Chart of accounts granularity: how many accounts is too many?

Chart of accounts granularity is the level of detail your account structure carries — whether "Expenses" is thirty accounts or three hundred. Get it too coarse and reports say nothing useful; too fine and every coding decision becomes a judgement call that different people make differently. Most messy ledgers aren't messy because people are careless. They're messy because the chart made consistency hard.

The three failure modes

Too coarse. If "General Expenses" is your biggest expense account, the P&L can't answer basic questions — is spend up because of software, travel, or subcontractors? You find out at year end, from the accountant, expensively.

Too fine. An account per supplier, or "Office Expenses — Sydney — Kitchen", multiplies decisions. Every extra plausible account is another way for two people to code the same purchase differently, and thinly-used accounts make trend reporting meaningless — a chart where half the expense accounts have fewer than five transactions a year is a strong signal.

Overlapping. The worst of the three, and the most common: "Software", "IT Expenses" and "Subscriptions" all live in the same chart, all defensible for the same Xero invoice. Where two accounts could both be right, both will get used, and the split between them is noise.

A practical test for each

1. The question test (too coarse): pick the three questions the owner actually asks — "what are we spending on contractors?", "what does the ute cost us?" If the chart can't answer them without exporting to a spreadsheet, it's too coarse where those questions live.
2. The five-transaction test (too fine): run a transaction count per account for the last year. Accounts under five transactions are candidates for merging — with an exception for accounts that exist for tax or compliance reasons (FBT-relevant categories, entertainment), which earn their place regardless of volume.
3. The new-hire test (overlapping): for each common purchase type, could a competent new bookkeeper pick the right account from the names alone? If two names are both plausible, merge them or rename them so the boundary is explicit — "Software — subscriptions" vs "IT — hardware & repairs" draws a line; "Software" vs "IT Expenses" doesn't.

Fixing granularity is ordinary maintenance: merge, rename, archive, and update the bank rules that point at retired accounts. Do it at a period boundary so comparatives stay readable, and expect the first month afterwards to surface a few rules and repeating invoices still aimed at the old accounts.

Where detection fits

Granularity problems show up statistically before anyone names them: overlapping accounts produce the same supplier split across accounts, and over-fine charts produce clusters of near-empty accounts. Ledger Optics surfaces both as part of its coding-consistency scan of a Xero or QuickBooks file — it can tell you that "Software" and "Subscriptions" are being used interchangeably and show you the transactions, though deciding which account survives the merge is a judgement about your business, not a statistic.

FAQ

How many accounts should a chart of accounts have? There's no universal number — the test is functional. Every account should either answer a question someone actually asks of the reports or exist for a tax/compliance reason, and no two accounts should plausibly claim the same routine purchase. For most small businesses that lands well under a hundred active accounts.

What problems does poor chart of accounts granularity cause? Too coarse and reports can't explain where money goes; too fine and coding becomes inconsistent because every decision has multiple plausible answers; overlapping names cause the same supplier to be spread across accounts, which distorts the P&L and slows year-end.

Ledger Optics is a free cleanup app for Xero and QuickBooks that shows you which accounts are being used interchangeably. Connect a file to see yours.