Almost every story told about a company is a growth story, and almost every growth number quoted inside one has been chosen because it flatters. Growth analysis is the small set of habits that stops a headline percentage from doing your thinking for you: knowing which growth rate is being quoted, what it is measured against, what the path looked like in between, and who paid for it. This topic covers year-on-year, quarter-on-quarter and compound annual growth, the specific way each one misleads, and the gaps between revenue growth, profit growth and earnings-per-share growth that reveal what actually happened inside the business.
Why does the growth rate you quote change the story?
Growth is not a property of a company. It is a property of a company and a chosen window, and changing the window changes the answer without anybody having lied.
The same set of filings will support a five-year growth rate, a three-year growth rate, a rate measured from the pre-pandemic year, a rate for the latest quarter against the same quarter last year, and a rate for the latest quarter against the one before it. All of those are arithmetically correct. They will not agree with each other, and whichever one appears in a headline is usually the largest of them.
This is why the first question about any growth number is never 'is it high?' but 'what exactly was measured, and against what?' A reader who asks that consistently is already ahead of most of the commentary they will encounter.
Note — Company presentations are not doing anything improper by selecting a favourable window - they are marketing documents, and the underlying filings are public for anyone who wants to recompute. The failure is in repeating the headline without checking which window produced it.
YoY, QoQ and CAGR - three different questions
Three measures cover almost everything you will meet. They answer genuinely different questions, and each carries a specific failure built into how it is constructed.
- YoY (year-on-year)
- This period compared with the same period one year earlier - for a quarter, Q2 this year against Q2 last year. Because it compares like with like, it cancels seasonality automatically, which is why company commentary almost always leads with it.
- QoQ (quarter-on-quarter)
- This quarter against the quarter immediately before it. Also called sequential growth. It is the fastest way to notice a turn, and it is meaningless for any business whose quarters were never comparable to begin with.
- CAGR (compound annual growth rate)
- The single constant annual rate that would have carried the starting value to the ending value over the whole period. It is a smoothing device: it describes two endpoints and deliberately discards everything between them.
| Measure | What it is good for | How it misleads |
|---|---|---|
| YoY | Comparing like with like; cancels seasonality | One weak or freak base period distorts it completely |
| QoQ | Spotting a turn a quarter or two before an annual number would | Useless for seasonal businesses, and four QoQ figures do not annualise |
| CAGR | Comparing long records across companies and time periods | Hides the path entirely and is hostage to the exact start and end years chosen |
How do you calculate CAGR?
CAGR asks a simple question: if this business had grown at one steady rate every year instead of whatever it actually did, what rate would have got it from the start to the finish?
CAGR = (Ending value / Beginning value) ^ (1 / Number of years) - 1- Ending value - the figure in the final year of the period
- Beginning value - the figure in the base year
- Number of years - the years spanned, not the count of data points. Five annual figures from FY1 to FY5 span four years, not five
- Multiply the result by 100 to express it as a percentage
Take Deccan Fasteners Ltd, an illustrative company invented for this topic along with every figure attached to it. It reported revenue of Rs 500 crore in its base year and Rs 1,000 crore five years later. The ratio of ending to beginning is 2.0. Raise 2.0 to the power of one-fifth and you get 1.1487; subtract 1 and the compound annual growth rate is 14.87%, call it 14.9% a year.
The check that catches most arithmetic mistakes is to run the rate forward. Rs 500 crore compounding at 14.87% goes 574, 660, 758, 871, 1,000. If your series does not land on the ending value, the exponent is wrong - almost always because the years were counted as data points.
Two properties of the formula are worth internalising. It only uses two numbers, so anything that happened in between is invisible to it. And it is not intuitive: most people asked to estimate the annual rate that doubles a business in five years say something near 20%, and the answer is under 15%. Compounding is consistently underestimated over short horizons and overestimated over long ones.
Pro tip — Never eyeball a CAGR. Any spreadsheet does it in one line and a phone calculator with an exponent key does it in seconds, and the mental estimate is reliably wrong in the same direction.
What does a CAGR hide?
It hides the path, and the path is often the more important half of the information. Consider the same Rs 500 crore to Rs 1,000 crore journey travelled two different ways.
| Year | Steady path (Rs cr) | Lumpy path (Rs cr) |
|---|---|---|
| Base year | 500 | 500 |
| Year 1 | 574 | 900 |
| Year 2 | 660 | 620 |
| Year 3 | 758 | 700 |
| Year 4 | 871 | 1,150 |
| Year 5 | 1,000 | 1,000 |
Both columns are illustrative. Both report a 14.9% five-year revenue CAGR, and on a screen sorted by that column the two businesses are indistinguishable. They are not remotely the same business.
