Due diligence for operators

Diligence has to answer one question a checklist never asks: will the business still run the morning after the founder hands you the keys. The standard checklists for a small business are written by accountants and lawyers, for accountants and lawyers, and they verify that the thing is legal and that the numbers foot. The rest is the part only an operator sees, and it decides whether you bought a company or a job.

Two layers, and the order is not optional

I diligence in two layers, and I run them in a fixed order. First the numbers, because if the earnings are not real nothing else matters and every hour spent on culture and systems is an hour spent decorating a lie. Then the operations, because a business with clean books can still fall apart the day its one indispensable person walks out the door. Skip the first layer and you overpay for a fiction. Skip the second and you buy a job you did not know you were applying for.

The reason the order matters is leverage. The numbers are cheaper to check and faster to kill a deal on. If the profit does not survive the bank statements, I never have to meet the operations manager, and only a business that clears the financial layer earns the slower work of the second.

Verify the number before you meet the people. A bad number kills a deal cheaply, and a good story never does.

Four earnings numbers, and why the seller picked that one

Before any of it, the vocabulary. A small business is sold on a multiple of earnings, and there are four different numbers that all get called earnings. Which one the seller quotes tells you something before you have opened a single file.

Revenue

What came in. Quoted when the profit is not flattering, and on its own it says nothing about whether the business is worth owning.

Net profit

What the filed accounts declare after everything, including the owner's pay and whatever the business has been carrying for them personally. Usually the lowest of the four and the most defensible.

SDE, seller's discretionary earnings

Net profit with ONE owner's salary and personal expenses added back. The standard measure for owner-operated businesses, on the logic that the buyer will take that seat and that pay. Only honest if the buyer really is going to work in it.

EBITDA

Earnings before interest, tax, depreciation and amortisation, with a market-rate manager costed in rather than added back. The right measure once a business is big enough to be run by someone other than its owner, and the reason the same company can be quoted at two very different numbers.

The numbers: is the profit real

Three numbers carry almost the whole financial question. Verify these and the books have told you what they can. Skip any one and the clean profit on the last page can still be hiding the thing that sinks the deal.

Quality of earnings

Start with quality of earnings, the single most important question in any small business diligence: are the profits real and recurring, or are they a good year, a one-off contract, and a pile of add-backs stacked to make the multiple look small. I trace the reported profit back through the accounts and out to the bank statements, because a profit that cannot be found in the bank is not a profit, it is an opinion. Every owner add-back gets a name and a reason. A personal car, a family member on payroll who does not work, rent paid to the owner's own building: none of these are fatal, but each one has to reconcile or it comes out of the number.

Working capital

Then working capital, the question buyers skip and regret. It is the cash the business needs to keep the lights on between doing the work and getting paid for it. Inventory on the shelf, invoices out but not collected, suppliers waiting to be paid: that gap is real money, and you fund it on day one. A business can be profitable on paper and still swallow a large slice of your cash the week you take over, because the receivables that funded last month's payroll belong to the seller and next month's belong to you. I want the gap quantified before I sign, not discovered the first Friday I have to make payroll.

Customer concentration

Customer concentration is the third number, and the one that turns a good business into a fragile one. If one client is a fifth, a third, half of revenue, you are not buying a business, you are buying a single relationship you did not build and cannot control. Pull the revenue by customer for three years, ask how each big account was won, and ask what happens to the earnings the day that account leaves. One client is not a customer base. It is a risk wearing a customer's clothing.

Quality of earnings

Whether the reported profit is real, recurring, and traceable to the bank, or a good year propped up by one-off contracts and add-backs. The first number to verify, because every other number sits on top of it.

Working capital

The cash a business needs to bridge the gap between doing the work and getting paid for it: inventory, receivables, and payables. You fund it on day one, so an unquantified gap is a surprise bill in your first week.

Customer concentration

How much of revenue leans on a single account. The higher it is, the less you are buying a business and the more you are buying one relationship that can leave the day the deal closes.

Clean books tell you the business made money. They do not tell you it will make money without the man who is leaving.

The operations: does it run without the founder

Now the layer the financial checklist ignores. A business can post clean, recurring, diligence-proof earnings and still be one resignation away from collapse, and no accountant is trained to catch it. This is the layer an operator reads for, because he has run the thing being sold and knows where the load actually sits.

Owner dependence

Owner dependence is the first thing I map. Who quotes the jobs, who holds the customer relationships, who knows the pricing, who fixes it when it breaks. If the honest answer to all of those is the founder, the business does not run, the founder runs and the business follows. That pulls the price down, but only if the dependence is fixable, not if the owner is the entire product. It is the same trap I write about in owner dependence, read from the buyer's side of the table: the founder's ceiling is my opportunity, but only if I can price it honestly.

The key people who are not the owner

Every small business has one or two employees who quietly hold it together: the operations manager who knows every account, the lead tech every customer asks for by name. Find them, and learn whether they are staying, what they are paid, and whether they have any reason to leave the day ownership changes. Losing the founder is a known cost, priced into the deal. Losing the person you did not know mattered is the one that blindsides you the month after close.

