Chart of accounts for cafes and retail: the categorization rulebook that decides whether your P&L lies to you.

Most cafe and retail owners outsource the chart of accounts to their accountant and never look at it again. That is a mistake. The chart of accounts is the categorization rulebook your P&L is built on — the list of buckets every euro flowing in or out of the business gets sorted into. Sort wrong and the same euro lands in a group it does not belong to: card fees inflate revenue, inventory purchases crater EBIT in the month you bought stock, owner salary disappears from the cost stack and the business looks profitable when it is not. The monthly P&L still prints clean. The numbers on it are quietly lying to you. This guide is the practical, plain-English version of what a chart of accounts is, the five groups every transaction belongs to, two full sample charts for a small cafe and a small boutique, the seven mistakes that wreck most owner-managed charts, and how nouz maps the same logic into the four daily buckets you actually act on.

TL;DR

• What it is: the master list of buckets — Revenue, COGS, Operating Expenses, Assets, Liabilities — that every transaction in your business gets sorted into.
• Why it matters daily: the bucket a euro lands in decides whether it hits gross margin, prime cost, fixed cost or the asset register. Same euro, four different stories.
• How many categories: for a small cafe or boutique, 8-12 revenue, 6-10 COGS, 12-18 operating expense lines. Fewer than that and you cannot see what is moving; more than that and nothing gets categorized consistently.
• The 7 traps: card fees in expense (inflates revenue), inventory in expense (craters EBIT), owner salary missing (overstates profit), equipment in expense (volatile P&L), VAT in revenue (overstates revenue), tips in revenue (overstates revenue, hides a liability), personal mixed with business (year-end hell).
• How nouz handles it: four daily buckets — revenue, COGS, variable costs, fixed costs — that map cleanly into any accountant's chart. You operate from the four; the accountant maps to the full chart for tax filing.

What a chart of accounts is, in plain English

A chart of accounts is the master list of categories every euro flowing into or out of your business gets classified under. Think of it as the labelled drawers in a filing cabinet. Every receipt, every invoice, every till total, every supplier payment, every payroll run gets dropped into one of the drawers. The set of drawers — not the receipts themselves — is the chart of accounts.

In an accounting system there are five kinds of drawers. Revenue accounts hold what came in from sales. Cost of goods sold (COGS) accounts hold what the things you sold actually cost you. Operating expense accounts hold the cost of being open — rent, payroll, utilities, software, insurance. Asset accounts hold what the business owns — the espresso machine, the inventory on the shelf, the cash in the till. Liability accounts hold what the business owes — supplier invoices not yet paid, the bank loan balance, VAT collected but not yet remitted.

Together those five groups cover every euro that ever touches the business. There is no transaction that does not belong somewhere. The question is never "does this fit in the chart?" — it always does. The question is "which bucket exactly?" — and the consequences of getting that wrong are the rest of this article.

The 5 top-level groups

Every account in any chart of accounts belongs to exactly one of five top-level groups. The same five groups apply to a single-location cafe and a multinational. The number of accounts inside each group changes; the groups themselves do not.

Three of those groups feed the P&L (revenue, COGS, operating expenses) and the other two feed the balance sheet (assets, liabilities). The P&L answers "did the month make money?" The balance sheet answers "what did the business have and owe at the end of the month?" Most owner decisions live on the P&L side, but the balance sheet matters too — buying €4,000 of inventory in January is not an expense, it is moving cash from one asset (the bank) to another asset (the shelf). Get that distinction wrong and January's P&L tanks for no real reason.

Why this matters for daily decisions

It is tempting to think the chart of accounts is a back-office concern — accountant territory, nothing to do with running the shop on Tuesday. That is exactly the assumption that gets owners into trouble. The categorization a transaction gets at the moment of entry decides everything downstream of it: whether it hits gross margin, whether it shows up in prime cost, whether it gets absorbed into the daily fixed cost slice, or whether it never touches the P&L at all because it landed on the balance sheet.

Take a €600 espresso machine repair. Categorize it under "Repairs and maintenance" (an operating expense) and it hits this month's P&L as €600 of variable cost — one bad-looking month, fine, you move on. Categorize it under "Equipment" (an asset) and treat it as extending the machine's useful life, and it depreciates across the next 36 months at roughly €17/month, smoothing the hit across three years. Same €600, two completely different P&L stories. Neither is wrong, exactly — but the choice changes how the month reads and whether you wrongly conclude that "March was a bad month."

