Cash flow forecast

Most budgeting tools tell you what you spent. That is the easy half, and it is the half that never changes anyone's decisions. The question people actually have is forward-looking: if nothing changes, what does my balance look like in March, and is there a week between now and then where it quietly goes underwater?

This projects your balance forward for twelve months, day by day, from what comes in and what goes out. It runs entirely in your browser - no account, no bank connection, nothing sent anywhere.

The number that matters is the lowest point

Ending the year up is not the same as getting through it. A forecast that finishes comfortably can still dip below zero in a particular week, because annual bills, insurance renewals and quarterly payments do not spread themselves evenly. That dip is the number to plan around, and it is the one a monthly average hides completely.

Look at the lowest point above and the date attached to it. If it lands close to zero, the fix is usually timing rather than income: move one annual payment to a different month, or hold back a small buffer in the weeks before it.

Why weekly and monthly items don't mix cleanly

A weekly expense is not a monthly one divided by four. There are 52 weeks in a year, not 48, so four times a week understates weekly spending by about eight percent - and that error compounds across every weekly line you have. Some months carry five paydays or five grocery weeks, which is why a budget that balances on paper still fails in specific months.

This tool places every occurrence on its real date rather than averaging. A monthly bill starting on the 31st lands on the 28th in February and returns to the 31st afterwards, which is what your bank will actually do.

What to do with the projection

Change one line at a time and watch the lowest point rather than the end balance. Cancelling a subscription moves the end figure by a rounding error; moving a large annual payment out of the month that already contains your insurance renewal can move the lowest point by hundreds.

The limitation worth naming: this assumes your commitments continue as entered. It cannot know about the car repair in August or the pay rise in June. It is a model of your current commitments, which is precisely why it is useful for spotting structural problems and useless for predicting a surprise.

Keeping a forecast accurate is the hard part

A forecast is only as good as the last time you updated it, and that is where this kind of tool usually dies. You build one carefully in January, a subscription changes in March, and by May the projection quietly describes someone else's finances.

That is the argument for the ledger and the forecast living in the same place as the spending itself. When a recurring payment posts its own transaction and the projection updates from real entries rather than from what you typed months ago, keeping it honest stops being a chore you have to remember - which is how the Money hub is built.