Loading
Business· Thomas Cuccittini 9 min read
For a small business, almost everything worth knowing lives in five figures. Learn to find these and you can analyse most small operations.
Analyze your Business Now
Get Startedvalidation isn't asking people if they like your idea. It's finding out whether they'll pay for it — before you build it.
Jul 22, 2026
Jul 22, 2026
You know the pattern. Before you've had a single customer, you're comparing project management platforms. You're picking brand colours. You're watching a comparison
Jul 20, 2026
The five-second version: analysing a small business is a skill, and it's mostly about knowing which four or five numbers matter and being willing to look at them honestly. The frameworks built for corporations actively mislead when applied to a ten-person operation. Here's the small business version.
Most of what's written about analysing a business was written about analysing a large one.
The frameworks assume audited financials, a finance department, clean historical data, and departments that exist. They assume the founder's personal spending isn't tangled up in the business account. They assume revenue arrives predictably enough that a quarterly trend means something.
Small businesses are not small large businesses. They're a different species — messier data, more concentration risk, an owner whose labour is unpriced, and economics that can shift completely on one client leaving. Analysing them well requires a different instinct, and it's a learnable one.
The first skill, and the one people learn last: don't open the data until you know what you're asking.
"How's this business doing?" produces a report that answers nothing. "Why did gross margin drop from 61% to 54% over three quarters?" produces an investigation. The difference is that the second question has a shape — you know what would count as an answer, so you know where to look and when you're done.
Beginners gather data and hope insight emerges. It doesn't. Data is enormous and mostly boring, and without a question you'll drown in it and produce a summary nobody acts on.
A good analysis starts with a question that would change a decision. If the answer wouldn't change anything, don't spend the afternoon.
For a small business, almost everything worth knowing lives in five figures. Learn to find these and you can analyse most small operations.
Gross margin. Revenue minus the direct cost of delivering the thing, as a percentage. This is the most important single number in a small business and the most commonly miscalculated, because people disagree about what counts as a direct cost. For a service business it's the labour and tools tied to client work. For a product business it's materials, manufacturing, and fulfilment. Not rent. Not marketing. Not the owner's admin time.
Margin matters more than revenue, and this is the lesson that takes people longest. Two businesses at $800,000 revenue — one at 70% margin, one at 30% — are completely different companies. The first has $560,000 to run on. The second has $240,000. Same headline. Different universe.
Unit economics. What does a customer cost to acquire, and what are they worth over their life with you? When acquisition cost approaches or exceeds lifetime value, growth destroys value — every new customer makes things worse — and the business looks like it's succeeding right up until it doesn't.
Revenue concentration. What share of revenue comes from your largest customer? Above about a quarter, you're not running a business, you're an employee with worse security. This is the risk that kills small businesses fastest and appears nowhere in a standard P&L.
Runway. How many months could you operate if revenue stopped? Most owners avoid this number because it's uncomfortable, and a vague sense that "we'd be fine for a bit" is not a plan.
The owner's real hourly rate. Take net profit, divide by hours actually worked. If you couldn't hire anyone at that rate, the business only functions because you're subsidising it with unpriced labour — which is fine until you want to step back, at which point the whole model collapses.
Here's what the textbooks don't prepare you for: in a small business, the data is a mess, and cleaning it is most of the job.
Personal and business spending share a card. Revenue arrives when clients feel like paying rather than when it was earned. Some expenses are genuinely mixed-use. The owner's compensation might be a draw, a salary, or nothing at all depending on the month. Categories are inconsistent because nobody defined them.
The instinct is to demand clean books before analysing. Don't — you'll wait forever. Work with what exists and be explicit about the uncertainty.
The practical route is to work from transaction data rather than from someone's summary of it. Export bank and card transactions as a CSV and upload them to Cashowa, then run a business profile alongside the personal one — same data, two lenses. That separation alone answers questions most small business owners genuinely cannot answer today, like what the business actually cleared last quarter as distinct from what landed in the account.
