In the Weeds: Running the Numbers on a Lemonade Stand (With AI Doing the Math)

I know a woman who runs a dog-grooming business out of a converted garage. She's good at it — booked three weeks out, regulars who won't go anywhere else, a waiting list for new clients. She grosses about ninety thousand dollars a year and has no idea whether she's actually making money.

That's not a joke. She knows what comes in. She knows what goes out. But the space between those two numbers — the margin, the overhead allocation, the part where you figure out if your Tuesday afternoon slot is profitable or if you're losing eight dollars every time you wash a Bernese Mountain Dog — that's the math she's been avoiding since she opened.

She's not alone. According to the Small Business Administration, roughly half of small businesses fail within five years, and the most common reason isn't a bad product or a bad market. It's running out of cash. Not because the money wasn't there — because the owner didn't see it leaving until it was gone.

This is a piece about the specific, practical, unsexy math that AI can do for a small business owner in an evening. Not "AI can transform your business" — that's a magazine cover, not help. This is the spreadsheet-and-calculator version: what the numbers actually look like, how to get them out of your head and into a format you can act on, and the three questions that tell you whether your lemonade stand is a business or a hobby.

Question one: What does it actually cost you to do the thing?

Every small business sells a thing. A haircut, a cake, a logo, a cleaned house, a groomed dog, a repaired fence, a tax return. The owner knows what they charge for it. They almost never know what it costs them to deliver it.

The dog groomer charges sixty-five dollars for a standard groom. She thinks her costs are "shampoo and stuff." Here's what they actually are:

• Shampoo, conditioner, ear cleaner, nail grinding discs: roughly $4.20 per groom (she buys in bulk)
• Water and electricity for the tub, dryer, and clippers: roughly $2.80 per groom (estimated from her utility bills, divided by monthly grooms)
• Wear on equipment — clippers need new blades every 400 grooms, the dryer lasts about three years, grooming table was $800: amortized, roughly $1.50 per groom
• Towels and cleaning supplies: about $1.00 per groom
• Insurance: $2,400/year, divided by roughly 1,200 grooms/year = $2.00 per groom
• The space itself: she pays the mortgage regardless, but if she rented the garage as a studio, she'd get $600/month — that's an opportunity cost of about $6.00 per groom

Total direct cost per groom: roughly $17.50.

That leaves $47.50 in gross margin per groom. That sounds great — until you factor in the thing nobody wants to talk about.

Her time.

Each standard groom takes about ninety minutes, door to door. She does six a day, five days a week. At sixty-five dollars per groom, she's grossing $390 per day. Subtract the $105 in direct costs (6 x $17.50), and she's netting $285 per day for nine hours of standing, lifting, and getting bitten by a Shih Tzu.

That's $31.67 per hour. Which is decent — better than a lot of jobs. But she's also the receptionist, the bookkeeper, the marketing department, the janitor, and the person who answers texts at 9 PM about a last-minute cancellation. If she's honest about the hours she works (not just the grooming hours), it's closer to eleven hours a day. Now it's $25.90 per hour.

Still decent. But the Bernese Mountain Dog takes two and a half hours and she charges eighty-five dollars for it. Run the same math: $85 minus roughly $22 in supplies (bigger dog, more product, more time on the dryer) = $63 margin. Divided by 2.5 hours = $25.20 per hour. That Bernese is her lowest-margin appointment, and she has three of them a week.

She didn't know this until she sat down with Claude for forty-five minutes and listed every cost she could think of.

How to do this with AI tonight

Here's the prompt that starts the conversation. Paste it into Claude, ChatGPT, or any AI tool:

"I run a [type of business]. I charge [price] for [service/product]. Help me figure out what it actually costs me to deliver that service. Ask me questions about my expenses — supplies, equipment, space, insurance, utilities, software, anything I might be forgetting. Then calculate my true cost per unit and my gross margin. Be specific and ask me for real numbers, not guesses."

The AI will ask you about expenses you've forgotten. Everyone forgets something. The dog groomer forgot that she pays $49/month for scheduling software. She forgot that she drives to the pet supply store twice a month and the gas costs something. She forgot that she throws away about $200 worth of towels a year.

None of these are big numbers. Together, they moved her cost per groom from "shampoo and stuff" to $17.50. That's the difference between thinking you're making forty-seven dollars per groom and knowing you're making forty-seven dollars per groom. Same number — completely different confidence level.

Question two: Which customers are actually making you money?

