Here is the plain version. A business has to pay for things before it gets paid for them. It buys materials, pays staff, and covers rent, and only later do customers pay their invoices. The cash that bridges that gap is working capital. If the seller collects every receivable, pays down every payable, and walks away with the cash on closing day, the new owner inherits a business that has the earnings on paper and an empty tank in real life.
That is why I treat working capital as its own review and not as a footnote to the earnings review. The add-back work tells me whether the profit is real. This work tells me whether I can afford to run the business that earns it.
What Working Capital Actually Means in a Deal
In a small business purchase, working capital usually means the current assets the business needs to operate, minus the current liabilities it owes in the near term. In practice, the parts that matter are these:
- Accounts receivable. Money customers owe the business. Only the collectible portion counts.
- Inventory. Goods or materials on hand. Only what can actually be used or sold counts.
- Prepaid expenses. Things already paid for that will benefit the new owner, such as insurance or software.
- Accounts payable. What the business owes vendors. A bill that is already late tells you something.
- Accrued expenses and payroll. Wages, taxes, and other costs earned but not yet paid.
- Customer deposits and deferred revenue. Cash the business has already collected for work it still has to do. This is a liability even though the money sits in the bank.
Many small deals are priced on a cash-free, debt-free basis with a normal level of working capital included. That sentence hides a lot. Who decides what normal is? How is it measured? What happens if the closing number is lower? The answers should be in the agreement, and they should be reasoned out before anyone signs.
I am not a lawyer or an accountant, and the exact mechanics belong to people who are. What I can do is arrive at that conversation knowing what the business really needs and where the numbers look soft.
How I Use AI to Rebuild the Numbers From Raw Records
The seller's balance sheet is a summary. I want the records underneath it. For a working capital review I ask for the following, usually for the last twenty-four months where it exists:
- Month-end balance sheets and the accounting system's receivables and payables aging reports.
- The invoice and bill detail behind those reports.
- Bank statements for every operating account.
- An inventory count or valuation report, with the method used.
- A list of customer deposits, prepayments, gift cards, or retainers outstanding.
- Payroll summaries and a schedule of accrued but unpaid costs.
I load the material into an AI workspace and give it the same sequence of jobs each time. The order matters, because each step makes the next one cleaner.
First, rebuild the balance by month. I ask it to produce one table of receivables, payables, inventory, and deposits at every month-end, from the underlying detail and not from the summary line. If the rebuilt number does not match the seller's balance sheet, that gap is the first thing I ask about.
Second, age everything. I ask for receivables bucketed by how long they have been open, and payables bucketed by how late they are. The aging report alone is not enough, because I also want to see how the buckets moved over time. A receivable that has crept from current to ninety days over four months is a different story from one that was always ninety days.
Third, look for the seasonal shape. Most businesses have a rhythm. Cash tightens before a busy season and loosens after. I ask AI to chart the working capital level month by month so I can see the low point, not just the average. The low point is the number I need to be able to fund.
Fourth, hunt for window dressing. This is where AI earns its keep. I ask it to compare the last two or three month-ends before the sale process began with the same months in prior years, and flag anything that moves in a way the business history does not explain. Payables that suddenly stretch, receivables that suddenly shrink because of a push to collect, or inventory that jumps without a matching purchase pattern are all things I want to ask about.
The Four Patterns AI Surfaces Most Often
After enough reviews, the same handful of patterns keep showing up. Knowing them means I know what to ask.
Old receivables carried at full value
Receivables that are well past due are usually worth less than the balance sheet says, and some are worth nothing. AI is quick to isolate customers with a long open balance and no recent payment, and to show which invoices were partially paid and then went quiet. Each one becomes a question: is this customer still active, is there a dispute, and has the seller ever written one like it off? If the seller counts a stack of stale invoices as part of the working capital being delivered, the real figure is lower than the stated one.
Payables stretched before the sale
When a business knows it is about to be sold, there is a temptation to hold cash and let vendor bills run late. It makes the closing bank balance look better and leaves the new owner with a pile of overdue invoices and irritated suppliers. AI flags this by comparing how long bills used to take to pay with how long they take now. A shift of a few weeks in the last quarter is worth a direct conversation.
