ownership require at least 10 percent equity injection from the buyer, and a seller note must be on full standby with no principal or interest payments for the entire SBA loan term to count toward that injection, and may not account for more than 50 percent of the total. Those are structural constraints on how a deal gets financed, not rules about your distribution policy. But they reinforce the same point: the cleaner your cash story, the more options you have.
The Hard Truth
The amount you can take out is not a number you negotiate with yourself. It is a number your forecast produces. If you want to take out more, the path is not to argue with the forecast. It is to improve the inputs: collect receivables faster, raise gross margin per job, reduce the working capital that growth consumes, or build the reserve that lets you take a larger draw without breaching the floor.
That is a slower answer than a percentage of revenue. It is also the only one that survives a slow quarter, a tax examination, and a buyer’s diligence. Set the draw from the cash, label it correctly, rank every dollar above it against the ladder, and write the policy down. Then stop negotiating with yourself and run the business.
Frequently Asked Questions
Can I just take distributions instead of a salary from my S corporation?
No. The IRS states that S corporations must pay reasonable compensation to a shareholder-employee for services before non-wage distributions may be made, and that it has authority to reclassify non-wage distributions as wages subject to employment taxes. Courts have upheld that authority repeatedly. The amount of reasonable compensation is decided on facts and circumstances, so it needs documentation and professional input, not a guess.
Is there a standard reserve level I should target?
Not one that applies across businesses. The floor should be built from your own forecast: your largest customer or receivable concentration, one payroll cycle plus tax deposits, and contractual debt service in the window. A reserve sized to someone else’s business is not a reserve for yours.
Does a one-participant 401(k) still make sense if I hire?
It can, but the rules change. The IRS notes that an owner with no common-law employees does not need to perform nondiscrimination testing, and that the advantage vanishes once employees who meet plan eligibility requirements are hired. Your policy should specify when the plan structure gets reviewed as headcount grows.
Will a documented distribution policy increase what my business sells for?
I can’t give you a measured valuation effect, and I would be skeptical of anyone who does without evidence. What a documented policy does is reduce diligence friction by making owner compensation and cash patterns visible and explainable. That is a qualitative benefit, and in my experience it is a real one, but it is not a multiple.