Or take a €0.48 card fee on a €32 brunch sale. Categorize the €32 as gross revenue and the €0.48 as an operating expense and you have just overstated your revenue by €0.48 for that sale. Multiply across 18,000 transactions a year and you have overstated revenue by €8,640 and understated the true cost-to-collect by the same amount. The annual numbers eventually reconcile because the total is the same — but every interim margin calculation is wrong, every benchmark against industry numbers is off, every "how much did we really earn per transaction" answer is misleading. The right treatment is to net card fees against revenue (or sit them in COGS) so the gross-to-net walk is honest.

These are not theoretical edge cases — they happen every week in every small shop. The chart of accounts is the rulebook that decides them. Get it set up right once and every transaction sorts itself correctly forever. Get it wrong and you spend the next three years wondering why your margin numbers do not match what you see on the till.

A working chart of accounts: cafe

Below is a sample chart of accounts for a single-location cafe doing roughly €15,000-€50,000 in monthly revenue, one to two staff, with light retail (beans, mugs) on the side. It is deliberately small — about 35 lines — because the chart that gets used consistently beats the chart that has every possible category but ends up half-empty. Treat this as a starting point, not gospel; adapt the labels to match what you actually sell and pay for.

Revenue (5 accounts)

COGS (4 accounts)

Operating expenses — fixed (12 accounts)

Operating expenses — variable (5 accounts)

Assets (4 accounts)

Liabilities (3 accounts)

Thirty-three lines, organized into five groups, covering everything a typical small cafe touches. Notice what is not on the list: no separate account for every milk supplier, no separate revenue line for each menu item, no twelve-line breakdown of cleaning supplies by sub-category. A category gets its own line only if the cost or revenue inside it is material enough to track on its own. Everything else gets folded into the closest umbrella line.

A working chart of accounts: retail boutique

Now the same exercise for a small fashion or homeware boutique — roughly €10,000-€40,000 monthly revenue, one or two staff, with a small Shopify side. The structure is identical (five groups, same five questions) but the lines inside each group change to match the operating reality.

Revenue (4 accounts)

COGS (3 accounts)

Operating expenses — fixed (11 accounts)

Operating expenses — variable (6 accounts)

Assets (3 accounts)

Liabilities (3 accounts)

Thirty lines for the boutique. The two charts look structurally similar — same five groups, similar account counts — but the inside of each group reflects what each business actually does. A cafe has a deep COGS section because ingredients dominate its variable cost; a boutique has a deeper assets section because inventory is its biggest asset and tracking it accurately is the difference between a real margin number and a fictional one. Use the structure; adapt the lines.

The 7 most common categorization mistakes

Across hundreds of owner-managed charts of accounts, the same seven mistakes show up over and over. They are not subtle. Each of them produces a P&L that looks plausible — totals add up, columns reconcile — but which structurally misrepresents what happened. Owners often live with the consequences for years without realising the chart itself is the problem.

1. Card fees in operating expense instead of revenue contra

The most common mistake. A customer pays €40 by card, the processor takes €0.60, the business receives €39.40. The right treatment: revenue is recognised at €40 (gross), and €0.60 is netted against revenue (or recorded in COGS) to land at €39.40 net revenue. The wrong treatment: revenue is recorded at €40 gross and the €0.60 is dropped into "Bank charges" or "Operating expense — card fees." Both arrive at the same EBIT in the end, but the wrong treatment overstates revenue by every card fee for the year — typically 1.5-3% of total card sales. The gross-to-net walk is broken, every margin ratio that uses revenue is off, and comparisons to industry benchmarks for "revenue per transaction" become misleading. Fix: net card fees against revenue, or treat them as variable COGS-adjacent. Either way, not operating expense.

2. Inventory purchases in expense instead of asset

A boutique buys €4,000 of new-season stock in January. The wrong treatment: drop the entire €4,000 into "Inventory purchases — expense" and watch January's P&L crater. The right treatment: the €4,000 buys an asset (inventory on the shelf). Cash moves from one asset (the bank) to another asset (inventory). Nothing hits the P&L yet. As items are sold across the spring, the wholesale cost of each item sold moves from inventory (asset) to COGS (P&L expense). If the boutique sells €1,500 of those items in January, only €1,500 hits January's COGS — not the full €4,000. The wrong treatment makes January look catastrophic and the months that follow look artificially profitable, because the cost of February's sales is already gone from the system. Fix: inventory purchases are an asset move, not an expense. Cost recognition happens when the item is sold, not when it is bought.

3. Owner salary in personal account instead of operating expense

You work 50 hours a week in the shop and pay yourself either nothing or "whatever is left at month-end." The wrong treatment: nothing on the P&L at all, or a "Drawings" line that is treated as equity rather than expense. The right treatment: a fixed monthly "Owner salary" line under operating expenses, set at the market-rate cost of replacing yourself — typically €2,500-€4,500/month depending on hours and role. Even if you do not actually pay yourself that amount, the cost should still be on the P&L for management accounting. Without it, EBIT is flattering by exactly the amount of your unpaid labour. The business looks profitable when it is actually breaking even on your back. EBIT explained goes deeper on the owner-salary trap. Fix: add the line; let it depress EBIT if needed; act on the honest number.

