Skip to content
PlanRetirement
Know Your Number. Plan Your Retirement.

Calculator methodology

Every assumption, convention and shortcut behind the numbers on this site, written out in full. If a result here differs from another calculator, the reason is almost certainly on this page.

Last updated 9 August 2026

One engine behind every calculator

All the tools on this site are powered by a single calculation engine rather than page-by-page formulas. This is a deliberate design choice with a practical consequence: the corpus the retirement corpus calculator tells you to build and the income the SWP calculator says that corpus can produce are the same month-by-month simulation, run in two directions. They cannot contradict each other, because they are not separate pieces of arithmetic.

How annual rates become monthly rates

This is the single most important convention on the site, and the one most likely to explain a difference between our figures and another calculator's.

When you enter 12%, we treat it as an annual effective rate and convert it to a monthly rate of (1.12)1/12 − 1 ≈ 0.9489% per month. Twelve of those months compound back to exactly 12%.

Most Indian SIP and SWP calculators instead divide by 12, using 1% per month. Twelve months of 1% compounds to 12.68%, not 12% — so a calculator using that convention is quietly modelling a higher return than the one you typed in. Over twenty years on a ₹10,000 monthly SIP the two conventions differ by roughly 8%.

We use the stricter basis because it is internally consistent (a lump sum compounds to exactly P × (1+r)n, as the textbook formula requires) and because it matches how Indian mutual fund returns are actually quoted, as CAGR or XIRR — both of which are annual effective figures. Our SIP calculator also displays the ÷12 result so you can reconcile against any other site.

When cash flows are assumed to happen

  • Investments are made at the start of each month, which is how a SIP mandate executes. This is an annuity-due calculation and gives each instalment one extra month of compounding.
  • Withdrawals are taken at the start of each month, before that month's return is credited. The remaining balance then earns one month of growth.
  • Annual increases — to a SIP step-up or to a withdrawal — are applied once every twelve months, after the twelfth instalment.
  • Inflation compounds annually, not monthly.

How the required corpus is solved

We do not use a rule of thumb such as 25× or 30× expenses, and we do not use a closed-form annuity formula. Instead the engine runs the full withdrawal simulation and searches, by bisection, for the opening balance that leaves exactly your chosen ending amount (zero by default) in the final month.

The simulation inflates your current expenses to your retirement date, withdraws monthly, increases the withdrawal every twelve months by your inflation assumption, and credits the remaining balance with your post-retirement return. The solver converges to within ₹1.

The same approach is used in reverse for the "sustainable withdrawal" figure: it is the monthly amount that lands the corpus at zero in the final month of the period you specified.

Inflation

Inflation is applied to expenses, not netted off returns, because those two approaches diverge over long horizons. Where we show a figure "in today's money" we divide the nominal figure by (1 + i)n. Where a real rate is needed we use the Fisher relation, (1 + nominal) ÷ (1 + inflation) − 1, rather than simply subtracting one from the other.

Our inflation calculator supports a different rate per budget category, because applying one blended average to a whole household budget understates the categories — chiefly healthcare — that inflate fastest and matter most in later life.

Dividend and distribution calculations

Dividend income is capital × yield. Payout growth compounds annually and is applied to the rupee amount, so yield on cost rises while the yield a new buyer would receive does not. The reverse calculation divides the target annual income by the yield. Where you supply a tax rate and mark the target as post-tax, we gross the requirement up by 1 ÷ (1 − tax rate).

We do not model share price movement, so these figures describe income only, never total return.

Returns measurement

CAGR is (end ÷ begin)1/years − 1. XIRR solves for the rate that discounts a series of dated cash flows to zero, using Newton-Raphson with a bisection fallback for awkward series, and a 365-day year. It returns no result where the cash flows contain no sign change, because no meaningful rate exists in that case.

Rounding and display

All arithmetic is carried out at full floating-point precision; rounding happens only at display. Rupee amounts are shown in the Indian lakh–crore system, with two significant decimals below ₹100 crore. Because we round for display, adding up the figures in a table by hand may occasionally differ from the total shown by a rupee or two.

What these calculators deliberately do not do

  • No volatility. Returns are applied smoothly, every month, at the rate you entered. Real markets deliver their average through large swings around it.
  • No sequence-of-returns modelling. A fall early in retirement is far more damaging than the same fall later, because you are selling units at depressed prices to fund living costs. A smooth-return model cannot show this. A dedicated stress-test tool is on the roadmap.
  • No tax on redemptions. Capital gains tax on SWP withdrawals is not deducted. Where a tax rate is offered, it applies only to the specific income stream being calculated.
  • No exit loads, expense ratios or transaction costs. Enter a net-of-cost return if you want the output to be realistic.
  • No pension, EPF, NPS annuity or rental income streams. Reduce your expense input by any guaranteed income to approximate this.
  • No product selection. Nothing here tells you what to buy.

Where your data goes

Nowhere. Every calculation runs in your browser. Inputs are saved to your device's local storage so that returning to a calculator restores your last plan; they are never transmitted to us or to anyone else, and clearing your browser data removes them permanently.

Corrections

If you believe a calculation on this site is wrong, we want to know. Write to us with the inputs you used and the result you expected, and we will check it, fix it if it is wrong, and note the change on the page. Our editorial policy sets out how we handle corrections.