Most budgets are written on a Sunday. You sit down with a blank spreadsheet, you feel a certain resolve, and you start filling in numbers. Groceries: a tidy round figure. Eating out: much lower than last month. Transport: optimistic. Subscriptions: the ones you remember. By the time you finish, the columns balance and there is even something left over, which feels like an achievement.

The problem is that the budget was already wrong on the Sunday. Not wrong by February, when you abandoned it — wrong at the moment of writing, because almost none of the numbers came from anywhere. They came from what you hoped was true. That is why it lasted six weeks. It was never a plan; it was a wish written in a format that looked like a plan.

A budget that survives is built the other way round. You do not start by deciding what you should spend. You start by finding out what you already spend, in detail, without flinching, and only then decide what to change. That single inversion is the difference between a budget that runs for years and one that dies quietly in a drawer.

Why do most budgets stop working within two months?

There is a common story that budgets fail because people lack discipline. It is a satisfying story if you have never struggled, and it is almost always wrong. The usual failure has three parts, none about willpower.

The numbers were invented. You wrote the grocery figure you would like to spend rather than the one your statements show. When reality arrives in week two and you are already over, the budget starts reading as a scoreboard on which you are losing. People do not keep using tools that tell them they are failing.

The irregular things were left out. A budget built only around the bills that arrive every month is a budget that works only in months where nothing else happens. There are very few such months. The car needs something, a birthday arrives, the dentist happens. None of these are extravagances. They are ordinary life, and a budget with no room for them will be broken by ordinary life.

There was no slack and no enjoyment. A budget in which every unit of money has a serious, responsible job is a budget you will resent by week three. Austerity is not a plan; it is a mood, and moods pass.

Notice what is missing from that list: overspending. Overspending is the symptom. The disease is that the budget described a person who does not exist — someone with your income, no unpredictable costs, and no interest in anything fun.

A budget is not a promise about the future. It is a description of the present that you then adjust deliberately. Get the description wrong and everything downstream is guesswork.

What does the honest look backwards actually involve?

This is step one, and almost nobody does it, which is why it is worth doing. Before you allocate a single unit of currency, you are going to find out what your money currently does.

Go back three months. One month is not enough — it might have been unusual, and you will not know in which direction. Three months gives you an average that means something, and usually spans at least one payment that only shows up occasionally. Six is better still, but three is where the exercise becomes useful.

Gather every source. Current account statements, every card, any digital wallet, any account you use for a specific purpose, and cash withdrawals. Statements can usually be downloaded as a file you can open in a spreadsheet, which saves a great deal of typing. Cash withdrawals show as a single lump and you will have to estimate what they became — note that as an estimate rather than pretending it is precise.

Go line by line, and label everything. Not from memory. Line by line. For each transaction, write one word: what it actually was. The goal is a complete list, not a flattering one.

Look for the things you had forgotten. This is where the exercise earns its keep. Almost everyone finds at least one of the following: a subscription still charging for something they stopped using, a service that quietly increased its price, a recurring charge they cannot immediately identify, a category that is two or three times larger than they would have guessed, or a duplicate — two things doing the same job. Recurring charges are designed to be unnoticeable; that is the whole business model.

Expect it to be uncomfortable. There is often a moment in this process where you see a total and feel something drop. That feeling is information, not a verdict. The number was already true before you looked at it. The only thing that changed is that you know it, which is the only condition under which you can do anything about it. The version of you who spent that money was solving a real problem at the time — being tired, being busy, needing something to be easy. The spending had reasons.

At the end of this, you want two things: a total for each category over three months, and that total divided by three to get a monthly average. Those averages are your real starting numbers. Everything you build sits on top of them.

What is the difference between fixed, variable and irregular costs?

Confusing these three is one of the most reliable ways to break a budget, so it is worth being precise. Most budgeting advice collapses them into two, and the missing third is exactly the one that causes the damage.

Fixed costs are the same amount at the same interval, and you have little short-term control over them. Rent or mortgage payments, insurance premiums, loan repayments, subscription fees. You can change these, but only by making a large decision — moving, switching provider, cancelling something. Not by trying harder this week.

Variable costs arrive every month but the amount moves. Food, transport, energy billed on usage, household bits, going out, personal spending. This is where your day-to-day choices live, and it is where a budget can genuinely influence things.

