The quiet ground rules
You already know what accounting is, who reads it, and how the accounting equation holds the five element types in balance. But every one of those numbers rests on assumptions so basic that practitioners rarely say them out loud. Think of them as the rules of a game everyone agreed to before the first hand was dealt — invisible until you ask why a ten-year-old factory still sits on the books at what it cost a decade ago. There are four of these foundational assumptions, and this guide is where we finally name them.
Assumptions are not the same as lies. A good assumption is an honest simplification: it lets us say something useful while openly admitting it is not the whole truth. The four we are about to meet — money measurement, entity, going concern, and periodicity — each trade a little realism for a lot of clarity. The skill you are building is not to forget the assumptions, but to remember exactly where each one bends.
The four foundational assumptions
First, money measurement: accounting records only what can be expressed in a stable monetary unit, and it treats that unit as if it does not change. This is the monetary unit assumption. It is why a loyal customer base, a brilliant founder, or a glowing reputation never appears as an asset — you cannot reliably price them in dollars. It is also why a balance sheet quietly ignores inflation: a dollar booked in 2010 and a dollar booked today are added together as equals, even though they are not.
Second, the entity assumption: the business is an accounting person in its own right, separate from its owners. You met this as the business entity; here it becomes a rule for the books. The owner's personal car is not the company's car; money the owner puts in is the company's debt back to the owner, not the company's income. Without this line, a sole proprietor's grocery receipts and her shop's purchases would blur into one unreadable smear.
Third, going concern: we assume the business will keep operating long enough to use up its assets and pay off its debts in the normal course of things. This going concern assumption is the secret reason a factory can sit on the books at cost and be expensed slowly over its useful life. If instead we expected the company to shut down next month, every asset would have to be revalued at what it could fetch in a fire sale — a very different, much grimmer set of numbers. Fourth, periodicity: even though a business's full story only ends when it closes, users cannot wait that long, so we chop its endless life into tidy reporting periods. This periodicity assumption gives us the month, the quarter, and the fiscal year — and it is precisely this slicing that forces the hard question the rest of this guide answers: which period does a given dollar belong to?
Cash-basis vs accrual-basis
Once you slice life into periods, you must decide when each event lands. There are two honest answers. Cash-basis accounting records a transaction when money moves: revenue when cash arrives, expense when cash leaves. It is simple, hard to fake, and exactly how your bank statement thinks. Accrual-basis accounting records revenue when it is *earned* and expense when it is *incurred* — regardless of when the cash actually changes hands. This is the cash-basis vs accrual-basis choice, and it is the single biggest fork in how the books see reality.
Why does the accrual view win for serious reporting? Because economic reality and cash timing routinely disagree. You can do valuable work in March and not be paid until May; you can pay a year's rent in January for an office you will use all year. Cash-basis would call March a poor month and January a catastrophe, even though nothing real went wrong. Accrual-basis lines each effect up with the period that actually caused it — the engine behind this is the matching principle, which pairs an expense with the revenue it helped produce, and its partner, revenue recognition, which fixes the moment a sale truly counts.
A tiny timing example
Picture Mara, a freelance designer. In March she finishes a logo and emails the client a 900-dollar invoice; the client pays in April. Also in March she pays 300 dollars upfront for three months of cloud hosting she will use across March, April, and May. Under cash-basis, March shows zero revenue (no cash in) and a 300-dollar expense (cash out) — a 300-dollar loss for a month she actually did great work. That reading is, frankly, a lie about her March.
Now read March on the accrual basis. The 900 dollars was *earned* in March when she delivered the logo, so it is March revenue — even though no cash arrived; the unpaid amount becomes a receivable, a promise the client owes her. The hosting is trickier: only one of three months was *used* in March, so just 100 dollars is March expense. The remaining 200 dollars is not an expense yet — it is a prepaid asset, future benefit she has already paid for. March now shows 900 of revenue against 100 of expense: an 800-dollar profit. That is the truth of her March, and it is the accrual basis doing exactly its job.
Cash-basis Accrual-basis
March revenue 0 900 (earned, not yet paid)
March expense 300 100 (1 of 3 hosting months used)
----------------- ------- -------
March result -300 +800
Still on the books at March 31:
Receivable (client owes) 900
Prepaid hosting (2 months left) 200Honest limits and a look ahead
Accrual accounting buys accuracy with judgment, and judgment can be stretched. Deciding *when* something is earned or used is sometimes genuinely fuzzy, which is exactly where aggressive companies push and where auditors push back. Cash-basis, for all its blind spots, has one virtue accrual lacks: cash is hard to argue about. Most very small businesses and individuals are even allowed to use cash-basis for simplicity and tax. The point is not that accrual is holy and cash is wrong — it is that each answers a different question, and serious financial reporting needs the question accrual answers.
Notice that Mara's example left two loose ends hanging at month-end: a 900-dollar receivable and 200 dollars of still-unused hosting. Those leftovers are not a flaw — they are precisely the threads a later step in the accounting cycle will pick up. When you reach the rung on adjusting and closing, the entries that move earned-but-unrecorded revenue and used-up prepaid costs into the right period are simply this same accrual idea, applied with a pen at period-end. You have just learned, in plain words, the principle that the machinery later automates.