ROI vs CAGR: Why a 50% Return Can Be a Poor Result
.png)
Somebody tells you their investment made 50%. Sounds great. And it might be, if it took a year. If it took ten years, a fixed deposit would have done about the same and you'd have slept better. The number on its own just doesn't say which one you're hearing about, and that's what this post is about.
If you only read one bit
- A 50% gain is 50% a year if it took one year, and only 4.14% a year if it took ten.
- An 80% gain over 8 years (7.62% a year) actually loses to a 40% gain over 3 years (11.87% a year).
- Tax and trading costs can take a 50% stock gain down to 39.73%.
- With 6% inflation, a 50% gain over 3 years is closer to 26% in real terms.
- A SIP that ends with a 36.1% ROI really earned about 12% a year. Plain ROI hides that.
Every figure here comes from standard formulas. The prices and rates are round numbers I picked so the maths is easy to follow, so please don't read them as quotes for any real stock, fund or bank.
ROI in one line
Return on investment is what you got back, as a percentage of what you put in.
Put in ₹1,00,000 and end up with ₹1,50,000, and that's 50%. End up with ₹80,000 and it's −20%. That really is all there is to it, which is why people use it for everything: shares, a flat, a course, a machine for the shop, an ad campaign.
But a total is only a total. It doesn't know how long your money was stuck, what you paid along the way, or what else you could have done with it. Let's go through those one by one.
Fifty percent of what, over how long?
Take the same 50%, and just change the number of years. ₹1,00,000 becomes ₹1,50,000 each time. The last column is CAGR, which turns the total into a rate per year so you can compare.
| Time taken | ROI (total) | CAGR (per year) |
|---|---|---|
| 1 year | 50% | 50.00% |
| 3 years | 50% | 14.47% |
| 5 years | 50% | 8.45% |
| 8 years | 50% | 5.20% |
| 10 years | 50% | 4.14% |
The ROI column doesn't budge. The yearly rate slides from superb to something plenty of savings schemes beat. So if anyone quotes you a return without saying how long it took, ask. It's the first question I'd put to anybody selling you something.
Why I wouldn't compare two returns using ROI
Say you're looking back at two investments and trying to decide which kind to repeat.
| Investment A | Investment B | |
|---|---|---|
| Start | ₹1,00,000 | ₹1,00,000 |
| End | ₹1,80,000 | ₹1,40,000 |
| Time | 8 years | 3 years |
| ROI | 80% | 40% |
| CAGR | 7.62% a year | 11.87% a year |
On ROI, A wins easily. On the yearly rate, B is clearly better, and B is the one that would have grown your next ₹1 lakh faster. Once two investments ran for different lengths of time, ROI stops being a fair way to rank them. That's the one place I'd always switch to CAGR.
Then there's the tax bill
ROI is only as honest as the number you type in, and most of us type in the sale price. Here's the same ₹1,00,000 to ₹1,50,000 share trade with the usual delivery-trade costs added: 0.1% STT on each side, a flat ₹15.34 depository charge on the sale, and 20% short-term capital gains tax if you sell within 12 months.
| Headline | Sold within 12 months | |
|---|---|---|
| Gross gain | ₹50,000 | ₹50,000 |
| STT (both sides) | - | ₹250 |
| Depository charge | - | ₹15.34 |
| Capital gains tax (20%) | - | ₹10,000 |
| What you keep | ₹50,000 | ₹39,734.66 |
| ROI | 50% | 39.73% |
More than ten points gone. Hold the same shares past 12 months and the picture changes again. A ₹50,000 long-term gain sits inside the ₹1.25 lakh yearly exemption, so you'd keep ₹49,734.66, or 49.73%. Same shares, same prices, about ₹10,000 apart, and the only difference is a date on the calendar.
So when you work out ROI for yourself, put the after-cost, after-tax figure in the final value box. The fees box in the calculator further down is there for exactly that.
Don't annualise a short trade
What if that 50% gain had come in just 8 months? Convert it to a yearly rate and you get 83.7% before tax, about 65% after. Tempting to quote. But it's mostly a trick of the arithmetic. You can't count on finding another trade like that for the other four months, and then another after it. Stretching one lucky result over a full year makes it look like a habit. For anything under a year, I'd just look at the rupees you made and leave it there.
