Budgeting

A budget you don't have to build.

Most budgeting apps hand you a blank grid and a moral obligation. feels.money asks two questions — what comes in, what roughly goes out — and computes a complete monthly plan from them: twelve category limits, a savings target, and guardrails that stop the plan quietly eating your savings.

This page explains exactly what it computes and how, the rules it refuses to break, and the one number that measures whether the plan survived the month.

  • 12 spending categories
  • 50/30/20 with guardrails
  • Explainable adherence score

the point of the thing

A budget is a decision you make once, so you don't have to make it forty times

The reason budgets are worth the trouble is not discipline. It is that deciding once — at the start of the month, calmly, with the numbers in front of you — is dramatically easier than deciding forty separate times in the checkout queue with a coffee going cold in your hand. A budget converts a long series of small, tiring, guilt-adjacent decisions into a single planning decision you can revisit whenever you like.

That only works if the plan is plausible. A limit set at half of what you have spent on groceries every month for a year is not ambitious, it is fiction, and the first week that it breaks is the week you stop opening the app. So the goal here is not to produce the most virtuous budget imaginable. It is to produce a defensible one: limits derived from real arithmetic on your real income, with enough room left over that saving happens automatically rather than heroically.

Everything below follows from that. The proportions come from a framework with a long track record. The guardrails exist to stop the plan from swallowing your savings. And the score that measures adherence is proportional, not binary — going $40 over on groceries is a rounding error, not a failure.

the framework

The 50/30/20 baseline, computed across twelve categories

Fifty percent of income to needs, thirty to wants, the rest left for saving and paying down debt — turned into specific dollar limits rather than a slogan.

feels.money splits the classic framework across twelve categories. Five are treated as needs — Housing, Utilities, Groceries, Transport, and Health & Fitness. Seven are treated as wants — Food & Drink, Shopping, Entertainment, Travel, Subscriptions, Fees, and Other. Income and Transfer are deliberately left out of the budget entirely: money moving between your own accounts is not spending, and treating it as spending is how a budget ends up telling you that you spent your whole salary twice.

The shares are specific rather than hand-waved. Housing takes the largest single slice at 28% of income, groceries 9%, food and drink 8%, shopping 7%, transport 6%, utilities and entertainment and travel 4% each, health and subscriptions and the catch-all 3% each, and fees 1%. Add the needs and you get exactly half your income; add the wants and you get exactly thirty percent. Limits are rounded to whole dollars, because budgeting to the cent is a way of looking busy.

Example

A $5,200/month plan, as the engine computes it

Monthly income $5,200

  • Needs$2,60050%

    Housing · Utilities · Groceries · Transport · Health & Fitness

  • Wants$1,56030%

    Food & Drink · Shopping · Entertainment · Travel · Subscriptions · Fees · Other

  • Left for saving and debt$1,04020%

    Never squeezed below 10% of income — more when your savings target is higher

An illustrative income run through the same suggestBudget() function the app calls when it generates your plan. The percentages and guardrails are real; the household is invented.

Three guardrails sit on top of those shares, and they are hard guarantees of the calculation rather than suggestions:

  • Housing is capped at 35% of income. No set of inputs can produce a proposed housing limit above that ceiling. If your actual rent is higher, the plan will not pretend the other categories can absorb it — you will see the pressure instead of having it quietly spread everywhere else.
  • A savings floor is reserved before anything else. At least ten percent of income is held back, or more when your savings target is higher, and the limits are fitted into what remains. Your target itself is computed from your own answers: the smaller of a fifth of your income and the gap between what you earn and what you spend.
  • Wants are squeezed before needs. When the limits do not fit inside the envelope, discretionary categories are reduced first and proportionally, and housing keeps its capped value for as long as possible.
Example

The twelve limits, and what each one costs you

Example monthly category limits at $5,200.00 of monthly income
CategoryLimitShare
Housing$1,45628%
Utilities$2084%
Groceries$4689%
Transport$3126%
Health & Fitness$1563%
Food & Drink$4168%
Shopping$3647%
Entertainment$2084%
Travel$2084%
Subscriptions$1563%
Fees$521%
Other$1563%
Left for saving and debt$1,04020%
The same illustrative income, broken out per category with its share. Every figure is produced by the engine; nothing here is typed by hand.

during the month

Category tracking that reads like a sentence, not a spreadsheet

A plan is only useful if you can see where you stand without doing arithmetic. Every transaction that syncs is placed into one of the fourteen platform categories automatically, and the spending view rolls those up into the twelve budgeted ones. Each category shows what you have spent, how it compares with last month, and — when you tap it — the individual transactions behind the number. Where a limit exists for that month, it is drawn as a marker on the bar, with the amount you are over or under stated in words underneath.

