Saving money without a spreadsheet: what to automate and what still needs you

Spreadsheets fail on entry, not arithmetic

A budgeting spreadsheet almost never fails because the formulas were wrong. It fails because a spreadsheet is a record that only exists when someone types into it, and by the third week the typing has fallen behind, the totals are stale, and a stale budget is one you stop trusting and then stop opening. The abandonment is not a discipline problem; it is a design problem. The tool put the most tedious task on the critical path.

So the first move is not a better spreadsheet. It is to remove manual entry from the loop altogether. Every purchase you make already produces a transaction record at your bank; the job is to have that record sorted and totalled without you copying it.

What automatic categorisation replaces

Automatic categorisation takes each transaction as it lands from your bank and assigns it to a category — groceries, transport, subscriptions, eating out — using the merchant, the amount and the pattern of previous transactions. It replaces the entry and the totalling: the category totals are always current because they update as the transactions arrive, not when you sit down.

It also fixes two things a hand-kept sheet gets wrong. Transfers between your own accounts and card repayments are recognised as movements rather than spending, so they do not inflate a category. And recurring charges are detected as they repeat, so a subscription appears as a known recurring item rather than a line you have to remember to add each month.

Where it needs a check is on the edges: a merchant that sells across categories, a new merchant it has not seen, a cash withdrawal with no merchant at all. Reviewing those few once a week takes minutes, and each correction teaches the sorting for next time.

What per-category limits replace

The other thing a spreadsheet did was hold a target next to each total. A per-category limit set before the period starts does the same job continuously: the category shows its limit, what has been spent, and how far through the period you are, so you can see mid-period whether the current pace lands inside the number.

This is the part that actually saves money, and it is worth being precise about why. Categorisation on its own is a tidy record of a period you have already spent. A limit set in advance changes the decision at the point of purchase, because you know the number before the shop rather than after. The saving comes from the limit; the categorisation just makes the limit checkable without effort.

The three decisions that still need you

First, the limits themselves. Automation can draft them from your own last three months, which is a far better starting point than a blank cell, but where to set each one — at the current figure, or a little under — is a choice about what you want to change, and no tool can make it for you.

Second, the savings target. Decide the amount to move to savings each period and have it transferred on payday, before the categories see it. A spreadsheet treated savings as whatever was left at the end; making it the first line instead is the single most reliable way to save, and it is a decision, not a calculation.

Third, what to do when a category runs hot. When the pace says a limit will be breached, there are three honest responses — slow down in that category, move some allowance from a category that is running under, or accept the breach and re-set the limit next period if the number was wrong. Choosing between them is the budgeting. Everything else is bookkeeping.

What this looks like in practice

A workable routine without a spreadsheet takes about ten minutes a week. Payday: the savings transfer runs automatically. Once a week: glance at each category's used-versus-limit against the point in the period, correct any transactions the sorting got wrong, and make one of the three decisions for any category running hot. Once a period: re-set any limit that was breached three periods running, because that is a wrong number, not a lapse.

In Fin, transactions are categorised as they land, per-category limits are set before the period opens, recurring charges are flagged as they repeat, and the budget-versus-actual view shows each limit against what has been spent so far. That covers the recording. The limits, the savings target and the response to a hot category are still set by you, which is exactly the part worth keeping.