Deposits treated as income
If customers pay in advance, the business owes them either work or a refund. That cash is not free to spend. AI can match deposits received to work completed and show how much is still owed in service. If the seller keeps the deposit cash at closing, the buyer inherits the obligation with no money to fulfill it. This one matters most in businesses that book events, installations, or annual plans.
Inventory that is really dead stock
An inventory number is only as good as the last count and the method behind it. AI can compare inventory purchases to sales by item and show what has not moved in a year. I do not trust that to value the inventory alone, since some slow items are slow for good reasons, but it gives me a specific list to walk through with the seller instead of a vague worry.
What Stays My Decision
I want to be clear about the division of labor, because this is where people over-trust the tool.
AI can tell me that a receivable is two hundred days old. It cannot tell me whether that customer is a slow payer who always pays eventually or a closed business. Someone has to call, or ask the seller to, or look at the pattern of that relationship. AI can tell me that payables stretched in the last quarter. It cannot tell me whether that was a deliberate move or a rough patch for the owner that ended on its own. AI can tell me that inventory is not turning. It cannot tell me whether the items are seasonal, collectible, or obsolete.
Those are judgments about people and about the business, and they are the whole point of doing the work. The tool gets me to the right questions faster. The answers come from conversations, documents, and my own read of the situation.
I also do not use AI to set the working capital target on its own. The target is a negotiation between what the business has historically needed, what the lender expects, and what the seller is willing to leave behind. I use the review to know my range going in.
How I Think About a Working Capital Target
There is no single right method, and the people who draft the agreement will have their own. But here is how I think about it when I am preparing for the conversation.
I start with the monthly average of the cleaned working capital over the last year or two, after I have taken out the receivables I do not trust and put back the payables that were stretched. That gives me a more honest picture of what the business actually runs on. Then I look at the seasonal low point, because the average can hide a month where the business needs far more than usual. If I cannot fund the low point from my own reserves or the lender's terms, I am buying a risk I have not priced.
Then I compare my number to what the seller assumes. If the seller expects to take the cash and the receivables and leave the payables, that is a very different deal from one where everything stays. Neither is wrong. They are different prices dressed up as the same price.
A simple table helps me keep the stories straight:
| Item | Seller's stated figure | My cleaned figure | Question to ask |
| Receivables | Balance sheet total | Collectible portion only | Which invoices are older than ninety days, and why? |
| Payables | Balance sheet total | Total plus any stretched bills | Has the payment timing changed recently? |
| Inventory | Book value | Value of items that actually move | What has not sold in twelve months? |
| Deposits | Often not listed | Full amount owed in future service | Who keeps the deposit cash at closing? |
| Accrued payroll | Often understated | Wages and taxes earned but unpaid | What is owed on closing day? |
Timing: Why I Do This Before the Letter of Intent
The best moment to run this review is before I sign anything. At that point, the corrected number can shape the price, the structure, and the conversation about what comes with the business. Once an offer is signed, bringing up a working capital gap feels like a retrade. The seller has already spent the money in their head, and every change becomes a fight.
If the review is early, I can say something like: here is what I think the business needs to run, here is what the records show, and here is how I would like the agreement to handle the difference. That is a reasonable thing to say before an offer. It is a harder thing to say after.
It also affects the financing. A lender looking at a small business loan will ask how the buyer plans to fund day-to-day operations. If I walk in with a clear view of the seasonal low point and a plan to cover it, the conversation goes better than if I treat working capital as an afterthought.
A Practical Working Capital Review Checklist
This is the order I follow. You can copy it and adjust it to your own deals.
- Collect twenty-four months of month-end balance sheets, receivables and payables aging, bank statements, inventory reports, and a list of deposits and prepayments.
- Have AI rebuild receivables, payables, inventory, and deposits at each month-end from the underlying detail.
- Compare the rebuilt figures with the seller's balance sheet and write down every gap.
- Age receivables and payables, and track how the buckets moved month by month.
- Chart working capital across the year and mark the seasonal low point.
- Compare the months just before the sale process with the same months in prior years, and flag any unexplained movement.
- List old receivables, stretched payables, slow inventory, and unearned deposits, with a question for each.
- Ask the seller or the accountant for the answers, and record what changes.
- Build a cleaned working capital figure and a seasonal low-point figure.
