Zero-Based Budgeting Lasted Five Weeks and Taught Me One Permanent Thing

Zero-based budgeting asks you to assign every unit of income a specific job before the month begins, so that income minus assignments equals zero. It is internally coherent and a lot of people swear by it. I lasted five weeks, and the way it failed was specific enough to be worth writing down, because I do not think I failed at it through laziness.

The collapse happened in week four and the cause was a dentist. £180, unplanned, not in any category. The system’s answer to this is that you move money from another category, which is correct and which I did — I took it from food and from a category I had labelled “clothes.” And then the month stopped working, because I still needed to eat, so the food adjustment was fictional, and by the end of the month my actual spending bore no relationship to the plan and I had spent several evenings maintaining a document that had become fiction.

What I understand now is that the method has a hidden requirement: it needs the unexpected to be rare. When the unexpected arrives once a quarter, moving money between categories is a small correction. When it arrives most months, which it does for me, every month becomes a rebuild, and rebuilding a budget four times is how you stop having a budget.

The thing I kept is a single category that I now think should be in every budget regardless of method. I call it “things I have not thought of.” It gets a real allocation — for me about twelve percent of the month — it is not earmarked for anything, and crucially it is not savings and does not roll into savings. If it is unspent at the end of the month it goes to savings, but it is not called savings, because a category called savings gets defended and this one needs to be spendable without any feeling of having broken something.

Twelve percent came from looking backwards. I went through eighteen months of records and totalled everything that had been unplanned at the time — the dentist, a broken washing machine, an unexpected trip for a funeral, a laptop charger. It averaged about eleven percent of my spending, remarkably steadily. So the unexpected is not actually unpredictable in aggregate. It is only unpredictable in particulars, and a budget with no line for it is a budget that assumes a category of thing that happens every month will not happen this month.

The wider point I would make about budgeting methods, having tried four of them, is that they are mostly variations on the same two decisions — how granular your categories are and how strictly you hold them — and the method’s name matters much less than whether the granularity matches how much attention you actually have. Zero-based is high granularity and high strictness. That is a lot of attention, and if you have it, it is genuinely the most informative system I have used.

What I run now is four categories and a rule, which is low granularity and moderate strictness, and it survives months where I am not paying attention. The four are fixed costs, food, the unthought-of category, and everything else. That is not enough resolution to answer interesting questions about my spending, and I do a separate detailed tracking exercise once a year for that. Splitting the two — a low-effort system that runs continuously, and a high-effort audit that runs occasionally — is the arrangement I should have arrived at years earlier.

None of this is advice about what you should do with your money; it is what happened when I tried a popular method and what I changed as a result.