The steady path describes customers who came back every year. The lumpy path - nearly doubling, falling most of the way back, then doubling again - is the signature of something quite specific: a few very large orders, a commodity price passing straight through the top line, or a project business where one contract dominates whichever year it gets recognised in.
That difference has practical consequences beyond neatness. Factory capacity and working capital have to be sized for the peak rather than the average, which is expensive and shows up as capital sitting idle in the trough years. And any forecast built off the lumpy company's final year is really an extrapolation of whatever happened to be in that year's order book.
The habit that fixes this costs almost nothing: never accept a CAGR without seeing the yearly series it was computed from. Then test its stability by moving the start year forward by one. If a five-year CAGR of 15% becomes a four-year CAGR of 4%, the entire number was carried by one unusual base year and it is not describing a trend at all.
What is the base effect, and how does it manufacture a headline?
The base effect is what happens when a growth rate is computed against an abnormally small denominator. It produces spectacular percentages that are arithmetically true and directionally misleading.
Stay with Deccan Fasteners. Suppose revenue collapses from Rs 1,000 crore to Rs 400 crore in a bad year - a plant shutdown, a lost anchor customer, a frozen end-market. The following year it recovers to Rs 760 crore. The press release writes itself: revenue up 90% year on year. Every word is accurate and the impression is false, because Rs 760 crore is still 24% below where this business was two years earlier.
Measure across the whole disruption instead - Rs 1,000 crore to Rs 760 crore over three years - and the compound rate is about minus 8.7% a year. A plus 90% headline and a minus 8.7% three-year CAGR are describing exactly the same company over overlapping periods.
This is rarely deception. It is simply what the arithmetic does when the denominator is small, and it appears on a predictable schedule: the year after a demand shock, the year after a one-off write-off crushed profit, the year after a fire or a regulatory suspension, and throughout the early years of any company whose starting base was tiny.
Watch out — The base effect is far more violent in profit than in revenue. Profit can fall by 90% and still be positive, so the recovery year prints a growth rate in the hundreds of percent. Any three-digit profit growth figure is nearly always telling you that last year was close to zero, not that this year was exceptional.
Two questions defuse it every time. What was the base year, and was it normal? And how does this year compare with the last normal year, rather than with the last year?
For a business that has been through a disruption, the most informative single number is usually a multi-year CAGR measured from before the disruption to today, because it forces the collapse and the recovery into the same calculation. A company that has fully recovered will show a modest positive rate; one that is still below its old peak cannot hide behind a large recovery percentage.
Revenue growth, profit growth and EPS growth - what does a gap between them mean?
Three growth rates are worth reading together, always in this order: revenue, profit after tax, and earnings per share. When they are close, the business is simply selling more of the same thing at roughly the same economics. When they diverge, the divergence is the actual story - and it is always locatable in a specific line between the top of the income statement and the bottom.
| What you observe | What it usually means | Where to check it |
|---|---|---|
| Profit growing faster than revenue | Operating leverage as fixed costs spread over more sales, better pricing, cheaper inputs, lower interest after repaying debt, or a lower tax rate | Operating margin across three years, the interest line, the effective tax rate, and the other income line |
| Profit growing slower than revenue | Input costs rising faster than selling prices, a new plant adding depreciation before it adds sales, or higher interest on the borrowing that funded the growth | Gross and operating margin, the depreciation line, the finance cost line |
| EPS growing slower than profit | The share count rose - a qualified institutional placement, a rights issue, warrants converting, ESOPs vesting, or shares issued to pay for an acquisition | Weighted average shares outstanding in the notes, and the quarterly shareholding pattern |
| EPS growing faster than profit | The share count fell, which in practice almost always means a buyback | Shares outstanding across years, and the financing section of the cash flow statement |
The third row is the one that catches people, because it is where total profit and per-share profit part company. A company can raise its profit by a quarter and still deliver almost nothing per share if it issued a large number of new shares to fund that profit. Total profit is a fact about the business; earnings per share is a fact about your slice of it, and those are not the same claim.
EPS = Profit after tax / Weighted average number of shares outstanding- Weighted average, not the year-end count - shares issued in March count for only part of the year
- This weighting is exactly why EPS growth and profit growth can point in different directions in the same year
Example — Deccan Fasteners, illustrative throughout, grows profit after tax by 25% while issuing new shares that lift its weighted average share count by 25%. Profit growth of 25% and EPS growth of roughly zero are both true statements about the same year. Which one gets quoted is a choice.
Who paid for the growth?