Inflation quietly takes its cut
A 50% gain over three years feels wonderful until you remember that prices went up as well. At 6% inflation a year, what cost ₹1 lakh at the start costs about ₹1.19 lakh three years later. Your ₹1.5 lakh is really worth about ₹1.26 lakh in today's money.
So the headline said 14.47% a year, and the real figure, after inflation, is about 8%. It's still a good result. It's just roughly half of what you'd have said out loud.
Losses are meaner than gains
Percentages are not symmetrical. If something falls 50%, it has to rise 100% to get you back to where you started.
| If your investment falls by | You need a gain of this much to break even |
|---|---|
| 10% | 11.1% |
| 20% | 25.0% |
| 33% | 49.3% |
| 50% | 100.0% |
| 75% | 300.0% |
It's the reason I'd rather avoid one big drop than chase one big win. Two portfolios with the same average return can end up in very different places because of it.
SIPs break the formula
ROI assumes all your money went in on day one. A SIP doesn't work that way. Money goes in every month, so most of your rupees were only invested for a fraction of the period.
Put ₹10,000 a month into something earning 12% a year for 5 years. You've paid in ₹6,00,000 and you end up with about ₹8,16,697. That's a 36.1% ROI. If you then try to turn it into a yearly rate by treating the whole ₹6 lakh as if it went in at the start, you get about 6.4% a year. The truth is 12%. Both ROI and CAGR make this investment look worse than it was, because the average rupee was only in there for around two and a half years.
For regular investing, use a SIP calculator or XIRR, which takes the timing of every instalment into account.
The flat that "made 60%"
You buy a flat for ₹50 lakh and sell it ten years later for ₹80 lakh. ROI is 60%, which sounds like a decent deal. Now look closer.
- The yearly growth rate is only 4.81%.
- An illustrative 7% a year deposit, compounded over those same ten years, turns ₹50 lakh into about ₹98.4 lakh. That's a 96.7% ROI.
To be fair, the flat may have earned rent, and the deposit has no upkeep or brokerage, so a proper comparison would include both. But the lesson holds: 60% only sounds good until you put it next to the simple thing you could have done instead. Economists call that opportunity cost. ROI never mentions it.
Try it with your own numbers
Reality check
Fill in what you put in, what you ended with, how long it took, and any fees or tax you paid. You'll get ROI, the yearly rate, and the yearly rate after inflation.
When plain ROI is perfectly fine
None of this makes ROI a bad number. It's the right one when a decision has a clear start and finish, like a piece of equipment that pays for itself, a course that got you a raise, or a single ad campaign. It's also fine when you're comparing things held for the same length of time, and it's the quickest answer to "did I make or lose money on this?" The trick is knowing when to stop at ROI and when to move on to a yearly rate.
What I'd check before believing any return
- How long did it take? A return with no time period isn't really information.
- What's the yearly rate, if I'm comparing it with something else?
- Is the final value after tax and costs, or just the sale price?
- What would a plain deposit or index fund have done over the same stretch?
- How much of it was eaten by inflation?
- Was the money put in all at once, or in instalments? If it's the second, plain ROI will mislead.
Questions people ask
What counts as a good ROI?
It depends on how long it took and how risky it was. 10% in six months is excellent. 10% over five years isn't. Turn it into a yearly rate first, then compare it with something low-risk over the same period.
Can ROI be negative?
Yes. If you end with less than you put in, it's negative. An ROI of −20% means you got back 80% of your money, and as the table above shows, you'd need a 25% gain to recover.
Is CAGR always better than ROI?
No, they answer different questions. CAGR is better when you're comparing investments that ran for different lengths of time. ROI is better for a single result or when the periods match. Looking at both is the best habit.
Does ROI include dividends or rent?
Only if you add them in. If a share paid dividends or a flat earned rent, count that income as part of what you got back. Otherwise ROI undersells the investment.
Why doesn't my mutual fund's return match my own ROI?
Fund fact sheets show annualised figures for fixed periods, while your own result depends on when you started and stopped. If you invested through a SIP, XIRR is the fairer measure.
This article is for general information and education. It is not financial, tax or legal advice. The prices, rates and examples are illustrative round numbers, and tax rules can change. Calculations use standard formulas. Past performance does not predict future results, and market-linked investments carry risk. Check current rules or speak to a qualified adviser before making investment or tax decisions.
Join the conversation