Reconciliation: why taking over a half-finished set of books is better redone than continued

Hook

Boss, here is a situation a lot of people do not understand. You engage a new accountant to take over, the books are half done, and what you have in mind is simple: just carry on from there. Then they tell you, "I would like to redo the past six months." And you think: is that not a waste of money? It has already been done. Today we settle it: why, when taking over old books, redoing them is often cheaper than continuing them.

First, what reconciliation actually reconciles

Start with the basics. Reconciliation is not "checking whether the numbers look right". It is taking two independent records and letting them verify each other. Three common kinds. Bank reconciliation — the bank balance in your books against the balance on the bank statement. Where is the difference? Cheques not yet presented, transfers that only landed at month end, bank charges. Supplier reconciliation — what your books say you owe against what their statement says you owe. Where is the difference? An invoice they issued that you have not received, goods you returned for which they have not raised a credit note, a payment you made that they applied to the wrong invoice. Customer reconciliation — the same set in reverse, and the ageing from the sixth lesson. The point: the value of a reconciliation is not that it agrees. It is that when it does not agree, you know which item the difference is. Not agreeing and not being able to say why is not reconciling at all.

Where the trouble is in a mid-year handover

Now the handover. Three situations come up most. One, you did half of it yourself — you or your wife keyed the first few months and the rest goes to an accountant. Two, the colleague who kept the books left; they got to June and then they were gone. Three, the accounting system changed: one in the first half of the year, another in the second. All three share the same trouble: you do not know what has already been done. Open it and there are numbers. But start asking. This receipt — is it a sale, or the boss putting money through (the company-versus-personal line from the third lesson)? This invoice — has it been posted, or only scanned and filed? This "other payables" balance — what is actually inside it? Nobody can answer, because whoever did it worked from memory, and the memory left while the books stayed. So the real trouble is not that the numbers are messy. It is that the books come with no manual — nobody knows how each balance grew.

Why redoing is cheaper

So why would the person taking over rather redo it? Work it out and it is obvious. Continuing means archaeology first: every balance has to be traced back to its source and confirmed for what it is, whether it is right, and whether it is duplicated. One balance can involve a dozen entries, and one entry you cannot trace jams the whole line. And even after tracing it you still cannot guarantee it, because you can only see the result, not the thinking at the time. Redoing? You already have the documents (the evidence pack from the thirty-first lesson), and you post the period once through, to one set of rules. A month's volume in experienced hands is quick. That is the difference: redoing is labour, continuing is a puzzle. Labour can be estimated; a puzzle cannot, and as the thirtieth lesson said, the older the puzzle the more it costs. And when the redo is done, what you hold is a full year on one consistent logic — comparable with last year, and presentable to a bank. Continuing leaves you with one logic in the first half, another in the second, and a permanent seam where they join.

The hurdle of changing systems

Now changing systems. Bosses often ask, "Can the data not just be moved across?" The balances can move; the understanding cannot. Every system has its own temperament: how accounts are grouped, how documents are attached, how stock is valued, whether a report can be clicked through to its source. As the twenty-second lesson said, one reason we use Bukku is that a report drills through to the document. So the most important thing when changing systems is not the data migration. It is deciding on a line: the old system is sealed, and from a given day everything is done anew. The line usually goes at the start of a financial year — clean, comparable and easy to explain at filing time. Really have to change mid-year? Then treat the line as a small year-end: count the stock, reconcile, strike the balances, and use those as the opening figures in the new system. Then keep the old system; do not delete it. Keeping it is not sentiment, it is evidence.

To close

In summary: reconciliation is two records verifying each other, and it only counts if you can say where the difference is. What is expensive about a mid-year handover is not the numbers but the missing manual. So redoing is often cheaper than continuing — labour can be quoted for, a puzzle cannot. And when changing systems, draw a line first and seal the old one rather than deleting it. If you are sitting on a set of books that is half done and that nobody can explain — that is not your fault; it is a structural problem with the handover. Grab a coffee first and talk about your business. For the accounting, come to LTT. I am LTT, helping SME bosses get their accounts straight. Follow us, and see you next time.