When a spreadsheet stops being enough for your books
Most businesses in the region started their books on a spreadsheet, and at the time that was entirely the right call. The trouble is that the moment it stops being enough arrives without warning — the file keeps opening and keeps calculating, and it has quietly started to lie.
The Fatouraty team
There is nothing wrong with keeping books in a spreadsheet. It is free, it opens instantly, it imposes no structure you did not choose, and anyone can understand it in minutes. A great many serious businesses in the region ran their first years on one with complete success. But a spreadsheet is a calculation tool, not a recording tool, and the difference between the two does not show on day one — it shows on the day you need to prove what happened.
What a spreadsheet is genuinely good at
This deserves saying plainly, because conversations about alternatives usually overstate the case: a spreadsheet is excellent at modelling — trying a pricing scenario, building a six-month cash forecast, analysing data you exported from somewhere else. It is excellent at one-off work that will not repeat. And it will stay open on your screen long after your books have moved elsewhere, which is normal and healthy.
What it is not good at is being a record: holding what happened, when, and who did it, and preventing it from being changed afterwards. A cell is editable by nature — its greatest strength for modelling and its fatal flaw for bookkeeping.
Five signals you have outgrown it
- 1You were asked about a number and could not explain it. "Why were receivables 310,000 in March?" has no answer in a file that overwrites itself every month.
- 2More than one person edits the file. That is where multiple copies begin, and shortly after, nobody knows which one is current.
- 3You entered the tax net. The return asks for separate buckets — standard, zero-rated, exempt — from invoices written at one blended rate.
- 4You started buying or selling in a second currency. An exchange rate makes a figure correct the day it is typed and wrong the day it is read, and revaluation differences have nowhere to live in a spreadsheet.
- 5You started carrying stock. A spreadsheet knows quantities; it does not know the cost of what you specifically sold — so your margin becomes an estimate rather than a number.
One of those five is enough. It is not about size or invoice count — a business with twenty invoices a month, two currencies and stock has already outgrown the spreadsheet, while one with two hundred cash invoices in a single currency and no stock may not have yet.
What breaks quietly
The dangerous part is that none of these produces an error message. The file opens, the numbers appear, and everything looks fine:
| What breaks | How | When you find out |
|---|---|---|
| A formula range | A new row added outside the SUM range | When one total stops matching another, months later |
| Rounding | Floating-point decimals each rounded on their own | When the columns add up to a unit or two away from the total |
| Numbering | Two people typing the next number at the same time | When two customers quote the same invoice number |
| History | A value edited in the row of a month already closed and filed | When you are asked to explain a difference against a past return |
| Bank reconciliation | A transaction recorded in the sheet that never happened at the bank, or the reverse | When you try to tie the two balances at year end |
| The second side of the entry | An asset moving with no matching source | Usually never |
That last row is the heart of it. A spreadsheet has no concept of double-entry, so it cannot tell you that a number arrived from somewhere nothing left. Accounting software refuses an unbalanced entry before accepting it at all — which is not an extra feature but the entire reason it exists.
What moving over actually involves
The most common fear about migrating is that it means re-entering years of history. It does not. A normal migration is five steps, most of it a single day's work:
- 1Pick a start date — usually the first of a month or of the financial year. Everything before it stays in the spreadsheet as an archive and is not re-entered.
- 2Enter opening balances at that date: cash and bank, receivables itemised by uncollected invoice, payables, inventory at value, fixed assets with their accumulated depreciation, and capital and retained earnings.
- 3Import your contacts from the spreadsheet. This is the one step where the old file genuinely helps, and it is one CSV.
- 4Issue every new invoice from the system from the start date onward. Do not run both in parallel for more than one cycle — running in parallel is longer and riskier than switching outright.
- 5Reconcile the bank balance at the end of the first month. If it ties, the migration worked; if it does not, the difference surfaces now while it is small rather than in a year when it is not.
Step two is the only one that deserves real time and attention, and it is also the moment double-entry asserts itself immediately: your opening assets must equal liabilities plus equity. If they do not, the difference has been in your old books for a while and there was nothing there to reveal it.
What you keep from the spreadsheet way of working
Migrating is not supposed to take away what you liked. You keep the ability to export any report to CSV and open it in your usual spreadsheet to slice it however you want — the one rule being that this is a copy for analysis, not a copy to be edited and typed back in. And CSV stays useful in the other direction where it helps most: your whole contact list imports in one go, and every bank statement arrives as a file to be auto-matched rather than lines to be re-typed.
What you gain in exchange is what no amount of discipline can give a spreadsheet: an entry that is not accepted unless it balances, invoice numbers that cannot repeat, an audit trail saying who changed what and when, a period lock that blocks edits after filing, and reports all derived from the same source so there is no second version of the truth. In Fatouraty every one of those is enforced in code rather than recommended in a manual.
Frequently asked questions
Can you keep books properly in a spreadsheet?
You can keep a correct cash log for a simple, single-currency business with no stock. What you cannot do is enforce balancing, prevent retroactive edits, or preserve a history of who changed what — which are the things that make a ledger auditable.
When exactly should I switch?
At the first of the five signals: a question about an old number you cannot explain, a second editor on the file, entering the tax net, a second currency, or stock. The best practical timing is the start of a financial year or quarter, because the opening balances are being prepared anyway.
Do I have to re-enter years of data?
No. You enter opening balances at the start date only, and everything before it stays as an archive in the old file. The one exception is itemising uncollected invoices and outstanding payables, because you will need to match payments against them later.
What if my opening balances don't balance?
This is very common, and it means a difference has been sitting in your old books with nothing to reveal it. The practical approach is to find the cause first — a forgotten asset, an unrecorded loan, a prior year's profit never carried forward — rather than pushing the difference into a suspense account.
Should I run both systems in parallel for a while?
One cycle at most. Long parallel running doubles the work and creates two sources of truth, at which point "which one is right?" has no answer. Reconciling the bank at the end of the first month gives you the same assurance for a fraction of the effort.
Can I keep using Excel after switching?
Yes, and that is its correct use: export any report to CSV and analyse it however you like. The one rule is that the copy is never edited and typed back in — the export is for reading, and the ledger stays the single source of truth.