Fifty shares bought at 120, now at 156, and 2.40 per share in dividends over the year. Most people call that a 30% gain and move on. The honest number is different, and how far off it is depends on two things the price chart does not show: the dividends, and how long you held.
I am a test manager, not an investment adviser, and I am not going to tell you what to buy. What I can do is arithmetic, and check whether the number on a statement is the number the inputs produce. That is what this post is: total return, annualised return, yield, what fees, tax and inflation take off before the money is yours, and the one comparison that says whether the result was any good.
Total return is the only number worth quoting
Total return is everything the position gave you: the change in price plus every dividend or distribution paid along the way.
Total return = (current value + dividends received, minus what you paid) / what you paid x 100
For the example: - Paid: 50 x 120 = 6,000 - Worth now: 50 x 156 = 7,800 - Dividends: 50 x 2.40 = 120 - Total return: (7,800 + 120 minus 6,000) / 6,000 x 100 = 32%
Two points on a single year is a rounding error. Over a decade or two it is not. Roughly a third of the long-run return on the S&P 500 has come from dividends, and a comparison that leaves them out makes every dividend payer look worse than it was.
The Percentage Calculator does the change-from-to step for you. Put in what you paid and what it is worth plus dividends, and read off the percentage.
Annualise before you compare anything
32% sounds good. Over four years it is less than a plain index fund did. Over six months it is exceptional. The number means nothing until you attach the holding period, and the way to do that is to convert it to a per-year figure.
Annualised return = ((1 + total return) ^ (1 / years) minus 1) x 100
For 32% over 2.5 years: (1.32 ^ 0.4 minus 1) x 100 = 11.7% per year. Check it on a calculator rather than trusting the rounding in someone's blog post, this one included.
The S&P 500 has averaged about 10% a year over the long run, so 11.7% is a little ahead of the index. Not a triumph, but not nothing either, provided the risk was comparable.
Where annualising changes the verdict is when two holding periods differ. 15% in 8 months annualises to about 23%. 25% in 18 months annualises to about 16%. The smaller gain was the better result.
The Compound Interest Calculator runs this the other way: put in a yearly rate and a horizon, and it shows what a position or a monthly contribution grows to.
32% sounds good.
Dividend yield and yield on cost answer different questions
Dividend yield is this year's dividend divided by today's price. A share paying 3.00 a year and trading at 75 yields 4%. That is the number to use when comparing two income stocks you might buy today.
Yield on cost is the same dividend divided by what you paid. If you bought that share at 60, your yield on cost is 5%. That is the number to use when deciding whether to keep a position you already hold.
The gap widens with time. A company that raises its dividend every year can show a 2% yield to a buyer today and a 6% yield on cost to someone who bought ten years ago, because the dividend grew and the purchase price did not. Both numbers are correct. They just answer different questions, and quoting one when the other was asked for is how people talk past each other about the same stock.
Fees, tax and inflation come off before the money is yours
The return on the statement is gross. Three things come off it.
Fees. Trading commissions are near zero at most brokers now, but fund costs are not. A fund charging 1% a year against one charging almost nothing, both earning 7% before costs, ends 20 years about 17% apart. Small annual numbers compound just like returns do, in the wrong direction.
Tax. This is where the country matters more than the stock. In the United States a gain on shares held over a year is taxed at a lower long-term rate, and a gain under a year as ordinary income. In the Netherlands there is no capital gains tax on shares held privately at all. Instead box 3 (the Dutch tax bracket for savings and investments) taxes a deemed return on your net assets above an allowance, every year, whether you sold or not and whether the shares went up or not. Which system you are in changes what a 12% year is actually worth, so work it out for your own case.
Inflation. 8% with 3% inflation is a 5% gain in what the money buys. Over ten years the difference between the two numbers is the difference between doubling and not.
Compare investments on the return after fees, and where you can, after tax. A high gross return with a high fee and unfavourable tax treatment can leave less in your account than a modest return with neither.