Systems, and the boring compliance nobody checks

Systems are the next line. Are the processes written down, or do they live in one person's head. Is there a real CRM and a real set of books, or a shoebox and a memory. A business that runs on documented process is one you can own from a distance and improve. A business that runs on the founder remembering everything is one you have to personally replace him inside of, which is not owning a company, it is inheriting his workload. Last, the boring compliance nobody checks: licenses current and actually in the buyer's name after close, insurance that covers what the work exposes, contracts that survive a change of ownership instead of letting the customer walk, employment paperwork, safety records, the permits the work legally requires. None of it is exciting, and all of it can void a deal or hand you a liability the seller quietly priced you to absorb.

The order I run it in

Diligence is a sequence, not a pile. Run it in order and each stage either kills the deal cheaply or earns the right to the next one. Each stage also has a failure mode, and every failure mode is the same shape: doing the motion without doing the work.

1. Verify the earnings

   Trace the reported profit through the accounts and out to the bank, and give every add-back a name. The failure mode is accepting the seller's adjusted-earnings schedule at face value, which prices the business on the number he wants rather than the number the bank can prove.

2. Fund the gap

   Quantify the working capital the business needs to run, so you know the cash you inject on day one. The failure mode is modelling the profit and forgetting the receivables that funded last month's payroll leave with the seller, so your first Friday arrives as a surprise.

3. Stress the revenue

   Pull revenue by customer for three years and ask what the earnings look like the day the biggest account walks. The failure mode is reading the healthy top line and never the concentration underneath it, so you buy a single relationship and call it a book of business.

4. Read the dependence

   Map who decides, who holds the relationships, and who fixes what breaks, then ask what stops if the founder takes a month off. The failure mode is mistaking founder heroics for a working system, and paying system money for one person's overtime.

5. Check what voids the deal

   Confirm the licenses, insurance, and contracts survive a change of ownership, and that the safety and employment paperwork is clean. The failure mode is assuming boring means safe, then inheriting a liability the seller quietly priced you to absorb.

The list an operator actually runs

The sequence above is the shape. This is the line-by-line I want to be able to tick before I sign, and it does not care how polished the pitch was. Anything I cannot tick honestly is either a discount I take or a reason I walk.

• The reported profit traces cleanly to the bank statements, not just to the P&L
• Every owner add-back has a name, a reason, and actually reconciles
• Earnings survive a soft year, not only the best one on record
• Working capital is quantified, so I know the cash I inject on day one
• Receivables are collectible, not aged invoices dressed up as assets
• No single customer is more than a fifth of revenue
• I know how each large account was won and who owns the relationship
• The founder can take a month off without the business stalling
• The key non-owner people are named, paid fairly, and have no reason to leave at close
• The recurring decisions are written down, not living inside one head
• Licenses, insurance, and contracts survive a change of ownership
• Safety, employment, and permit paperwork is current and clean

What only an operator sees

Here is what the numbers hide. A business can post clean, recurring, diligence-proof earnings and still be held together entirely by founder heroics: the owner personally saving every near-miss, working the sixty-hour week, being the reason nothing has broken yet. That effort does not show up as a line item. It shows up as the absence of a system, and it walks out the door with him.

So I ask the operator's questions. What breaks first when the founder takes a month off. How many decisions a day route through one person. What is not written down that would have to be relearned from scratch. The answers tell you the real price, because the gap between how it runs today and how it should run is the work you are signing up for.

That gap can be the opportunity, a fixable dependence bought at a fair discount, or a trap, a business that only ever worked because one exhausted person refused to let it fail. Same clean books on both sides of that table. Only an operator who has run the thing can tell you which one is on offer, and that reading is the entire reason I do the second layer at all.

A worked example, plainly hypothetical

Picture a distribution business in the 1M to 10M band, ten years trading, profit clean and growing. On paper it clears every financial test. Then you run the operations layer. The founder quotes every large job himself, holds the two accounts that make up most of the margin, and is the only person who knows which suppliers bend on a rush order. None of that is in the accounts, and all of it leaves the day he does. The number did not lie. It just was not the whole sentence.

What to walk away from

Three findings end the conversation for me, and all three survive a lawyer's checklist untouched. First, books nobody can explain: not always fraud, sometimes just chaos, but what I cannot verify I will not buy. Second, one-customer concentration: when a single account is the business, the earnings leave the day that account does, and no price is low enough to make that safe.

Third, and the one I have to talk myself out of most often, the turnaround dressed as a bargain. When a business is cheap because it is broken, the low price is not a discount, it is a warning. I buy things that already work and price the gap I can close myself, not a rescue I have to call a deal.

That is also why I would rather help someone pressure-test a deal honestly through advisory than watch them fall in love with a number, and why the traits I read for here are the same ones that make a business worth buying in the first place.

Diligence is not the hunt for a reason to say yes. It is the discipline to say no fast enough that the one deal which clears every line is the one you can actually keep.