Revenue can be bought. Growth funded by the business's own cash, growth funded by lenders and growth funded by new shareholders can look identical on the top line and mean entirely different things about the years ahead.
- Funded by operations - cash from operations covers the capex and the extra working capital the growth needs. The business is paying for its own expansion, which is the version that compounds without anybody's permission
- Funded by debt - the growth is real, and so are the interest bill and the repayment schedule. Revenue and borrowings rising together means the return on the new capital has to beat the cost of that debt, or the growth is subtracting value while adding sales
- Funded by issuing shares - cash arrives with no repayment obligation, and every existing shareholder ends up owning a smaller fraction. Total profit can grow while profit per share does not move
- Funded by acquisition - revenue jumps because another company's revenue was added to it. This is not the same as the existing business growing, and most companies disclose an organic figure separately for exactly that reason
The fastest check available is to line up three series for the same five years: revenue, total borrowings, and shares outstanding. Growth with debt and share count flat is a completely different animal from identical growth with both climbing, and the comparison takes a couple of minutes from the annual reports.
How much debt is too much for a given business depends on how stable its cash flows are, which the debt and leverage topic works through. Whether the operating cash flow funding the growth is genuine is the subject of the cash flow statement topic, and the extra working capital that fast growth consumes is taken apart in the working capital and cash conversion cycle topic. Growth analysis is where those three threads first meet.
Note — None of this makes debt-funded or equity-funded growth wrong. Capacity that takes three years to build has to be paid for long before it earns anything, and refusing all external funding would rule out most industrial expansion. The point is only that growth and the cost of that growth are read together, never separately.
Why is quarter-on-quarter growth meaningless for many Indian businesses?
The Indian financial year runs April to March, so the quarters map onto the Indian calendar in a way that matters enormously for demand. For a large number of listed businesses, consecutive quarters were never comparable in the first place.
- Q1 (April to June)
- The summer quarter. Air conditioners, fans, cold beverages and paints do a disproportionate share of their annual business here, and construction is active in the weeks before the rains arrive.
- Q2 (July to September)
- The monsoon quarter. Construction slows, cement dispatches soften, outdoor activity falls. Agricultural inputs and rural demand move with how the rains actually turn out rather than with anything the company did.
- Q3 (October to December)
- The festive quarter. Diwali pulls consumer durables, jewellery, apparel, two-wheelers and gifting-linked demand forward into it, and the wedding season adds to the same window.
- Q4 (January to March)
- The year-end quarter. Budgets get spent before they lapse, projects get closed out, and companies that recognise revenue on completion book a bulge that says more about the calendar than about demand.
So a paint company reporting lower revenue in Q2 than in Q1 has not shrunk - it has had a monsoon. A jeweller whose Q4 sits below its Q3 has not lost customers - Diwali was in Q3. Comparing consecutive quarters in these businesses measures the calendar, not performance, and no amount of commentary makes the comparison valid.
There are two fixes. Compare the same quarter year on year, which is why nearly all company commentary leads with a YoY number. Or use the trailing twelve months, which always contains exactly one of each quarter and is therefore seasonally neutral by construction.
TTM figure = sum of the last four reported quarters- Rolls forward every quarter - drop the oldest quarter, add the newest
- Always contains one of each seasonal quarter, so seasonality cancels without waiting for the annual report
- Useful for any metric reported quarterly: revenue, profit, EPS, and the ratios built from them
QoQ is not useless. For a business with genuinely level demand - a subscription-like service, a steady industrial supplier - a sequential number is the first place a change shows up, usually two or three quarters before an annual figure would register it. It is misleading only when the two quarters being compared were never comparable, and knowing which case you are in requires knowing what the business sells.
The reading quarterly results topic later in this curriculum covers the order to read a results release in, and why the segment note and the commentary often carry more signal than the headline numbers.
How do you sanity-check a growth number in five minutes?
This runs against a company presentation, a news headline or a screener column, and it catches most of the ways a growth figure can be misleading.
The five-minute growth check
- Identify which measure is being quoted - YoY, QoQ, CAGR, or something bespoke like growth since FY19
- Find the base year and ask whether it was a normal year for this business
- Look at the full annual series, not just the two endpoints
- Move the start year forward by one and see whether the CAGR survives
- Read revenue, profit and EPS growth side by side, and explain any gap between them
- Check whether the top line grew because a company was acquired
- Put total borrowings and shares outstanding next to the revenue series
- For any quarterly figure, confirm the comparison is like quarter against like quarter
Is there such a thing as a good growth rate?
