Practice Years of Money in Minutes
Browse more posts about Money Simulations →
The strongest argument for saving is one nobody gets to hear in time. A choice made this month shows its full effect in about a decade, by which point the choice is long past changing and the person who made it has forgotten making it.
That delay is the central difficulty in teaching money, and no amount of explaining fixes it. What can help is compressing the distance, so the consequence arrives while the decision is still fresh. This guide is about doing that honestly, and about why the assumptions behind any long projection matter more than the figure it produces. Nothing here is financial advice.
Why a decade is the hard part
Short-term money is legible. Overspend this week and the account tells you by Friday. The feedback is fast, unambiguous, and it teaches.
Long-term money gives you nothing. Put aside an extra amount each month and for the first two years almost nothing observable happens. The balance is a little larger. It does not feel like a different life, because it is not yet one. Everything interesting is in years six through fifteen, and no habit survives on a promise that distant without some way of seeing it.
This is why the standard advice fails even when people believe it. They are not unconvinced. They are unable to picture it, and a thing you cannot picture loses every argument to a thing you can, such as whatever you would rather buy this week.
One decision, two futures
The useful format is not a projection. It is a comparison.
A single number ten years out is easy to dismiss, because the assumptions are invisible and it has no alternative to be measured against. Run two versions of the same decade that differ by exactly one decision, and put them side by side, and the argument makes itself. This is the one thing a simulation can do that lived experience cannot: nobody gets to run their own life twice.
The decision worth testing is usually one of two. Save a bit more each month, or spend a bit less on the recurring things. They sound like the same choice and they produce noticeably different shapes, because one adds to the pile and the other reduces what is drawn from it. Watching them diverge is considerably more persuasive than being told that small amounts add up.
The assumptions are the whole product
Any projection is only worth what its assumptions are worth, so an honest one states them where you can see them and lets you change them.
The version in CustomBank is deliberately narrow. It models cash and savings interest only. It works in today's money, with no inflation applied. It applies no market returns, because a projection that quietly assumes a rate of stock market growth is really just presenting that assumption as a conclusion. The starting figures come from your own months where the app has them, and every one of them is editable.
Narrow is the right choice here, and it is worth being clear why. A model with more moving parts is not more accurate; it is more confident, and each additional assumption is another place for a wrong number to hide behind a plausible total. A model you can hold in your head is one you can argue with.
It also means the output should not be read as a forecast. It is the arithmetic of a decision carried forward, which is a different and more modest claim.
What the compression actually teaches
Three things tend to land, and none of them is the final number.
The first is that the early years look like nothing and the later years do all the work. Seeing the curve stay flat and then move is the most common moment of genuine surprise, and it is the exact intuition that makes someone start earlier rather than later.
The second is how much of the outcome is decided by the monthly amount rather than by the interest rate. People arrive expecting the rate to be the lever and leave understanding that what you put in dominates, which is fortunate, because the amount is the part you control.
The third is a sense of scale. Most people either wildly overestimate what a small monthly amount becomes, or wildly underestimate it. Both errors lead to giving up, for opposite reasons. Seeing the real figure, with stated assumptions, corrects both. The mechanics underneath are covered properly in the glossary entry on compound interest.
Practice Tip: Run a decade, then run it again changing only the start date by five years. The gap between those two is the cost of waiting, and it is almost always larger than the gap produced by changing the monthly amount.
Using it with a class or at home
This works where a compound interest lesson usually does not, because the student supplies the inputs.
A worked example on a whiteboard belongs to the teacher. A projection built from a number the student chose belongs to the student, and the difference in attention is obvious immediately. Having everyone run the same decade with their own assumptions, then comparing, also surfaces the real lesson faster than any explanation: the people who put in more got more, and the ranking barely moved when they fiddled with the rate.
At home it does something slightly different. It is a way to have the conversation about long-term saving without it being about anyone's actual finances, which tends to make it a shorter and less defensive conversation. Age-appropriate saving lessons cover what to introduce when.
What it cannot tell you
It cannot tell you what will happen. Income changes, costs change, and a decade contains at least one thing nobody modelled. Those are not edge cases; they are what a decade is made of.
It also deliberately excludes investment returns, so it is not a retirement calculator and should not be used as one. Anything involving real money over that horizon deserves your own research and, where the stakes justify it, a qualified professional who knows your situation.
What it can do is make the shape of a long decision visible while there is still time to make it. Nothing in the app is a real account: the balances are simulated, no financial institution is involved at any point, and it is a free download on iPhone and Android.