The one rule that runs every transaction.
Double-entry bookkeeping says every transaction touches at least two accounts, and total debits must equal total credits. That is the entire rule. The reason it exists is self-correction: if you record a sale of $100,000 against cash, you also record $100,000 of revenue. The numbers must agree, or you missed something. A single-entry ledger cannot catch a typo because there is nothing to compare against. A double-entry ledger catches it the moment the trial balance refuses to add up. That is why banks, auditors, and your tax authority all assume your books are double-entry.
The five account types are assets, liabilities, equity, revenue, and expenses. Debits increase assets and expenses. Credits increase liabilities, equity, and revenue. Memorize that and you can decode any journal entry in the country. Nonari builds every transaction on this rule and refuses to save anything that does not balance.
Worked example: a cash sale at a Atlanta shop.
Say your Tariq Road branch sells a kettle for $8,500 cash. Two accounts move. Cash goes up by 8,500 (asset increase, so debit). Sales revenue goes up by 8,500 (revenue increase, so credit). The journal entry is: DR Cash 8,500 / CR Sales 8,500. That is it. If the kettle cost you $5,200 from the supplier, a second entry fires when the inventory ledger relieves stock: DR Cost of Goods Sold 5,200 / CR Inventory 5,200. Now both your P&L and your balance sheet update in the same transaction.
A spreadsheet would let you record only the cash side and forget COGS. Six months later your gross margin looks like 100% and your inventory ledger looks like nothing ever sold. Double-entry forces both halves to fire together.
Where SMBs actually lose track.
The pain points repeat across every shop we have audited. Owner takes $50,000 from the till for personal use and writes nothing. That is a debit to drawings or owner equity, credit to cash. Skip it and equity is overstated by 50,000 forever. Owner pays a supplier from his personal HBL account: that is a debit to the expense or inventory, credit to owner contribution. Skip it and the expense never hits the books. A bank charge of $350 hits the statement: debit bank charges, credit cash. Skip it and reconciliation breaks.
These are not exotic edge cases. They are the everyday transactions that cause a January trial balance to be off by $73,224 and nobody knows why. Double-entry catches them only if every leg is recorded.
- Owner draws cash from till without an entry.
- Personal account pays a business expense.
- Bank charges, FED, or withholding tax deducted at source.
- Inventory shrinkage written off mentally, not in books.
- Returns and refunds posted on one side only.
Debits and credits, the cheat sheet.
Memorize the directions. Assets: debit to increase, credit to decrease. Liabilities: credit to increase, debit to decrease. Equity: credit to increase, debit to decrease. Revenue: credit to increase. Expenses: debit to increase. The famous accounting equation Assets = Liabilities + Equity is a direct consequence. Anything that increases the left side (debit asset) must either increase the right side (credit liability or equity) or decrease the left side somewhere else (credit another asset).
When in doubt, ask: what did the business get, and what did it give up? You got cash, you gave up inventory. You got a service, you gave up a future cash payment (so a payable). Once you frame each transaction as get/give, the debit and credit follow automatically.
Why your trial balance is the daily smoke test.
A trial balance lists every account with its debit or credit balance. Total debits should equal total credits. If they do not, you have a missing leg, a transposition error, or a one-sided entry. SMBs that do month-end close on a spreadsheet usually find their trial balance off by some round number, and they spend a full day hunting. With double-entry software the trial balance balances by definition because the system rejects any entry where DR does not equal CR.
That does not mean the books are correct, only that they are arithmetically consistent. You can still post to the wrong account. But the universe of errors is now small enough to debug. Run the trial balance every Friday afternoon, and a problem from Monday cannot hide for four weeks.
Software vs spreadsheets, honestly.
Excel can do double-entry if you are disciplined. Most owners are not, because life happens. The day a customer disputes a $250,000 invoice and you need to trace 14 related entries across 8 sheets, you will switch. Modern accounting software enforces the rule at the database level, gives you a trial balance on demand, and keeps an audit log on every change. Nonari does this with multi-branch isolation, so the Manchester branch books cannot accidentally bleed into the Atlanta books.
The migration cost is one weekend of opening balances and a chart of accounts setup. The payback is every reconciliation, every audit, every loan application, and every tax filing for the rest of the company life.
Your first 30 days on a real ledger.
Week one: enter opening balances from your last spreadsheet. Cash, bank, inventory, payables, receivables, owner equity. Run the trial balance, confirm it ties. Week two: enter every sale, purchase, expense, and bank deposit as it happens. Resist the urge to batch-enter at month end. Week three: reconcile the bank account against the statement. Every cleared transaction matches; every uncleared one is investigated. Week four: run the trial balance, P&L, and balance sheet. Compare against your gut feel of how the month went.
By day 30 you will trust the books for the first time. That is when an SMB graduates from "I think we made money" to "we made $412,000 net of tax, here is the proof." Nonari keeps every step on rails so the first 30 days feel like a checklist, not an exam.