4. Equipment in expense instead of asset and depreciation

A cafe buys a new espresso machine for €8,500 in March. The wrong treatment: drop the full €8,500 into "Equipment — expense" and watch March's P&L go negative. The right treatment: capitalise the machine as an asset, then depreciate it across its useful life (typically 5-7 years for catering equipment). At 5 years that is €142/month of depreciation — a steady, sensible line on every month's P&L for the life of the machine. The wrong treatment produces one terrible month followed by 59 months that look unrealistically good because the equipment cost is invisible. Pricing decisions made off those flattering months are wrong; the next equipment purchase is treated as a shock instead of a planned-for cost. Fix: anything with a useful life over one year and material value (typically €500+, but the threshold depends on jurisdiction) is an asset that depreciates, not an expense.

5. VAT in revenue (instead of as a separate liability)

A customer pays €12 for a sandwich at a VAT-inclusive price. The wrong treatment: record €12 of revenue. The right treatment: record €10 of revenue (the net portion that belongs to the business) and €2 of VAT accrued (a liability, owed to the tax office at the next return). Recording the full €12 as revenue overstates the top line by exactly the VAT rate — 10%, 19%, 20%, 21% depending on jurisdiction and category — and inflates every margin ratio that uses revenue as the denominator. Worse, it makes the VAT remittance feel like an expense ("we paid €2,400 to the tax office in March") when in reality the business was only ever the collector, not the owner, of those euros. Fix: revenue is always net of VAT; VAT collected is a liability that sits on the balance sheet until it is paid over. Gross vs net revenue walks through the full stack.

6. Tips in revenue (instead of as a pass-through liability)

A customer leaves a €3 tip on their card. In most jurisdictions and in most business arrangements, that €3 belongs to the staff — the business is just the collector. The wrong treatment: record €3 of revenue and €3 of "Tip payout" expense, leaving EBIT unchanged but overstating both revenue and operating expenses. The right treatment: the €3 never touches revenue or expense — it lands as cash received (asset increase) and tips payable (liability increase), then nets out when paid to staff (asset decrease, liability decrease). Tips pass through the business without ever being its money. If tips are kept by the owner (some markets, some structures), they are revenue — but for the standard cafe and restaurant arrangement where tips go to staff, the right treatment is pass-through. Fix: tips that belong to staff sit on the balance sheet as a liability until paid out, not in revenue.

7. Personal and business mixed in the same account

The owner uses the business card to buy fuel for personal errands. The business pays the owner's phone bill even though half the use is personal. A €60 lunch with the family ends up coded as "Marketing." The wrong treatment: leave it all in business expense accounts and let your accountant untangle it at year-end. The right treatment: keep personal and business strictly separate from day one. If a personal expense lands on the business card by mistake, code it as "Owner drawings" (a reduction in equity) so it does not pollute the P&L. Owners who let the two mix produce a P&L that misrepresents real costs by 5-15% — and a year-end clean-up that costs hundreds of euros in accountant time to unwind. The damage is not the individual transactions; it is the loss of any reliable signal on what the business actually costs to run. Fix: separate cards, separate accounts, separate brains.

Granularity: too few categories vs too many

A chart of accounts can fail in two opposite directions. Too few categories and you cannot see what is moving — "Operating expenses" is one €8,400 number per month with no breakdown of whether the move was rent, payroll or marketing. Too many categories and nobody categorises consistently — every transaction takes 30 seconds to file because the chooser has to scroll through fifty options, so within three months 30% of transactions end up in "Miscellaneous" and the breakdown is meaningless anyway.

The sweet spot for a small cafe or boutique sits in a narrow band:

The principle: split when the split changes a decision. Splitting "Beverage sales" into "Coffee" and "Other beverages" matters because the two have very different margin profiles and trend differently across the year. Splitting "Coffee" into "Espresso," "Cappuccino," "Latte" and "Americano" rarely changes a decision — the menu engineering happens off product-level POS data, not off the chart of accounts. Resist splits that produce data nobody will act on. The job of the chart is to support decisions, not to demonstrate diligence.

If you ever find yourself adding a new account because a single transaction needed it, that is a smell. The right move is usually to fold the transaction into the closest existing line. Add an account only when the cumulative volume in a category is material enough that "where did this go?" is a question you ask more than once a quarter.