Irregular costs are real, predictable in the sense that they will happen, but they do not arrive monthly. An annual insurance renewal, a vehicle service, professional registrations, gifts clustered around certain months, course costs.

Here is why the distinction matters. Treat a variable cost as fixed and it drifts upward without you noticing. Treat a fixed cost as variable and you spend energy shaving at something that will not move. And if you leave irregular costs out entirely — the most common error — your budget will look perfectly balanced in most months and then be destroyed by the one where a large annual bill lands. You will experience that as failure. It was not failure. It was a missing category.

How do you turn annual costs into a monthly number?

An annual cost does not become manageable by being ignored for eleven months. It becomes manageable by being converted into a monthly figure and treated exactly like a bill.

The arithmetic is simple: take the annual amount and divide by twelve. An illustrative example — a cost of 600 per year becomes 50 per month. Do this for every irregular expense you found in your three-month review, plus any you know about that did not fall inside the window, and add the monthly figures together. That combined number goes into your budget as a single line, every month, alongside rent and utilities. It is not spare money. It is money that already belongs to a bill that has not arrived yet.

Two cautions. If an amount varies year to year, use the higher recent figure. And if a large renewal is only three months away, divide by three rather than twelve for that item, so the money is there in time.

Where you keep that money, how to catch up if you are starting late, and how to structure it are a topic in their own right — here is how to build a sinking fund.

Which budgeting method should you actually use?

There are four approaches in wide use. They are usually presented as competing philosophies, and one of them is usually being sold to you. They are better understood as tools with different failure modes. None is correct in the abstract; the right one matches your income pattern and your tolerance for admin.

Method How it works Suits Fails when
Zero-based Every unit of income is assigned a job until nothing is unallocated. Income minus all allocations equals zero. People who want full visibility, who have variable or irregular income, or who have found that vaguer methods leave money unaccounted for. You do not want the admin. It needs a session at the start of each month and a check-in or two; done half-heartedly it produces a stale plan.
50/30/20 Split take-home pay into rough thirds: 50% needs, 30% wants, 20% saving and debt repayment beyond minimums. Someone who wants a quick orientation, has moderate housing costs relative to income, and finds detailed tracking off-putting. Housing is expensive relative to income — which is very common. If rent alone is 45%, needs cannot fit in 50%, and the framework becomes a way of being told you are doing it wrong. See below.
Pay yourself first Move a set amount to savings immediately on payday, then live on what remains without tracking it closely. People with steady income who dislike detail and mainly want to make sure saving actually happens rather than being whatever is left over. What remains genuinely is not enough, so the money comes back out later — sometimes at a cost. Also hides where the rest goes, so it cannot help you diagnose a problem.
Envelopes or pots Money for each category is physically or digitally separated. When a category is empty, that category is done for the period. People who overspend in specific identifiable categories, and anyone who finds a single large balance impossible to reason about. Costs are lumpy or timing is unpredictable, so you constantly move money between pots and the boundaries stop meaning anything. Cash envelopes also clash with card-based and online spending.

A specific word about 50/30/20

This rule is repeated more than any other, and it does real damage applied to the wrong situation. It assumes essential costs fit inside half your take-home pay. For a large number of people, they cannot. If housing takes 40% or 50% on its own, then before you have eaten, travelled, insured anything or paid a utility bill, the “needs” bucket is already spent. Add the rest and needs might be 70% or 80% of income. That is not a personal failing or a miscategorisation. It is a description of what housing costs relative to what many people earn.

Being told in that situation that needs “should” be 50% is not guidance. It is a reason to conclude that budgeting is for other people. If your needs come out well above 50%, the honest response is: fine, that is your starting position. Use the structure — needs, enjoyment, future — but set your own percentages from your actual numbers, and treat any improvement in the ratio as progress rather than treating the gap from 50 as failure.

How much should you allow for things you enjoy?

Whatever number lets you keep going. That is a real answer, not an evasion. A budget with no enjoyment line is not a strict budget; it is an unfinished one. You will spend money on things you enjoy — everyone does — and if the budget does not contain a line for it, that spending happens anyway and gets recorded as a failure. Do that twice and you stop opening the spreadsheet.

