The real company hiding inside the annual report.
Segment analysis strips away the averages and shows you what each business actually earns
The chairman's letter said 'strong diversified growth.' Page 180 told a different story.
A friend bought shares in a conglomerate last year. Good brand. Stable dividend. He read the summary financials and felt reassured. He never opened the segment note on page 180. That note would have shown him one division was carrying the entire company. The rest existed on thin margins — quietly subsidised by the one business that worked. He had priced four companies as one.
A consolidated P&L is a weighted average. When a company runs three or four different businesses, the final numbers blend everything together. A 15% EBIT margin might be one division at 40% pulling three others at 5% up to something acceptable. The consolidated number looks decent. The underlying reality is messier — and it matters enormously to what you are actually buying.
Ind AS 108 exists for exactly this reason. Every listed Indian company must disclose segment-wise revenue, profit or loss, and assets in the notes to its financial statements. The standard aligns with IFRS 8, the same logic used globally. The data is there every year. It usually sits on page 150 to 200 of the annual report — far past where most investors stop reading.
The most useful number segment analysis gives you is segment ROCE, not blended ROCE. Take a segment's operating profit. Divide it by the capital the company has deployed in that segment. You might find one division earns 60% on its capital while the division beside it earns 8%. The blended ROCE might show 22%. That gap is the real story. Capital creating 60% returns and capital creating 8% returns are fundamentally different assets. Averaging them does not tell you what you own.
Capital employed per segment also shows where management is deploying your money — regardless of what the chairman's letter says. A company that claims to be investing in its highest-potential businesses might be quietly channelling capital into a low-ROCE segment. Segment capex allocation is a forward-looking signal. Words are easy. The capex line is a decision that costs real money.
Two further traps are easy to miss. First, inter-segment revenue. When one division sells to another inside the same company, that sale appears in both divisional totals. Look for the elimination line in the segment note — without it, you are comparing numbers that overstate consolidated revenue. Second, unallocated corporate overhead: head office costs, group-level expenses, the CEO's salary — these sit outside all segment tables. They reduce total returns but appear in no segment's margin. A segment showing 30% EBIT can translate to 22% at the company level once unallocated overhead is absorbed.
The practical payoff of this analysis is sum-of-parts valuation. Pure-play companies trade at their own multiples: a standalone cigarette business commands a different price than a standalone hotel chain or a pure FMCG company. When you own a conglomerate, you own all three at a blended price. Sometimes the market prices the whole at a discount to what the parts would fetch separately — because most investors look only at blended numbers and miss the gap. Sometimes it prices the whole at a premium, embedding a complexity charge you may not want to pay. The segment note is where you check which situation you are in. That step — from reading financials to asking what each piece is independently worth — is what separates a price-taker from someone who actually knows what they own.
The tapri that's paying for the café
Imagine the same owner runs two chai businesses. One is a tapri on a railway platform — simple stall, low rent, constant footfall. Every rupee of capital invested earns back ₹4. The other is a fancy café — expensive rent, polished staff, Instagram-worthy décor. Every rupee invested earns back ₹1.10. The blended business looks acceptable on paper. But only the tapri is creating real value. The café exists because the owner loves it — and the tapri quietly funds it. Segment analysis forces you to look at each stall separately before you decide what to pay.
Why this matters
When you buy a stock, you buy all the segments — profitable ones and loss-making ones together. If the dominant segment is high-ROCE, asset-light, and growing, you own something genuinely good. If the dominant segment is commodity-driven and capital-heavy, the impressive brand in the chairman's letter is not what you actually own. Segment analysis moves you from what a company is called to what it does with money. That is the shift that separates a careful investor from someone reading headlines.
Set each segment's share. Watch the real company take shape.
Try the widget below. Set each of three segments' revenue share and EBIT margin. Watch the blended company ROCE and a plain-English verdict update live. You'll see how quickly one weak segment can drag down an otherwise excellent business — and how easy it is to miss that in a consolidated number.
The hidden company inside the annual report
Segment A generates 85% of profits but only 40% of revenue (a 2.1xx concentration). This gap shows the consolidated P&L is hiding where value is created and lost.
One segment is the company. Find it first.
Where people go wrong
- Trusting consolidated EBIT margin without checking segment breakdownOne high-margin segment can flatter the blended number significantly. The average tells you nothing about which business creates value and which quietly consumes it.
- Ignoring capital employed per segmentA segment can show respectable profit and still destroy value if it requires enormous capital to generate those profits. Profit without the capital cost is an incomplete and misleading picture.
- Treating rising consolidated revenue as encouraging newsRevenue growth concentrated in the lowest-ROCE segment is bad news dressed as good news. Growth in a capital-heavy, low-return division dilutes shareholder value even as the top line climbs.
- Missing unallocated corporate costs that sit below the segment tableThese belong to no segment's margin and are easy to overlook. Ignoring them makes every segment look more profitable than the company actually is at the group level.
The segment with 40% of revenue but 80% of profit is the real company — understand it before you price the stock.
Segment ROCE, not blended ROCE, tells you which division actually creates shareholder value.
Unallocated costs belong to no segment but reduce all group returns — check them before trusting any margin.
We read the chairman's letter. We admire the brand. We never open page 180. That is where the real company lives — and it is almost never what the cover suggests.