TL;DR. A sinking fund is money set aside every month for a specific future expense you already know is coming. An emergency fund covers the unknown. A sinking fund covers the entirely predictable thing you keep pretending is a surprise: the car service, the insurance renewal, December. Sizing one is trivial arithmetic, which is why the failure is never mathematical. It is the missing monthly ritual. The sinking fund planner does the division, and the transfer itself belongs in whatever you already do at the end of the month.
Here is the pattern, and I have run it myself more than once.
Something expensive arrives. The car needs two tyres, the insurance renews, the laptop that has been making a noise since spring finally stops. You cover it, slightly grimly, out of whatever was meant to be the buffer. And the story you tell about it is that you got unlucky this month.
You did not get unlucky. Tyres wear out. Insurance renews annually. Laptops last about four years. Every single one of those was on a schedule you could have read in advance, and the only genuinely unexpected thing about them was that you had decided not to look.
That is the entire problem a sinking fund solves. It is not a clever financial instrument. It is a decision to stop being surprised by the calendar.
The definition, in one line
A sinking fund is money you put aside monthly for a known future expense, so that when it arrives it is already paid for.
The term is borrowed from corporate finance, where a sinking fund is money set aside to retire a debt by a fixed date. The household version keeps the useful structure: a target, a date, and a monthly amount that connects the two.
The distinction most articles blur
Emergency fund and sinking fund get used interchangeably, and they are not the same object at all.
An emergency fund is for the unknown. You cannot name what it is for. That is the definition. Job loss, the sudden medical thing, the failure you did not see coming.
A sinking fund is for the known. You can name it, price it and date it. That is also the definition.
The reason the distinction is worth defending is not pedantry. It is that when you have no sinking funds, every predictable expense becomes an emergency by default, and your emergency fund spends its life being drained by things that were never emergencies. Then when something actually unpredictable happens, the buffer is thin, and you conclude that you are bad with money.
You were not bad with money. You had one bucket doing two jobs.
The categories that catch people out
Run through this list and mark the ones that have actually disrupted you in the last two years. Not the ones that theoretically could.
| The expense | Why it hides | Roughly how often |
|---|---|---|
| Car servicing, tyres, the test | Never the same month twice | Annual, plus a surprise |
| Insurance renewals | Arrives as one large annual charge | Annual, on a fixed date |
| Christmas and the birthdays | Feels emotional, not budgetary | Annual, entirely predictable |
| Travel | Booked in a burst, paid across months | One or two a year |
| Dental and optical | Skipped until it is urgent | Every six to twelve months |
| Annual software renewals | Auto-renews while you are not looking | Annual, quietly |
| The replacement laptop or phone | Has no date until it does | Every three to five years |
| Vet bills | Not optional and not scheduled | Unpredictable timing, predictable existence |
The last one is a useful edge case. You cannot know when your dog needs something, but you can know with near certainty that a dog costs money at some point, which makes it a sinking fund with a soft date rather than an emergency.
The planner on this site ships with eleven named categories plus a custom one, and the named list covers most of the table above: vacation, car maintenance, gifts and holidays, home repair, insurance, medical, education, tech, clothing, pets and taxes. The category is only a label and an icon. The work is in the numbers you attach to it.
Sizing each one
Total needed, minus what you have already set aside, divided by the number of whole months until you need it.
That is the whole formula, and the sinking fund planner is essentially that division with a progress bar attached. You give each fund a name, a target amount, a target date, the amount you have saved so far and a priority of high, medium or low, and it returns the monthly figure plus a weekly version for people who think in weeks.
Two things about it are worth knowing before you trust the numbers.
The monthly figure is recalculated from today, every time. If you skip two months, the tool does not flag it. It just quietly divides the same remaining amount by fewer months, and the required contribution goes up. That is arithmetically correct and psychologically sneaky, so check whether a number that has grown means the target got bigger or you got slower.
Priority sorts, it does not allocate. There is a what-if panel where you enter a monthly budget and it shows how far that budget gets each fund. The split is proportional to what each fund needs, and priority only controls the order of the list. If you genuinely want one fund funded first, do that yourself by lowering the targets or pushing back the dates on the others.
