Tell it your best recent month, your worst recent month and a typical one, mark which months of the year are your strong ones, and choose a percentage. It returns an estimated annual range, a baseline monthly spending figure a bad month can actually support, and the same savings percentage translated into dollars for a good month, a typical month and a bad month — one rule, three amounts. The 12-month table underneath shows the rule running against your own strong-and-weak pattern so you can check it by hand. Save a percentage of what arrives, not a fixed amount, and size your fixed monthly commitments to the bad month rather than the average.
Why fixed-dollar saving fails people with variable income. A fixed target — “$600 a month” — is set from an average. In a strong month it is far too little and you spend the rest. In a weak month it is impossible, so you skip it. Two or three skips in and the rule feels broken, so it gets abandoned altogether, usually within a year. A percentage never asks for money that has not arrived. On a $1,900 February it takes a small amount and on an $8,000 June it takes a large one, and because it is never impossible, it does not get abandoned. That is the entire idea on this page.
Your months
One rule, three amounts
The baseline — what a bad month can actually carry
The last row is the one to sit with. That is the fixed monthly amount this percentage rule saves on average — and the figure beside it is what that same fixed amount would represent in a bad month. Once a fixed target crosses roughly a quarter of a bad month’s deposits, it stops being a savings plan and starts being a month you skip. The percentage never crosses that line, because it is defined as a share of whatever arrived.
Your year, month by month, with the rule applied
Strong months are modelled at your best month and every other month at your worst. That is deliberately the widest version of your year — real years contain middling months too — and it is the version that makes the mechanism visible. The cushion column is the running total of everything the rule set aside. The last column is the one that makes the argument: it measures each month against a life sized to your typical month rather than your worst one. Every negative figure in it is a month you would have to cover from the cushion, from a card, or by skipping the saving.
| Month | Strength | Deposits | Saved | Left to live on | vs a typical-month life | Cushion |
|---|
Sizing your fixed commitments to the bad month, not the average
The second half of the idea matters as much as the first. Rent, a car note, insurance, a phone bill and a gym membership do not care what kind of month you had. They arrive at the same size every time. Every one of them that you size against your average month is a bill you cannot pay in February without touching the cushion — and a cushion spent on ordinary bills is not a cushion when something actually happens.
So the discipline is: fixed commitments fit inside the baseline; everything discretionary lives in the space above it. A good month does not raise the baseline. It funds the cushion, then the retirement account, then the things you wanted. This is the opposite of how a variable-income year usually goes, and it is the single change that most reliably stops a strong year from being followed by a bad one.
If your income arrives on a 1099 rather than a W-2, there is a further slice to take off the top before any of this: the tax nobody withheld. Self- employment tax alone is 15.3% of net earnings before income tax is considered at all. Our take-home pay calculator covers the W-2 side, and the plan chooser shows the self-employment tax arithmetic in full.
Assumptions used — read these before trusting the number
- Every figure is money deposited, before tax. Nothing here withholds anything. If you are paid on a 1099, set aside your quarterly estimated tax before you apply this rule, not after — otherwise the cushion you build in the spring becomes the tax payment you make in the summer.
- The annual range is built from your three months, not from history. The low end is your strong months at the typical figure plus every other month at the worst; the high end is your strong months at the best figure plus every other month at the typical. The modelled year in the table sits between the two by construction.
- The baseline is the worst month less the savings percentage. It is the amount a bad month can carry while the rule still runs. It is not a budget, it does not know your rent, and it says nothing about whether that number is livable where you live.
- No investment return is applied to the cushion. An emergency cushion belongs somewhere boring and immediately available. Growth is not the point of it and none is assumed here.
- No inflation, no raises, no rate changes, and the same twelve-month pattern repeats. Real years do not.
- The strong/weak model uses two values, not a distribution. Real months land everywhere between your best and your worst. Two values make the mechanism inspectable; they do not make it a forecast.
- Nothing here is a recommendation to save any particular percentage, to spend any particular amount, or to hold any particular account. It is arithmetic on figures you supplied.
This is an estimate, not advice. The output above is arithmetic performed on the numbers you entered, under the assumptions printed on this page. It is not a recommendation, not a projection you should rely on, and not a substitute for a conversation with your CPA or a qualified adviser about your own situation.
Who this was built for
Most people with this problem are not unusual earners. They are ordinary earners on an unusual schedule: servers and bartenders whose income changes every shift, rideshare and delivery drivers with no withholding and a vehicle to feed, real estate agents and loan officers paid in lumps by a market they do not control, owner-operators living on settlement statements, dancers working a short peak-earning window with no employer plan of any kind, travel agents paid months after the work and sometimes clawed back, and entertainment crew whose years swing by a factor of three. All seventeen audiences we publish for are here.
Common questions
What percentage should I use?
We are not going to name one for you, because the honest answer depends on your fixed costs, your tax situation and your debts. What we will say is the structural point: pick a percentage you can apply in your worst month without flinching, and let the good months do the heavy lifting. A rule that survives February is worth more than a rule that looks impressive in June and is abandoned by March.
Should the cushion come before retirement saving?
For most people with variable income, a working cushion comes first, because without one every slow month becomes a credit card balance or an early withdrawal from the retirement account — and an early withdrawal usually costs tax plus a 10% penalty. Once the cushion is genuinely funded, the same percentage rule can split between the two. What we would not do is stop the retirement contribution for years while the cushion is built. Both can run at once, at smaller percentages.
My income dropped and I cannot hit the baseline. What now?
Then the baseline was set from a worst month that was not actually your worst. Re-enter the real figure and watch the baseline fall — that lower number is what your fixed commitments need to fit inside. This is uncomfortable and it is also the useful part: the tool is showing you which commitments are sized to a version of your income that is not reliably there.
Where should the cushion actually sit?
Somewhere liquid, boring and separate from the account your deposits land in — separate enough that moving money out takes a deliberate act. Beyond that, this page does not recommend specific accounts or products, and no calculator should. What matters far more than the vehicle is that the transfer happens on the day the deposit lands rather than at the end of the month.
Income that changes every month is a planning problem, not a character flaw
We work with people whose income arrives in lumps — tips, fares, commissions, settlements, residuals. Fee-only, fees published, $0 to open and $50 a month ongoing, and nothing to sell you.