How to Read CAGR: The Honest Way to Annualize Multi-Year Growth

Someone hands you two numbers: an account that grew from $1,000 to $2,000, over five years. How fast did it actually grow per year? The tempting answer is to take the 100% total gain, divide by five, and call it 20% a year. That answer is wrong, and it is wrong in a direction that flatters whoever is quoting it.

The compound annual growth rate, or CAGR, is the fix. It is the single steady yearly rate that would carry your starting number to your ending number if every year compounded on the one before. For the example above, the real figure is about 14.87% a year, not 20%. That gap of five percentage points is the whole reason CAGR exists.

The CAGR formula, written out

CAGR is one clean equation:

CAGR = (end / start)^(1 / years) − 1

You divide the ending value by the starting value, raise the result to the power of one over the number of years, and subtract one. The output is a decimal you read as a percentage.

The structure tells you why it works. Dividing end by start gives the total growth multiple. Taking the years-th root spreads that multiple evenly across each year. Subtracting one strips out the original principal so you are left with just the rate. There is no averaging of yearly numbers here, no path dependence, no calendar guesswork. Two values and a time span are all it needs.

A worked example you can check by hand

Take the $1,000 to $2,000 case over five years.

CAGR = (2000 / 1000)^(1 / 5) − 1
     = 2^(0.2) − 1
     = 1.1487 − 1
     = 0.1487, or about 14.87% a year

Now prove it forward. Start with $1,000 and apply 14.87% five times:

• Year 1: 1000 × 1.1487 = 1148.70
• Year 2: 1148.70 × 1.1487 = 1319.51
• Year 3: 1319.51 × 1.1487 = 1515.72
• Year 4: 1515.72 × 1.1487 = 1741.11
• Year 5: 1741.11 × 1.1487 = 2000.00

The money lands exactly on $2,000. That is the defining property of CAGR: it is the rate that hits the ending value on the nose. The simple-average rate of 20% would overshoot, because each year it would compound on a base that is too large. You can run any pair of values through the CAGR Calculator and watch the same reconciliation happen, with the simple total sitting right next to the annualized figure for comparison.

Why CAGR beats a simple average

The simple average treats each year as if it grows on the original base. Compounding does not work that way. Every year grows on last year's larger balance, so the same finish line is reachable at a lower yearly rate. Divide-by-years always overstates, and the longer the period, the bigger the lie.

Here is a sharper case. Suppose 100 becomes 200 over three years. Total growth is 100%. Divide by three and you get 33% a year. But check it: 100 at 33% would go to 133, then 177, then 235 — well past 200. The honest CAGR is about 26% a year, the rate that lands on 200 and not a dollar more. The simple average was off by seven percentage points on a three-year span, and that error compounds with the timeline.

I learned to distrust the simple average the hard way. Early on I reported a side project's revenue as having "grown 300% in five years" and let a reader divide that to 60% a year in their head. The real CAGR was about 32%. When a sharper reader did the math and asked why my headline number was nearly double the annual rate, I had no good answer — just a sloppy total dressed up as a rate. Now I run every multi-year claim through the annualized number first, because that is the figure people actually benchmark against.

Comparing investments and revenue on equal footing

CAGR's real power shows up when the time spans differ. One fund returns 60% over four years; another returns 90% over seven. The raw totals say the second one won. But annualize them and the first is about 12.5% a year while the second is about 9.6% a year. The shorter holding was the stronger annual performer the whole time — the longer one just had more years to pile up a bigger total.

The same logic cleans up a revenue story. Going from $1.2M to $4.8M over five years is a 300% total gain, which sounds enormous and means little. The CAGR is about 32% a year, and that is the line a board or investor can compare against peers, against your last raise, against a market. A total return cannot be compared across different lengths of time; an annualized rate can. That is also why pairing CAGR with a quick Rule of 72 estimate is handy — once you know the annual rate, the rule of 72 tells you roughly how long the money takes to double, which sanity-checks whether a quoted return is even plausible.

The limitation: CAGR smooths volatility

CAGR is honest about the start and the finish, and silent about everything in between. It reports the single smooth rate that connects two endpoints, which means it erases the journey entirely.

Consider an investment that goes 100 → 200 → 50 → 100 over three years. The CAGR is 0% a year, because it ended exactly where it started. That number is technically correct and practically useless if you cared about the white-knuckle ride: a double, then a 75% crash, then a recovery. Two investments with identical CAGRs can have wildly different risk profiles — one a steady climb, the other a roller coaster — and CAGR cannot tell them apart.

So treat CAGR as a summary, not the whole story. It answers "what was the equivalent steady annual rate?" cleanly and fairly. It does not answer "how bumpy was the path?" or "what was the worst drawdown?" For those you need the year-by-year series, standard deviation, or a max-drawdown figure. CAGR is the right headline number; it is the wrong number to make a risk decision on by itself.

The quick takeaways

• Use CAGR = (end / start)^(1 / years) − 1, never total growth divided by years.
• The simple average always overstates the true annual rate, and the error grows with the time span.
• CAGR lets you compare investments and revenue across different holding periods on equal footing.
• It smooths volatility, so two very different rides can share one CAGR. Pair it with a measure of dispersion before judging risk.

Get the formula right once and a five-year, an eight-year, and a one-year result all line up on the same per-year scale. That comparability is the entire point — and it is the one thing a raw total can never give you.

---

Made by Toolora · Updated 2026-06-13