How to set up your chart of accounts from scratch

If you are starting from nothing — no inherited chart, no template, just a real business and a set of bank statements — the setup is six concrete moves. Block ninety minutes, open three months of bank and card statements, and walk through:

1. 01
   undefined

   Read your last three months of bank deposits and till summaries. Every distinct way money comes in becomes a candidate revenue line. A cafe might have: food, coffee, other drinks, retail, catering. A boutique: clothing, accessories, online, alterations. Five to eight lines is usually right. If you find yourself listing more than ten, group the smaller ones.

2. 02
   undefined

   Walk through three months of bank outflows. Identify everything that comes out on a regular cadence — rent, payroll, social contributions, software subscriptions, insurance, accountant, utilities. These become your fixed operating expense lines. Annual invoices count too: divide them by 12 and treat as monthly. Aim for 10-15 lines.

3. 03
   undefined

   Bank outflows that are not recurring — supplier deliveries for COGS, packaging suppliers, marketing spend, hourly staff payments, repairs, cleaning supplies. Group them into 5-8 variable cost categories. The grouping rule: same category if the spend would scale together with sales.

4. 04
   undefined

   Walk through the physical shop. Espresso machine, fridges, oven, dishwasher, furniture, signage, POS hardware, fit-out, mirrors, rails, mannequins. Anything with a useful life over one year and material value (typically over €500) goes on the asset register. Note purchase price and estimated useful life for each.

5. 05
   undefined

   Put every line you have listed into one of Revenue, COGS, Operating Expenses (split fixed/variable), Assets, Liabilities. The chart is now drafted. Cross-check: every account has exactly one group, every group covers a clear question, no orphan lines.

6. 06
   undefined

   Show the draft to your accountant. They may want to add specific account codes that match their jurisdiction's tax filing format, or split a line for tax-treatment reasons. The structure stays yours; the accounting codes layer on top. Ninety minutes once, then the chart runs itself.

Once the chart is set up, the discipline is to categorise every new transaction against it the same day. The longer the gap between transaction and categorisation, the more likely it gets dropped into "Miscellaneous" — and "Miscellaneous" is the dead-end of every good chart. If "Miscellaneous" creeps above 5% of transactions in any month, you are missing an account; add it before the misc pile becomes the actual chart of accounts.

How nouz handles categories

nouz is not a full accounting system — it does not produce statutory balance sheets, it does not file tax returns, it does not replace the accountant. What it does is operate the daily P&L layer in real time, using a deliberately focused operating set of four buckets that map cleanly into any full chart of accounts your accountant maintains.

Four buckets is the minimum granularity needed to produce honest daily EBIT. Revenue minus COGS gives gross profit. Gross profit minus variable costs minus today's slice of fixed costs gives EBIT. That is the whole formula. Every euro that flows through the shop lands in one of those four buckets the day it happens; the daily P&L falls out automatically. The fifth and sixth groups in a formal chart — assets and liabilities — sit with your accountant, get touched at month-end and year-end, and do not interfere with the daily operating signal.

In practice this means an owner can run nouz daily without ever opening their full chart of accounts. The chart still exists — your accountant maintains it for the annual filing — but the daily operating layer is the four buckets, and they map back into the chart cleanly whenever the accountant pulls a monthly export. Fixed costs are entered once with a start_date and optional end_date, so a cost that begins on the 14th starts slicing into the daily floor from the 14th, and a cost that ends on March 31 stops slicing into April. COGS is snapshotted at the moment of sale, so updating a product cost tomorrow never rewrites yesterday's P&L. The two design rules — active-window for fixed costs, snapshot for COGS — are what make the daily layer reliable enough to act on without doubling up the accountant's work.

For deeper context on the formulas that sit on top of the chart, EBIT explained walks through the operating profit formula and the owner-salary trap. Fixed vs variable costs covers the split and the daily slice mechanic. COGS snapshot explained covers why product cost edits never rewrite history. Gross vs net revenue covers the VAT and card-fee stack on the revenue line. Depreciation for non-accountants covers how equipment moves from asset to expense over its useful life. Accruals vs cash basis covers when costs are recognised versus when they are paid. How to read a P&L statement covers the resulting report end-to-end. Fixed vs variable costs — small business gives the sector-by-sector worked examples. The cross-vertical synthesis is the daily P&L pillar guide.

If you want to sanity-check the operating side of your chart against your actual numbers, the operating expense ratio calculator takes net revenue and total operating expenses and returns the ratio that a properly categorised chart should produce. Healthy small cafes land at 70-85%; healthy small boutiques at 60-75%. If your ratio is wildly outside those bands, the chart is usually the first place to look. And if you want to run the four-bucket daily layer on your own shop, the pricing page is monthly only — no annual lock-in for a tool you can stop using tomorrow.

FAQ