There is no universal threshold, and any number offered as one is being quoted without its context. Growth only means something measured against three things: the company's own history, what its sector is growing at, and nominal GDP growth. That last comparison is the one most often skipped - nominal GDP includes inflation, so a business growing slower than nominal GDP is shrinking as a share of the economy even while its rupee revenue rises.
The size of the base changes the meaning entirely. A company going from Rs 50 crore to Rs 100 crore and a company going from Rs 5,000 crore to Rs 10,000 crore both report 100% growth, and the second did something vastly harder. High percentage growth off a small base is arithmetically easy and says very little; sustained double-digit growth on a large base is rare and says a great deal.
And growth is not free. It consumes working capital and capital expenditure, and it is entirely possible for a business to grow itself into a cash crisis. A company growing revenue at 30% while its receivables grow at 60% is financing its customers rather than selling to them - a pattern the working capital topic takes apart, and one of the recurring signatures examined in the accounting red flags topic.
The most useful reframing is this: growth is an input to value, not a synonym for it. Growth that earns more than the capital it consumes adds value; growth that earns less destroys it while looking impressive on a chart. Which of the two you are looking at is answered by the return ratios covered in the profitability topic, not by the growth rate itself.
Key points
CAGR = (Ending value / Beginning value) ^ (1 / Number of years) - 1
Example — Deccan Fasteners Ltd is an illustrative company - the name and every figure are invented for teaching. Revenue of Rs 500 crore grows to Rs 1,000 crore over five years, which is a CAGR of 14.9%. Compounding forward gives 574, 660, 758, 871, 1,000, confirming the arithmetic. But the same 14.9% would also describe a path of 900, 620, 700, 1,150, 1,000 - identical endpoints, an entirely different business underneath.
Pro tip — Read the three growth rates as one line, in order: revenue, then profit, then EPS. Every gap between them is locatable in a specific line of the accounts, and hunting down which line moved teaches you more about the business than the growth rate itself ever will.
Warning — A three-digit growth number is a statement about last year's base, not about this year's performance. Before repeating any large percentage, find out what the denominator looked like and whether that year was normal.
Frequently asked questions
How do you calculate CAGR?
Divide the ending value by the beginning value, raise the result to the power of one divided by the number of years, and subtract 1. Revenue rising from Rs 500 crore to Rs 1,000 crore over five years gives 2.0 raised to the power of 0.2, which is 1.1487, so the CAGR is 14.87%. Count the years spanned rather than the number of data points - five annual figures from FY1 to FY5 span four years.
CAGR vs YoY growth - which is better?
They answer different questions. YoY tells you what happened in the most recent period against the same period a year earlier, so it is current but hostage to a single base year. CAGR smooths several years into one rate, so it describes a record but discards the path completely. Reading them together is the point - a high CAGR alongside a weak latest YoY figure usually means the growth belongs to the past.
What is a good CAGR for a company?
There is no universal number, because the answer depends on the sector, the size of the base and where the cycle is. The useful comparisons are the company's own history, what its sector is growing at, and nominal GDP growth - a business growing slower than nominal GDP is losing share of the economy. High percentage growth from a very small base is arithmetically easy and tells you far less than a moderate rate sustained on a large base.
Why is quarter-on-quarter growth misleading?
Because for many businesses consecutive quarters were never comparable. In the Indian financial year, Q1 is the summer quarter, Q2 is the monsoon, Q3 carries the festive season and Q4 carries year-end budget spending, so a fall from Q3 to Q4 in a consumer business is usually the calendar rather than performance. Comparing the same quarter year on year, or using the trailing twelve months, removes the effect.
What is the base effect in earnings growth?
It is the distortion created when a growth rate is calculated against an abnormally small prior figure. If revenue collapses from Rs 1,000 crore to Rs 400 crore and then recovers to Rs 760 crore, the headline reads plus 90% year on year even though the business is still 24% below where it was two years earlier. The defence is to compare against the last normal year, or to compute a multi-year CAGR that spans the disruption.
Can CAGR be negative?
Yes. When the ending value is lower than the beginning value, the formula returns a negative rate, which is the correct description of a business that has shrunk over the period. Negative CAGRs are genuinely useful for cutting through recovery headlines - a company can post a large positive YoY figure while its three-year CAGR is still negative.
Why is EPS growth different from profit growth?
Because EPS divides profit by the weighted average number of shares outstanding, so anything that changes the share count changes EPS without changing profit. New shares from a placement, a rights issue, converting warrants, vesting ESOPs or an acquisition paid for in stock all push EPS growth below profit growth. A buyback does the reverse. The share count is disclosed in the notes to the accounts and in the quarterly shareholding pattern.