- Decide how the purchase agreement should treat working capital, and take that view to the people who draft it.
- Note anything that needs a professional, such as an accountant's review of the closing balance sheet.
The whole review, from raw records to a cleaned number, takes me a day or two now instead of most of a week. That speed is what lets me do it early, while it can still change the deal.
The Mistakes I See Buyers Make
Almost every problem I hear about comes from one of a few habits.
The first is trusting the summary. A balance sheet total looks authoritative, but it can hide old invoices, late bills, and obligations that are not recorded at all. The detail underneath is where the truth is.
The second is assuming that a cash-free, debt-free price means everything is handled. It only means the structure is a certain shape. What counts as cash, what counts as debt, and what counts as normal working capital still have to be spelled out.
The third is ignoring the low point. A business can be comfortable ten months of the year and badly stretched in the other two. If you only plan for the average, the bad months arrive as a surprise.
The fourth is not planning for the first ninety days. Even a well-run business needs cash to absorb the transition. Vendors may tighten terms with a new owner, customers may pay slower during a handover, and you may want to invest in something right away. A cushion you do not have to touch is cheaper than one you scramble to build.
The fifth is skipping professional help on a larger deal. AI makes me better prepared, but an accountant can verify the records and a lawyer can write the terms. I use the tool to make that help cheaper and sharper, not to replace it.
Where This Fits in the Rest of the Process
I run this review alongside the others, not instead of them. The add-back review tells me whether the earnings are real. The customer concentration check tells me how fragile the revenue is. The working capital review tells me how much cash it takes to keep the machine running. Together they give me a picture I can price and finance with confidence.
If you want to see how the pieces connect, start with the add-back review, then the full due diligence checklist. For the revenue side of the same risk, the customer concentration check is the natural companion. And when it is time to put terms on paper, the guide to structuring an LOI shows where working capital belongs.
FAQ
What is working capital in a small business acquisition?
Working capital is the cash a business needs to operate day to day, usually measured as current assets such as receivables, inventory, and prepaid items minus current liabilities such as vendor bills, accrued wages, and customer deposits. In a deal, the question is how much of it stays with the business when ownership changes, and the answer should be written into the purchase agreement.
What does cash-free, debt-free with normal working capital mean?
It describes a price where the seller keeps the cash and pays off the debt, while the business is delivered with a normal level of operating assets and liabilities. The word normal is the part that needs work, because the buyer and seller may each have a different number in mind. It is worth defining how it is measured and what happens if the closing figure falls short.
How does AI help review working capital?
AI rebuilds receivables, payables, inventory, and deposits from the underlying records, ages them, charts them across the year, and flags movement that the business history does not explain. It does not decide whether an old invoice will be paid or what level of working capital the business needs. Those judgments stay with the buyer and their advisers.
What are the biggest working capital red flags?
The ones I look for first are receivables that are far past due and still carried at full value, vendor bills that stretched in the months before the sale, customer deposits the seller expects to keep, and inventory that has not moved in a year. Each is a reason to ask a direct question before making an offer.
Should I still use an accountant for working capital?
On larger deals, yes. AI helps you arrive prepared and ask better questions, but an accountant can verify the records and review the closing balance sheet, and an attorney can draft the working capital terms. I use the tool to make that work faster and more focused, not to skip it.
Working capital is not the exciting part of buying a business. Nobody brags about a clean receivables aging. But it is the part that decides whether the first few months feel like a business you own or a problem you inherited. A careful review turns a vague worry into a number you can plan around, and it is far cheaper to find the gap before closing than after.
If you are working through an acquisition and want help building the AI workflows that make this kind of diligence fast and repeatable, that is the kind of work I do with a small number of clients. You can request a strategic AI consulting conversation and we can talk through what makes sense for your situation.
Current Search Intent Check
Recent Search Console data shows people arriving through "how can small business owners use ai to get practical guidance when planning their next steps?". That changes the bar for this post: it needs to answer the operator question directly, name the workflow being improved, and give the reader a practical decision rule instead of another broad AI opinion.
Recent Search Console data shows people arriving through "small business acquisition due diligence checklist". That changes the bar for this post: it needs to answer the operator question directly, name the workflow being improved, and give the reader a practical decision rule instead of another broad AI opinion.