And every number Cashowa produces is clickable. The math expands underneath — formula, inputs, the specific transactions that fed it. This matters enormously in small business analysis precisely because the data is messy: when a margin looks wrong, you need to open it and find the miscategorised transaction, not trust a black box. A general-purpose AI will hand you a confident figure it invented and offer no way to check. That's not analysis, that's astrology with decimals.
The most common beginner error is treating one period as a verdict.
A single quarter in a small business is noise. One large invoice, one slow month, one client paying late — any of these swings the numbers enough to tell a completely wrong story. A business can look like it's collapsing because a payment arrived on the second of the month rather than the thirtieth.
So look at rolling periods. Three or four quarters of gross margin tells you something a single quarter cannot: whether the underlying economics are improving or eroding. Everything else is weather.
This is also where the discipline breaks down in practice. Quarterly analysis requires actually doing it quarterly, and assembling the data takes hours, which means it gets postponed. Cashowa's quarterly review does the assembly automatically — re-auditing every ninety days and reporting what improved, what slipped, and what changed. The analysis becomes thirty minutes of reading rather than a day of spreadsheet archaeology, which is the difference between a habit and an intention.
Small business analysis that stops at the P&L misses most of the story, because in a small business the biggest problems are usually operational or structural rather than financial. The financials are where the symptoms show up.
Look at the operations for leaks — the software nobody uses, the process that costs more than it produces, the scope creep quietly eating your margin on every project. Look at where customers actually come from and which channels produce buyers rather than activity. And look hard at the website, because for most small businesses it's the largest unexamined failure point in the whole operation: it's turning away people who were about to buy and it never tells you.
Cashowa's business analyst crawls the site as part of the audit and reports on the SEO, conversion, and trust problems — the things costing you customers you never knew you almost had. That's revenue that doesn't appear as a loss anywhere in your books, which is exactly why nobody ever finds it.
Three habits separate people who analyse well from people who produce reports.
Prefer being wrong early to being confident late. The point of analysis is to find problems while they're cheap. An owner who discovers margin compression in quarter two has options. One who discovers it in year three has a crisis.
Distrust the story you're telling yourself. Every business owner has a narrative about why things are how they are. That narrative is shaped by what's memorable, not what's important. The data exists specifically to challenge it, so when the numbers contradict your story, the numbers are usually right.
Always end with what you'd do differently. Analysis that doesn't change a decision was entertainment. Three specific actions, prioritised by impact against effort, and an honest note about what you're choosing not to do.
What's the first thing I should look at when analysing a small business?
Gross margin, and its trend across the last three or four quarters. It's the fastest read on whether the underlying model is healthy and whether it's improving. Revenue tells you about activity; margin tells you about the business.
How do I analyse a business where the owner's personal and business finances are mixed?
Separate the view even if you can't separate the accounts. Running a business profile against a personal one from the same uploaded transaction data gives you the split without waiting for a bank appointment. Be explicit about anything genuinely ambiguous rather than pretending to precision you don't have.
How much data do I need before analysis is meaningful?
Ideally a year, minimum three or four quarters. Less than that and you're reading weather rather than climate — small business numbers swing too much month to month for a short window to mean anything reliable.
What's the difference between analysing a small business and a large one?
Concentration risk matters far more, the owner's unpriced labour distorts everything, the data is messier, and single events swing the numbers much harder. Corporate frameworks assume clean data and diversified revenue, and applying them uncritically to a small business produces confident nonsense.
Do I need accounting software to do this properly?
Not to start. Categorised transaction data will answer most analytical questions for a long time. Formal accounting becomes necessary for compliance, payroll, and volume — but you can understand your margins, unit economics, and concentration risk long before you need proper books.
How do I know if my analysis is any good?
By whether it changed a decision and whether the change worked. Analysis is judged downstream. A beautiful report that nobody acts on is worth less than one honest number that made someone raise their prices.
What does the tooling cost?
Cashowa's tracking suite is free forever with no card required. The AI features including the business audit run on credits, every account gets free credits each month, and you always see what a task costs before running it. Your data stays yours throughout — no bank login, row-level secured, exportable and deletable at any time.