Buying in instalments changes the cost basis, not the method
Most people do not buy once. They buy a fixed amount every month, which means more shares when the price is low and fewer when it is high, and a cost basis that is the sum of all of it.
Total return still works the same way. Add up everything you paid, compare it with what the position is worth plus the dividends.
- Paid: (10 x 100) + (10 x 80) + (10 x 120) = 3,000
- Worth now at 115 a share: 30 x 115 = 3,450
- Total return: (3,450 minus 3,000) / 3,000 = 15%
Annualising is where instalments bite, because the money went in on different dates and each euro had a different amount of time to grow. The correct tool is the internal rate of return, and the practical way to get it is the XIRR function in Excel or Google Sheets: one column of dates, one column of cash flows (purchases negative, today's value positive), and it returns the per-year rate. I would not attempt it by hand, and I would not trust a per-year number from anyone who did.
Good compared to what? The index you could have bought instead
12% a year is a good result until you learn the index did 18% over the same period. A return only means something next to the alternative you turned down.
The alternative for most people is a cheap, broad index fund: the S&P 500 for US shares, a world index for a global portfolio, your own country's main index for domestic holdings. Same start date, same end date, dividends reinvested on both sides. That last part matters; index funds reinvest, and a comparison against the price-only index flatters your own picks.
If your chosen shares trail the index over several years, the index fund was the better decision, and it is not a personal failure. The S&P Dow Jones SPIVA reports have shown for two decades that most professional managers trail their benchmark after fees over any fifteen-year window.
Risk belongs in the comparison too. 8% a year with a worst drop of 10% and 12% a year with a worst drop of 40% are not the same result, and for most people the second one gets sold at the bottom. The Sharpe ratio (return above the risk-free rate, divided by volatility) puts a number on that, if you want one.
The practical test is one question: did this beat a total market fund over the same dates, after costs? If yes, the picking earned its time. If no, ask what the time was for.
12% a year is a good result until you learn the index did 18% over the same period.
FAQ
Do reinvested dividends count?
Yes. If a dividend bought more shares, those shares are part of what the position is worth today, and the cash you put in is still only the cash you put in. Value all the shares, subtract your own money, and the reinvested dividends are in the answer automatically. Do not add them again as income; that counts them twice.
What does a stock split do to the return?
Nothing. 100 shares at 200 become 200 shares at 100, and the position is worth the same. Halve the per-share cost basis in your own records so later arithmetic still works. Most brokers adjust it for you, but check.
What is a good yearly return?
The US market has averaged about 10% a year before inflation and about 7% after, over the long run, with individual decades far above and far below that. Anyone sustaining 15% a year for a decade is doing something most professionals cannot. Treat a claim like that with the same suspicion you would give any other number nobody has verified.
I bought US shares from the Netherlands. Which return is real?
The one in euro, because that is what you spend. A share that gained 10% in dollars while the dollar lost 5% against the euro gained about 4.5% for you. Convert the purchase at the rate on the day you bought and the value at today's rate, and calculate the return on the two euro amounts.
Tax Calculator for Self-Employed Freelancers: 2026 US Numbers
What a US freelancer owes in self-employment tax, the 2026 quarterly deadlines, the deductions worth the paperwork, and the share of each invoice to set aside.
Emergency Fund: How Many Months, Worked Out From Your Own Numbers Instead of a Rule
Three to six months of expenses is a range too wide to act on. Here is how to get to one figure: essentials from bank statements, a multiplier for how your income can stop, a deduction for the safety nets you already have, and where the money sits.
Financial Calculators: Loans, Compound Interest & Discounts
How to use financial calculators for loan amortization, compound interest projections, discount analysis, and everyday money decisions that matter.
Car Loan Calculator: The Monthly Payment Is the Wrong Number
How a car loan payment is calculated, why the 72-month offer costs the most, what a down payment buys you, and the one number to negotiate on.