Kakeibo in Google Sheets

Kakeibo does not need paper and it does not need an app. It needs a place you will actually write in, sorted into four categories, with the savings promise taken off the top before you look at anything else. A spreadsheet satisfies all of that, and it does one thing neither paper nor a phone does particularly well: it adds up for you and remembers every month you have ever run.

This is the sheet, built from an empty tab rather than described from a finished one, so every formula in it is one you typed yourself and can therefore fix.

The one sheet you need

Use two tabs, not twelve. The first is the plan, one section per month with income, the savings promise and the four envelope targets. The second is the ledger: one continuous list of every entry you make, for every month, in one place. Splitting the ledger by month into separate tabs is the single most common mistake people make setting this up, because it means a formula written in January has to be rewritten in February.

The five columns

Across the top of the ledger tab, five headers and nothing else.

• Date. The day the decision happened, not the day a card statement settles it.
• Description. Two or three words. This is a ledger, not a diary.
• Amount. One number, always positive. Income lives on the plan tab, not mixed into this list.
• Envelope. Survival, Optional, Culture or Extra, and nothing else typed by hand here. Select the column, open Data, then Data validation, and set the criteria to a list of those four items. A dropdown means every entry sorts correctly on the first try, and it means a formula that reads this column later can trust what it finds.
• Month. A short label like 2026-08. Typing this by hand for a year of entries is tedious enough that most people set it with a formula off the date column instead, something like TEXT(date cell, "YYYY-MM"), and then never touch it again.

Five columns, one row per entry, and the row is the whole unit of work. Nothing about the layout should ever require you to scroll right.

The three formulas that matter

Everything else on the plan tab is arithmetic, and it is three formulas, not thirty.

1. The savings line, off the top

This one is not a formula so much as a decision typed into a cell. Put your income for the month in one cell and the amount you are saving in the cell directly beneath it, decided before you look at a single expense. The rest of the plan tab subtracts from that second cell, not from income, which is the entire structural point of the method: saving is a promise made first, not whatever is left over.

2. The four category totals

On the plan tab, one row per envelope, each with a formula that reads the ledger tab:

=SUMIF(Ledger!D:D, "Survival", Ledger!C:C)

Column D is Envelope, column C is Amount, adjust the letters to match your own sheet. Repeat the same formula three more times with Optional, Culture and Extra in place of Survival, and the whole month's spending sorts itself the moment you enter it, with no pivot table and nothing to refresh by hand.

3. What is left to spend

One more cell, income minus the savings promise minus the running total of all four envelopes so far this month:

=Income - Savings - SUM(Category totals)

This is the number worth checking most often, because it is the one that tells you whether the month is still on plan or already spent past it. If you want the same figure worked out to a daily allowance rather than a month total, the safe to spend calculator does that division for you against the days actually left.

Reading the month back

Once the four totals exist as formulas, the four questions from the method are almost free to answer. How much came in and how much did you decide to save are the two cells at the top of the plan tab, typed before the month began. How much did you spend is the four SUMIF results, read together. How you can improve is the one question a spreadsheet cannot answer for you, and it is worth writing that answer in a real sentence, in a sixth column or a cell at the bottom of the tab, rather than skipping it because the numbers already looked tidy.

To see whether the shape of a month is the one you intended, the four category split calculator takes the same numbers you have just totalled and returns each envelope as a share of income, which reads differently from an absolute total and is usually the more useful comparison against last month.

Comparing months without fooling yourself

A single month's totals do not tell you much on their own, because income moves and a bigger number does not automatically mean a worse month. Add one more column to the plan tab, dividing what you actually saved by income for that month, and you have a rate rather than an amount, which is the figure that survives a pay rise or a quiet month. The savings rate calculator works through the same division and the definitions worth arguing about, gross against net income chief among them, before you commit to a version in your own sheet.

Add a new block of the same rows every month rather than resetting the plan tab, and a year in you have twelve rates sitting in a column, which is a chart away from showing you a trend a single month never could.

What a spreadsheet is genuinely better at

Two things, and they are the two that make the method worth continuing rather than the two that make it easy to start. The arithmetic you actually want checked, the four totals, the savings rate, the running comparison of plan against actual, is exact every time and never depends on you doing mental subtraction at the end of a long month. And the reflection has somewhere permanent to live: a year of entries in one file means the January question and the December question are answered from the same evidence, which a notebook you replace every twelve months cannot offer without you keeping every old one on a shelf.

What it is genuinely worse at

The counter. A spreadsheet lives on a laptop far more often than it lives in a pocket, and the method depends on the entry happening close to the moment of the decision, not on the bus home or on Sunday from memory. Waiting until you are back at the laptop to log five days of card spending is not a spreadsheet problem specifically, it is the same reconstruction failure paper suffers from when the notebook is left at home, and it is dealt with properly in kakeibo with a credit card and digital payments. If you check the sheet on your phone rather than a laptop, most of that gap closes, but the entry is still two taps slower than opening an app built for exactly this one job.

The honest fix most people land on is not choosing one medium forever. Log at the moment on whatever is already in your hand, and let the sheet be where the month gets read back rather than where every entry originates. That is also roughly what the paper version of this looks like, laid out page by page in the templates guide, which is worth reading if the idea of a bound notebook appeals more than a browser tab.

Keeping it honest for a full year

The sheet gets unwieldy exactly once, around month nine or ten, when the ledger tab is long enough that scrolling to this month's entries stops being instant. Freeze the header row so the column labels stay visible, and sort the ledger by date if it ever falls out of order, rather than starting a new tab. A single continuous ledger is what makes a year end reflection possible at all, and splitting it defeats the entire advantage a spreadsheet has over paper.