The first month of a budget almost always goes well. You write down what comes in, you write down what goes out, the numbers agree, and there is that feeling of having finally got a grip on things.
Then the second month arrives, and something turns up that was not on the list. The car insurance. The service. A trip to the dentist. Road tax. A present that could not not be given. It is not overspending and it is not a character flaw. It is a perfectly ordinary cost that simply does not happen every month, which is exactly why it was not there when you made the list.
And this is where the real damage starts: the budget is now wrong, and a wrong budget is one you stop opening. It does not get abandoned out of laziness. It gets abandoned because it stopped telling the truth.
Why is the second month the one that kills a budget?
Because the first one is built from memory, and memory is good at holding what repeats and terrible at holding what happens once a year.
If you sat down now and wrote out your spending, you would get almost everything that leaves every month. Housing, food, transport, subscriptions, the gym. Those repeat, and because they repeat they are close to hand.
What will not come to mind is the thing that happened seven months ago and will happen again in five. And the problem is not forgetting one of them. It is that added up across a year they usually come to more than any single monthly category. A budget that ignores them is not slightly optimistic. It is looking somewhere else.
What is missing from a budget built from memory?
A third drawer.
Almost everyone starts with two: what is fixed and what varies. Rent is fixed, groceries vary, and the split looks like enough. It is not, because some costs are entirely predictable in amount and unpredictable in date, and those fit in neither drawer.
The split that works has three:
- Commitments, with a known date and a known amount: rent or the mortgage, utilities, instalments, monthly insurance.
- Variable spending, uncertain in amount and happening continuously: food, transport, going out, whatever else.
- Occasional spending, reasonably known in amount and scattered across the year: annual insurance, road tax, services, health, school supplies, presents.
The third drawer is the one that saves the second month, and it is the one almost nobody writes down. In a table, you can see why:
| Drawer | Do you know the date? | Do you know the amount? | How it fails in a budget |
|---|---|---|---|
| Commitments | Yes | Yes | Counted as spent on the 1st, when it leaves on the 28th |
| Variable spending | No | No | Estimated from your best month, not your average one |
| Occasional spending | No | Yes | Not in the budget at all |
The column that matters is the last one. Three different failures, and not one of them is fixed by more discipline.
How do you plan for something that only turns up now and then?
You give it a monthly share, including in the months when nothing goes out.
Write down everything from the last year that does not happen every month. Add it up. Divide by twelve. That number becomes a line in your budget like any other. Treat it as a bill, because that is what it is.
The practical difference is this: in the months when none of those costs land, that share is not money left over. It is money waiting. In the month the insurance arrives, it has already been paid, spread across the months before it, and it does not have to come out of anywhere else.
It is the same idea as a fund for the unexpected, with one important difference: this is not for the unexpected. It is for the expected that simply is not monthly. Confusing the two leads to spending your emergency fund on car services, and then having no fund when a real emergency shows up.
Have I spent it, or have I only promised it?
They are two different things, and confusing them is what breaks the most budgets.
It is the 5th. There is money in the account, and the mortgage goes out on the 28th. How much can you spend before then?
The answer almost every spreadsheet gives is "whatever is in the account". For twenty-three days that money looks like yours, and it is not: it already has an owner and a date. Anyone deciding against that number is deciding against an inflated one, and the end of the month arrives as a surprise that was never surprising.
The right answer takes the dated bills out before it answers. That is three things, not two:
- What has already gone. Spent, done, and out of the account, so it needs no line at all.
- What is promised and has not gone yet. It is in the account, it has a date, it has an amount, and it is not yours to decide about.
- What is left. It is the only number that answers "can I?".
With the whole month worth 100, so the example does not depend on what you earn:
| Still the 5th | The wrong reading | The right reading |
|---|---|---|
| What is in the account | 100 | 100 |
| Bills already dated this month | not counted | minus 35 |
| I can spend | 100 | 65 |
A hundred against sixty-five, with exactly the same money in the account. That is not a rounding difference: it is the difference between a month that closes and a month that does not, and the decision that separates them gets made on the 5th, against one of those two numbers.
A budget that only watches the balance is right on the last day of the month and lying on all the others. And the others are when the decisions get made.
How much should you leave unassigned?
A share of it, not nothing.
There is a strong pull towards assigning every last unit to a category, because it feels more rigorous. In practice it is what makes a budget brittle: anything unexpected forces you to rework the plan, and reworking the plan three times is how you stop believing in it.
Leave a slice unassigned, a proportion of what comes in rather than a fixed amount, so it moves when your income does. Do not give it a name. It is not for anything. It is precisely for what was not planned, and its job is to absorb the shock instead of letting it reach the other categories.
If that slice is never touched, it is too small to matter and too big to waste, so cut it. If it is gone every month before the 20th, the slice is not the problem: your category estimates are.
How many categories do you need?
Fewer than you will want to create.
The natural urge is to split everything into many small categories, because each one looks like more information. What it gives you is more decisions: every purchase now needs a choice, some purchases fit two categories, and the sorting starts happening at random. A budget with twenty badly filled categories tells you less than one with six filled properly.
Start with five or six, the ones that actually weigh. If after a few months one of them is large and opaque, a "miscellaneous" eating a fifth of the month, split that one only. Categories should come from a question you have, not from an urge to tidy.
What do you do when a month goes badly?
You close it and open the next one.
A month where you blew a category is not a failed budget, it is a data point. It shows either a wrong estimate or an unusual month, and both have an answer. What has no answer is deleting everything and starting again, because the information you had goes with it.
And if the same category blows three months running, the problem has stopped being the month and has become the number. Raise it and lower another. A budget that never adjusts is not discipline, it is fiction.
How do you know it is working?
There is one test, and it is not "I saved more".
The test is: the number at the end of the month has stopped surprising you. If three days out you can say roughly what will be left, and you are right, the budget is working, even if the amount is not what you would like. And if it still surprises you, more discipline will not fix it; what is missing is one of the drawers.
Saving more comes after, and it comes almost by itself. It is far easier to decide to cut something when you can see the whole year than when you are guessing on the 20th of every month.
Where to start, today
Three things, in this order. None of them takes more than twenty minutes.
- Write down everything from the last year that does not happen every month. Insurance, road tax, services, health, school, presents. Add it up and divide by twelve. That is the third drawer, and it is the one that was missing.
- Separate what is promised from what is available. If you have bills with a date on them this month, take them out of the number you make decisions against. That alone changes the 20th.
- Pick five categories, not twenty. The ones that weigh. The others can appear when you have a question about them.
We keep a page of frequently asked questions about budgeting and everyday money, with a short answer to each one and the full piece one click away. If you want to see it, it is here: frequently asked questions.
This is educational writing. It is not financial advice or an investment recommendation; for advice on products, speak to a professional registered with the CMVM.