Transfers and anything you have marked as excluded never count. Pending transactions are visible but flagged. If a charge landed in the wrong bucket, you can recategorize it in one tap and the whole month recalculates; your override is stored alongside the provider's original value, so the next sync cannot silently undo your correction. The mechanics of that are on the expense tracker page.

When a category runs past 110% of its limit, a single notification is raised naming the category, the amount, and the percentage — once per category per month, not a running commentary. Alongside that, a set of deterministic detectors watches for patterns worth naming: lifestyle inflation as your baseline creeps upward, clusters of impulse purchases, a dining trend moving in one direction, your largest purchases, and how this month compares with the last. These are descriptions, not verdicts.

the score

Budget adherence: proportional, explainable, and only 7.5% of the story

One number for how closely the month tracked the plan — and a deliberate decision not to let it dominate.

Adherence is calculated in the least dramatic way possible. Add up the amount by which every category exceeded its limit. Divide that total by the sum of all your limits. Subtract from one. Categories you came in under do not earn credit that offsets an overspend elsewhere, because a $200 underspend on travel does not undo a $200 overspend on rent — and categories with no limit set are skipped entirely.

The result is a figure between 0 and 1 that behaves the way you would expect: small overruns move it slightly, and blowing through several categories moves it a lot. With no limits at all the calculation has nothing to violate and returns 1 — and in your health score, a month with no budget on file is scored at a neutral 0.6 instead, so you are neither rewarded nor punished for not having started.

Example

One month, scored against those limits

Over budget

$271

Across 3 categories, on $4,160 of limits

Adherence

93%

1 − overspend ÷ total budget

Health score points

7.0 / 7.5

Adherence carries 7.5% of the 0–100 score

  • Groceries$494of $468
  • Food & Drink$604of $416
  • Shopping$421of $364
Illustrative spending scored by the same budgetAdherence() function the health score uses. The weight shown is the real weight adherence carries in the 0–100 score.

That figure then becomes one of eight weighted subscores in your financial health score, and it is deliberately not the biggest one. Savings rate carries 20%, cash flow, debt load and emergency fund carry 15% each, investing and income stability 10% each, and adherence and spending habits 7.5% each. The reasoning is simple: staying inside self-imposed limits is a useful habit, but having six months of expenses in the bank matters more, and we would rather the score reflect that than flatter your discipline.

Every subscore comes with a plain-English explanation of what produced it — the actual percentages, the actual dollar amounts — so the number is never a black box you have to trust. The coach reads the same subscores, which is why its answers and your dashboard never disagree.

how it behaves

Four things this budget will and won't do

It is arithmetic, not a questionnaire

Two answers — income and typical spending — produce twelve limits, a savings target and a set of guardrails. The same function runs every time, so the plan is reproducible rather than a matter of which day you filled the form in.

The month is read live

Category totals, month-over-month deltas and the position of each limit are recomputed on every sync, so where you stand is a page you look at rather than a spreadsheet you maintain.

Your corrections outrank the machine

Any transaction can be recategorized, excluded from analytics, flagged as an impulse or annotated. Those choices are stored separately from the provider's data, so a later sync never overwrites them.

It cannot move your money

The budget is a plan, not an instruction. Bank connections are read-only, so nothing here can transfer, pay, or freeze anything on your behalf.

The split
50% of income across five needs categories, 30% across seven wants categories, and the remainder left for saving and debt.
Housing cap
No proposed housing limit ever exceeds 35% of monthly income, whatever the other inputs say.
Savings floor
Limits never consume more than 90% of income — and less than that whenever your savings target is higher than ten percent.
Squeeze order
When the limits do not fit inside the envelope, discretionary categories are reduced first and proportionally; needs are touched last.

One more boundary worth stating plainly: this is a planning and education tool, not financial advice. feels.money is not a registered investment adviser, broker, or fiduciary. It computes your numbers, explains the arithmetic, and lays out options — the decisions, and the consequences, remain yours. For anything with real stakes, talk to a professional who knows your whole situation.

questions

Before you start

Your first plan takes about four minutes.

Answer the onboarding questions and the twelve limits are computed on the spot. Connect an account and the spending side fills in behind them. Free forever, no card, and you can delete everything in two clicks if it isn't for you.