JOVANA
Library Glossary Getting Started Three Levels Fields How it works Mission
Join the mission
All guides

Perpetual vs Periodic Inventory Systems

Before we argue about FIFO or LIFO, we have to answer a plainer question: when does a business actually know what it has left and what its sales cost? Two systems give two very different answers — one watches every sale as it happens, the other waits until the shelves are counted.

Two ways to keep score

You already know, from the income-statement rung, that cost of goods sold is usually a retailer's single biggest expense, and that it lives right under sales: revenue minus the cost of what you sold equals gross profit. And from the glossary you know inventory is carried at *cost*, not at the hopeful price on the tag. This rung is about the machinery underneath those facts — and the very first decision a merchant makes is not *which cost-flow assumption* to use, but *when the books are allowed to know the answer at all*.

There are two answers, and they name the two systems. A [[perpetual-inventory-system|perpetual system]] keeps a running score: every time a unit comes in or goes out, the inventory account is updated *on the spot*, the way a scoreboard ticks over the instant a goal is scored. A [[periodic-inventory-system|periodic system]] keeps no running score during play; it simply waits until the period ends, walks the floor, counts what survives, and works the cost of goods sold out backward from there. Same goods, same purchases, same sales — but one knows continuously, and the other knows only at the whistle.

The perpetual system: every sale moves cost in real time

Picture the supermarket scanner. The instant the cashier beeps a carton of milk, two things happen, not one. First, the obvious thing: a sale is recorded — say 3 of revenue. But a second, quieter entry fires at the very same moment: the milk's *cost* — say 2 — is lifted out of the inventory account and dropped into cost of goods sold. The perpetual system never lets a sale leave the building without also moving its cost. That is what "perpetual" really means: continuously, transaction by transaction, the books are kept current.

PERPETUAL -- one sale, TWO entries fire at once:

  (1) record the sale
      Dr Cash / Accounts receivable   3
          Cr Sales revenue                3

  (2) move the cost out of inventory, same instant
      Dr Cost of goods sold           2
          Cr Inventory                    2

At any moment:  Inventory account balance = goods on hand (in cost)
                COGS account balance      = cost of all goods sold so far
In a perpetual system every sale triggers a paired entry: the second line is what keeps inventory and cost of goods sold current to the minute. Notice the debit to cost of goods sold and the credit to inventory — a reminder from the double-entry rung that a debit is not automatically an "increase"; here it raises an expense while the credit *lowers* an asset.

The payoff is control. Because the inventory account is always live, a manager can ask at lunchtime "how many phones are left, and how much gross profit has today produced so far?" and get a real answer without anyone walking the stockroom. The system flags low stock for reordering before the shelf empties, and it lets the business compare what the books *say* is on hand against what is *actually* there. That comparison is where shrinkage — theft, breakage, spoilage, miscounts — finally surfaces, instead of hiding.

The periodic system: count first, then compute backward

Now strip out the scanner. In a periodic system the inventory account is left *untouched* all period long. Purchases of goods do not go into inventory; they pile up in a separate "purchases" account. And here is the part beginners find strange: when a sale happens, the cashier records the revenue but makes *no* entry for cost. Nothing is lifted out of inventory at the moment of sale, because there is no running cost record to update. The cost side of the story is simply postponed.

So how is cost of goods sold ever found? At period-end, the business physically counts what remains, and then works backward through the cost of goods sold formula. Start with what you had at the beginning, add what you bought, and that grand total is the [[cost-of-goods-available-for-sale|cost of goods available for sale]] — the whole pool that *could* have been sold. Subtract the ending count, and whatever is missing from the shelf is *assumed* to have been sold. A worked example in words: begin with 2,000, buy 8,000 more, count 3,000 left at the end. Then cost of goods sold = 2,000 + 8,000 − 3,000 = 7,000. One backward calculation, done once.

When each is practical — and why barcodes tipped the balance

For most of accounting's history, the periodic system won by default, because the perpetual system was simply *too expensive to run by hand*. Imagine a 1950s hardware store: to keep a perpetual record, a clerk would have to write down the cost of every single nail, washer, and tin of paint the moment it left the counter, all day, every day. The bookkeeping labor would dwarf the value of knowing. So small and mid-sized merchants used the periodic system: record purchases, sell freely, and count the shelves a few times a year. Cheap, simple, good enough.

Then came the barcode and the point-of-sale terminal. The barcode (first scanned on a pack of chewing gum in 1974) turned that crushing clerical burden into a single beep. Now the cash register *is* the bookkeeper: scanning an item simultaneously rings the sale and tells the accounting software exactly which product, at what cost, just left the building. The marginal cost of keeping a perpetual record collapsed from "a full-time clerk" to "essentially zero." That is the real reason perpetual became the norm — not a change in accounting theory, but a change in the price of information.

So who still chooses periodic? A handful of niches where scanning earns its keep poorly: a tiny market stall, a craft business with a few sales a day, or an operation whose stock is so low-value or so uniform that tracking each unit is pointless — think a sand-and-gravel yard. For nearly everyone else with a real product business, perpetual is now the default, baked straight into the software. Each unsold item, recall, still sits as inventory on the balance sheet under either system; the systems differ only in *when and how* the books learn that an item has gone.

Why both systems still walk the floor and count

Here is the twist that surprises people: the physical count never goes away, even under a perpetual system. The roles it plays are simply different. Under the *periodic* system, the count is the *only* way to find cost of goods sold at all — without it, the books literally cannot compute the number, because the cost side was postponed. The count is not a check; it *is* the measurement. The cash register knows revenue; only the warehouse floor knows what was sold.

Under the *perpetual* system, by contrast, the books already claim to know what is on hand — so the count becomes a *reality check*. The whole point of "perpetual" is that the inventory account is continuously updated, but updated is not the same as *correct*. The scanner cannot see a shoplifter, a dropped bottle, a wrong key-press, or a box that two clerks both forgot to scan. So once a year (often at quiet times, or rotating shelf by shelf as a "cycle count"), staff physically count the goods and compare reality against the books. The gap is recorded as an inventory shrinkage adjustment — an entry that lowers inventory and raises cost of goods sold (or a loss) to bring the records back to the truth on the shelf.