Everything is formatted in US dollars regardless of your currency, and everything stays in your browser.
How many is too many
More than you think, and fewer than you want.
The tempting version of this system has fourteen immaculate categories, one for every kind of expense you can imagine, colour coded. It lasts about six weeks. The reason is not discipline, it is friction: fourteen funds means fourteen decisions and fourteen transfers on the same tired evening, and a ritual that takes forty minutes is a ritual you will start skipping.
Four to eight is the workable range for most people. Merge anything you can. "Car" instead of servicing and tyres and the test. "Health" instead of dental and optical. You lose a little precision and you gain the thing that actually matters, which is that the monthly action stays small enough to survive December.
If you only ever run one, make it the annual-renewals fund. It is the least emotionally interesting and it removes the most unpleasant surprises.
Where to keep them, and what I will not tell you
Separate enough that you do not spend it by accident. Reachable enough that you do not resent getting it out.
Beyond that I am not going to give you an answer, and you should be a little suspicious of blogs that do. The right arrangement depends on where you live, what is actually available to you, and things about your circumstances I know nothing about. Anyone confidently naming a product to a global audience is guessing.
What I will defend is the principle. Money that sits in the same balance you spend from gets spent, and not because you are weak. It gets spent because you plan against the number you can see, and if the number includes eleven months of Christmas savings then your sense of what you can afford this week is simply wrong.
The part nobody writes about: the transfer is a habit
Every sinking fund guide stops at the spreadsheet, which is why so few of them work.
The maths takes four minutes. The system then depends on a small, recurring, deeply unexciting action: moving the money, once a month, forever. No one will remind you. The planner will not chase you, because it is a page in a browser with no account and no notifications.
Which makes this, structurally, a habit problem rather than a money problem. And habits that depend on remembering fail in exactly the way you would expect. What works instead is attaching the transfer to something that already happens on a schedule.
That is why I would run it inside an end-of-month routine rather than as its own free-floating intention. The monthly reset checklist is built around exactly that window: three phases, thirteen default tasks, and a nudge that appears in the last three days of the month. Two of the default items in the first phase are already financial, reviewing the month's spending and cancelling one subscription you did not use, so adding "fund the sinking funds" as a custom item to that phase puts the transfer next to the only other time you look at money on purpose.
If your natural reset is weekly rather than monthly, the same logic applies to the Sunday reset. The point is not which ritual. It is that the transfer has to ride on something that already exists.
When a sinking fund is the wrong tool
- If the immediate problem is that money is going out faster than it comes in, funding twelve future expenses is not the first move. Find the leak first. Running the annual arithmetic on the recurring stuff usually takes about four minutes and reorders the list.
- If you are mid-way through a hard reset like a no-spend challenge, do not start eight funds at the same time. Finish one thing.
- If the expense is genuinely unknowable, that is an emergency fund. Do not dress it up with a target date to make it feel organised.
- If you have tried this three times and abandoned it, the problem is the ritual, not the categories. Cut to two funds and one transfer.
The version I would actually run
- Name the three to six expenses that have actually disrupted you recently.
- Give each one a real amount and a real date, even if the date is a guess.
- Let the planner do the division and write down the total monthly figure.
- If that total is not affordable, push dates back or lower targets now, not in month four.
- Put the transfer in your end-of-month routine as a named item, so it survives the month you are tired.
None of that is clever. Sinking funds are not a clever idea. They are the observation that most of what wrecks a month was on the calendar the whole time, and that fifteen minutes of looking at it in advance is cheaper than every alternative.
Last updated: December 2026.
Written by Eusebiu, the solo founder building Loggd in public. I build the app in the evenings around contract work and a small daughter, and contract income arriving in lumps is exactly the situation that taught me predictable expenses need their own bucket. I share what the aggregate data keeps showing on Threads.
The transfer is a habit, so track it like one. Loggd shows a monthly recurring action as a forgiving contribution grid, which means a skipped month reads as one lighter square instead of a broken streak. Start free.