Some practical ways to handle it:

  • Start from your real average, not from zero. Your three-month review already told you what you spend on this. Set the budget line near that figure to begin with. Cut it later if you decide to — but not in month one, and not to nothing.
  • Make it a single flexible line rather than five specific ones. A single “personal spending” line works better than one per hobby, because your interests move around and the specific lines will be wrong.
  • Add a genuine buffer as well. Separately, leave an unallocated amount for small unforeseen things — a replacement charger, a prescription. A budget with zero slack is broken by the first surprise, and there is always a surprise.

If your numbers are tight enough that both of these feel impossible, keep them anyway and make them small. A small enjoyment line that survives is worth far more than a large one that gets cancelled in a fit of resolve.

How do you budget when your income is not the same every month?

Nearly every budgeting guide assumes a fixed salary on a fixed date. If you are freelance, on shifts, on commission, on tips, on seasonal work, or juggling several sources, that assumption makes most advice useless to you. Your situation is not harder to budget — it needs a different sequence.

Find your floor, not your average. Go back twelve months if you can, or as far as your records go. Identify the lowest month, not the typical one. That figure — or slightly below it — is your baseline. You build your core budget on the floor, because a budget built on the average is a budget that fails in every below-average month, and there will be many.

Separate essentials from everything else, sharply. Work out the absolute minimum you need in a month: housing, utilities, food, transport to work, insurance, minimum obligations, the monthly slice of your irregular costs. Everything above that total is variable.

Use a buffer account as a shock absorber. The mechanism that makes irregular income manageable is putting a month between earning and spending. In a good month, the surplus does not get spent — it goes into a holding account. In a poor month, you draw from that account to top your income up to your baseline. The aim is that your spending life becomes steady even though your earning life is not. Building it takes time, and the first good month or two will feel like you are not allowed to enjoy them. That is the price of the stability that follows.

Pay yourself a fixed amount. Once the buffer exists, take the same amount out each month regardless of what you earned, and let the buffer absorb the difference.

Set aside tax and deductions immediately, if they apply to you. If you are self-employed or paid without deductions taken off, a percentage of every payment is not yours. Move it out on the day it arrives, into an account you do not touch. Treating gross income as spendable is the most common way irregular-income budgets collapse, and it collapses them with a deadline attached. What percentage to hold back depends on where you live — check your own tax authority rather than a rule of thumb from an article.

Budget by period, not by calendar month, if that fits better. If you are paid weekly or per job, monthly boundaries are artificial. Some people do better running a four-week cycle, or budgeting per payment received.

What should you do in the first two months?

Collect data. That is the job. Not succeed — collect data.

Even after a careful review, your first budget will be wrong in places. You will have missed a category. One number will be too low because the review period was quiet. Something unanticipated will arrive. This is expected, and it does not mean the method is failing.

So run the first two cycles as measurement. Record what actually happened against what you planned, then at the end of each month spend twenty minutes on three questions:

  1. Which lines were wrong, and in which direction? Adjust them to reality rather than trying harder next month. If groceries ran over three months running, the budget figure is wrong, not your behaviour.
  2. What arrived that has no category? Give it one.
  3. Which lines do I want to change deliberately? Now — and only now — is the moment for aspiration. Pick one category, and move it by a modest amount.

By month three or four you have something built from evidence rather than hope, and it starts holding. People who say budgeting does not work for them usually quit in month two, at exactly the point where the process was doing its job.

After that the maintenance is light: a short session on payday to allocate, a mid-period glance for anything running hot, and a proper review once a quarter to catch price rises and subscriptions that crept back in.

What if the numbers do not balance at all?

Some readers will finish the honest look backwards and find that income genuinely does not cover outgoings. Not “if I were more careful” — genuinely does not, on essentials alone.

If that is you, the first thing worth saying is that a budget will not fix it, and no amount of category discipline will either. You cannot arrange a shortfall into a surplus. Being told to track more carefully when the gap is structural implies the problem is your habits when it is arithmetic.

The second thing worth saying is that this is a common position, and it is far more often caused by circumstance than by choices — housing costs rising faster than pay, reduced hours, a health problem, a relationship ending, caring responsibilities, a contract that stopped. None of that is a character question.

