Core answer: Annualized ROI converts any return to a per-year rate: for total return R over n years, annualized = (1+R)^(1/n) − 1. +50% over 3 years = 1.5^(1/3) − 1 = 14.5%/year — NOT 50÷3 = 16.7% (that's the simple average, which overstates). For cash flows in and out (SIP, dividends reinvested), use IRR/XIRR — the money-weighted truth.

Simple vs annualized: why the average lies

+100% then −50% over two years: simple average = +25%/yr; actual: 100 → 200 → 100 = 0% total, 0% annualized. Volatility always drags the simple average above the compound truth (volatility drag ≈ variance/2).

The conversion table

Total returnPeriodAnnualized
+21%2 yr10.0%
+33.1%3 yr10.0%
+50%3 yr14.5%
+100%5 yr14.9%
+100%10 yr7.2%
+300%10 yr14.9%

Doubling time check: Rule of 72 — at 10%/yr money doubles in ~7.2 years.

XIRR for real-world investing

You invest ¥2,000/month into a fund for 3 years (¥72,000 total) and it's worth ¥85,000: simple math says +18% total. But each installment had a different holding period — XIRR solves the rate that discounts every flow: ≈ 11.3%/yr here. Spreadsheets: =XIRR(values, dates). Any comparison between a SIP and a lump-sum needs XIRR on both sides.

Worked examples

Example 1 — Two funds. Fund A: +60% in 4 years (12.5%/yr). Fund B: +45% in 3 years (13.2%/yr). B wins on annualized despite the smaller total — duration matters.

Example 2 — The fee drag. Two identical portfolios, one charging 1.5%/yr more in fees: at 8% gross for 20 years, ¥1M becomes ¥4.66M at 8% vs ¥3.87M at 6.5% — fees consumed 17% of the outcome.

Example 3 — Inflation adjustment. 10% nominal with 3% inflation ≈ 6.8% real ((1.10/1.03)−1). Long-term planning must run on real returns or future-you gets a pay cut.

Common mistakes and myths

  1. Simple averaging — dividing total return by years overstates; compounding is multiplicative, not additive.
  2. Ignoring cash-flow timing — +20% on money deployed for 2 months ≠ +20% on 2-year money; annualize before comparing.
  3. Cherry-picking windows — 2019–2021 vs 2021–2024 fund returns tell opposite stories; always check multiple windows and full cycles.
  4. Nominal ≠ real — a 4% deposit with 3% inflation is a 1% real return; purchasing power is the only scoreboard that matters.
  5. Survivorship bias — average fund returns exclude dead funds; indices include everything. Compare your picks against the index, honestly.