This is the one that hurts.

Most small business owners treat all revenue as equal. A dollar from Client A is the same as a dollar from Client B. It isn't. Some clients take more time, require more follow-up, cancel more often, pay late more often, and generate more stress per dollar than others.

The AI prompt for this:

"Here are my top 10 clients by revenue. For each one, I'm going to tell you: how much they pay per month, how many hours I spend on them, how often they need revisions or callbacks, whether they pay on time, and any other time costs. Help me rank them by actual profit per hour — not just total revenue."

A freelance designer I know ran this exercise and discovered that her largest client by revenue — $3,200/month — was also her least profitable by the hour. They required an average of four revision rounds (her other clients averaged 1.5), they scheduled calls during her evenings, and they paid net-60 instead of net-30. Her actual hourly rate on their work was $22. Her smallest client, who paid $800/month for a retainer, required almost no revisions and paid within a week. Her hourly rate on that work was $67.

She didn't fire the big client. But she restructured the contract: revisions beyond two rounds are billed hourly, calls are scheduled during business hours, and payment terms are net-30. The client accepted. They were a profitable client three months later.

The AI didn't make that decision. She did. The AI just made the numbers visible.

Question three: What's your break-even — and do you know when you hit it each month?

Break-even is the point where your revenue covers all your fixed costs and you start making actual profit. Below that line, every dollar goes to keeping the lights on. Above it, every dollar is yours.

Most small business owners have never calculated their break-even point. They know it exists in a vague way — "I need to do at least twenty jobs a month to cover everything" — but they haven't pinned it to a number.

The prompt:

"Help me calculate my monthly break-even point. My fixed monthly costs are [list them: rent, insurance, software subscriptions, loan payments, phone, internet, anything you pay regardless of how much work you do]. My average revenue per [unit/job/client] is [amount]. My average variable cost per [unit/job/client] is [amount]. How many [units/jobs/clients] do I need per month to break even?"

The dog groomer's fixed costs: mortgage allocation ($600), insurance ($200), software ($49), phone ($85), internet ($60), equipment amortization ($125). Total: $1,119/month.

Her average margin per groom (after variable costs): $47.50.

Break-even: $1,119 / $47.50 = 23.6 grooms per month. She does about 120. She's well past break-even by day five of each month.

That's reassuring. But here's what the number actually tells her: she has pricing power. She could raise her prices by ten dollars per groom, lose fifteen percent of her clients, and still make more money. She could also drop the Bernese Mountain Dogs entirely, replace those slots with standard grooms, and make an extra $48 per week.

These aren't hypotheticals — they're math. The AI can model every scenario in seconds. "What happens if I raise prices by $10 and lose 15% of clients?" "What happens if I add a Saturday but only work half days?" "What happens if I hire a part-time assistant at $18/hour and do eight grooms a day instead of six?"

Each question takes sixty seconds to answer and would take the business owner hours to model in a spreadsheet — if they even knew how to set up the spreadsheet.

The thing the spreadsheet people don't understand

There is a version of this article that says "just use Excel." And yes, a spreadsheet can do all of this math. The problem is that most small business owners don't think in spreadsheets. They think in stories: "Tuesday was a good day, I had a full book." "That big client is my bread and butter." "I'm making good money — I think."

Stories are fine for motivation. They're terrible for decisions. The gap between "I think I'm doing okay" and "I know exactly which clients, services, and time slots make me money" is the gap between a business that survives and one that grows.

AI bridges that gap because it speaks in conversation, not cells. You don't have to know what a VLOOKUP is. You don't have to format a column. You tell the AI what you know, it asks for what's missing, and it gives you the answer in English.

That's not a technology story. It's a literacy story. Financial literacy — the practical, Tuesday-afternoon kind — has been locked behind spreadsheets and accountants for decades. AI puts it in a conversation. For a small business owner who has never had a CFO and never will, that conversation might be the most valuable one they have this year.

What to do tonight

Pick one of the three questions:
1. What does it actually cost me to do the thing?
2. Which customers are actually making me money?
3. What's my break-even point?

Open Claude or ChatGPT. Paste in the prompt. Answer the AI's questions honestly — including the expenses you've been ignoring. Take the number it gives you and sit with it for a day.

If the number is good, you'll sleep better. If the number is bad, you'll know exactly what to change. Either way, you'll know. And knowing beats guessing every time, especially when it's your money and your time and the business you built with your hands.

The lemonade stand deserves real math. Go run the numbers.