What the exercise has given you is still valuable: a documented picture of income against outgoings. That is precisely what you need to have a useful conversation with anyone who might be able to help — a landlord or lender about changing payment terms, a utility provider about a hardship arrangement, an employer about hours, or a benefits or entitlements check to see whether there is support you are not claiming.

It is also worth knowing that free, independent, non-commercial debt advice services exist in most countries. They are typically charities or publicly funded organisations, they do not sell products, and they can look at a full picture and explain options in a way a general article cannot. Search for the ones operating where you live, and check that the service really is free rather than a commercial firm charging for something available free elsewhere.

Spreadsheet, paper, app or bank tools — which should you use?

The tool matters far less than the review that precedes it. Any of these work. What follows is what each is good and bad at — and free versions of all four approaches exist.

Spreadsheet. Total flexibility, no cost, no third party, and your data stays with you. You can build exactly the categories you need, including the irregular-costs line most tools handle poorly. The cost is setup effort and manual entry, and if you dislike spreadsheets you will not maintain it. Best for unusual circumstances — irregular income, a business mixed with personal — because it is the only option that will bend to fit.

Paper. A notebook and a pen. It works well for some people, because writing something down by hand makes it register in a way that tapping a screen does not. The downsides are obvious: no automatic totals, easy to lose, and reviewing three months means flicking through pages. Good as a companion to a spreadsheet rather than a replacement for one.

Dedicated apps. These automate the tedious parts: importing transactions, categorising them, showing you where you are mid-month. That automation is the main reason some people stick with budgeting who otherwise would not. The trade-offs: many charge a subscription, which is itself a budget line; categorisation is frequently wrong and needs correcting; they usually assume a regular monthly salary, so irregular income fits awkwardly; and you are handing financial data to a third party.

Bank-provided tools. Built into many banking apps, free, and they already have your data. Convenient, zero setup. The limitations: they generally see only accounts held with that institution, so if your money is spread around the picture is partial; categories are usually fixed; and they are good at showing what happened but weak at planning what should happen.

Start with whichever one you will actually open. If you are unsure, a spreadsheet for the first three months is worth it purely because building it forces you to understand your own numbers. You can always automate later, once you know what you are automating.

A budget is only the accounting. What you do with the gap it reveals is a separate question with a clear order of priorities.

A budget is only as good as the income figure at the top of it, and that figure is easy to misread — here is what to check on a payslip.

If the budget will not balance however it is arranged, Citizens Advice offers free help with exactly that situation.

If the shortfall is structural, it is worth checking benefits and entitlements — a great deal goes unclaimed simply because nobody looks.

Frequently asked questions

How long does it take to set up a budget properly?

The three-month review typically takes two to four hours if your statements download cleanly, longer if you are working from paper or have many accounts. Building the budget itself from those numbers is under an hour. Split it across two or three sessions — the review is tedious, and people who attempt it in one sitting often stop halfway.

Should I budget by calendar month or by pay period?

By whichever matches how money arrives. If you are paid monthly, calendar months are simplest. If you are paid weekly, fortnightly or irregularly, a cycle starting on payday works better, because it avoids a month boundary falling in an awkward place with half your income in the wrong column.

Do I need to track every single transaction?

For the initial three-month review, yes — that is where the value is. After that, no. Once your categories reflect reality, checking category totals is enough, and you only drill down when one behaves unexpectedly.

What if I share finances with someone else?

Do the review separately first, then compare. Doing it jointly from the start turns into a negotiation before either of you has seen the numbers. Once you both have real figures, decide which costs are shared, which are individual, and give each person an unquestioned personal spending line. Budgets between two people fail most often over the absence of that line rather than over the big shared costs.

My income changed. Do I have to start the whole process again?

No. Your category structure and irregular-costs line stay valid — those describe your life, not your income. Rework the essentials against the new income figure, adjust the flexible lines, and carry on. Only redo the full review if your circumstances changed substantially, such as moving home.

Is it worth budgeting if my income is very low?

The mechanics are the same and the review is arguably more valuable, because when there is little slack, knowing exactly where money goes matters more than when there is plenty. But be clear-eyed about what a budget can do: it improves allocation, it does not create income. If the review shows the shortfall is structural, the useful next step is the free advice route described above, not a stricter spreadsheet.