Half the salary, thirty-three years sooner
Someone on £120,000 spending £110,000 has a savings rate of 8% and roughly fifty-five years of work ahead of them. Someone on £45,000 spending £27,000 saves 40% and reaches independence in about twenty-two. The second person earns less than half as much and finishes thirty-three years sooner, which is the most counter-intuitive result in early retirement planning: how much you earn barely matters. The share you keep is nearly everything.
It works because the savings rate does two jobs at once. The money you save is what builds the pot, and the money you don't spend lowers the pot you need. Cut spending by £1,000 a year and you have added £1,000 to annual savings while removing £25,000 from the target. No other input touches both sides.
From 10% to 70%
Starting from zero, with a 5% real return and a 4% withdrawal rate:
| Savings rate | Years to independence |
|---|---|
| 10% | ~51 |
| 20% | ~37 |
| 30% | ~28 |
| 40% | ~22 |
| 50% | ~17 |
| 60% | ~12 |
| 70% | ~9 |
The curve is steepest at the bottom: going from 10% to 20% removes almost fifteen years, while 60% to 70% removes fewer than four. So a low rate is the opposite of hopeless. Sitting at 12% simply means the highest-value change available to you is a spending one, and you are standing exactly where the moves pay most; going from 12% to 20% removes roughly a decade. At the far end the maths turns fragile instead, because a 70% rate leaves little room to absorb a shock without derailing the plan entirely.
Notice, too, that the table never mentions pounds. A 40% savings rate produces roughly the same timeline whether you earn £30,000 or £300,000, because the target scales with spending. That target is the 25× calculator seen from the other end: it gives you the pot, this page gives you the timeline to reach it. Same model, viewed from opposite ends.
The same £5,000, spent twice
Kofi takes home £48,000 and spends £32,000. He saves £16,000 a year, a 33% savings rate, and his target at 4% is £800,000. From zero, that takes about 26 years.
Hand him a £5,000 pay rise and let him save every pound of it: his rate rises to 40% and the timeline falls to about 22 years. Nearly four years saved. Now rewind and give him a £5,000 spending cut instead. The rate rises to 44% and the target drops to £675,000, so the timeline falls to roughly 20 years: six years saved, from the same £5,000.
The cut is worth more than one and a half times the raise. That asymmetry is the single most useful thing this calculator has to teach, and it is why the sensitivity analysis puts spending at the top of the levers you control.
Measuring your own rate honestly
Use take-home income, not gross, and count pension contributions as savings. Measuring against gross makes your rate look better while telling you nothing about the spending your pot must eventually replace. The mortgage splits down the middle: the interest is spending, the capital repayment is saving because it builds equity, and many people simplify by counting the whole payment as spending, which is conservative and fine. Calculating the rate honestly takes the rest of the awkward cases one at a time, and says what each band actually buys you.
Where the saving goes matters almost as much as how much of it there is. Saving 40% into a pension is not the same as saving it into an ISA you can draw on at any age, because if you are under 50 today your pension stays locked until at least 57, an access age the government has already legislated. Tax treatment shifts over a career as well, and none of that appears in the timeline above. The pension bridge exists for exactly that split between accessible and locked money.
Two more assumptions sit under every number on this page. The model starts you from zero, and existing savings shorten everything, often dramatically; the FIRE calculator accounts for what you already have. And it holds your savings rate constant for decades, which no real career does. Pay rises, redundancies, children and house moves all bend the line, and the return assumption still matters as well, just less than the rate. Treat the answer as a direction, not a date.