Valuation Basics — P/E, Earnings Yield & DCF Intuition

In the last lesson, Reading a Company, Suresh taught you to open a company up and read it — the profit it makes, the earnings per share (EPS — the year's net profit sliced across every share), the book value (what the company would be worth on paper if it paid off every debt and sold everything at the value in its accounts), and how efficiently it turns shareholders' money into profit (its return on equity, ROE). You learned to tell a well-run business from a shaky one. This lesson asks the next question, and it is the one that quietly decides whether you make money or lose it: even if the company is wonderful, what is it worth — and is today's price a fair one to pay?

Here is the fear worth naming out loud, because almost every thoughtful beginner carries it: a great company can still be a terrible investment if you overpay for it. You can be completely right about the business — the best product, the cleanest balance sheet, the smartest managers — and still lose money, for years, simply because you paid too much on the day you bought. That is not a rare tragedy; it is the single most common way careful, intelligent people lose money in the stock market. They fall in love with the company and forget to check the price tag. So the honest question — 'how do I know what's a fair price?' — is not a beginner's anxiety to be soothed away. It is the exact right question, and the whole of this lesson is the answer.

That estimate of what a business is really worth has a name you'll meet again and again: its intrinsic value — the value that comes from the actual cash the business will earn its owners over its life, as opposed to the price the market happens to quote today. The market's price and a company's intrinsic value are two different things. Usually they are close. Sometimes — in a euphoria or a panic — they drift far apart, and that gap is where both the opportunity and the danger live. Valuation is simply the craft of estimating intrinsic value, so you can tell whether the price on offer is a gift, a fair deal, or a trap.

Suresh, 55, is a chartered accountant in Kochi — he earns about ₹40,00,000 (forty lakh) a year, sits in the top tax slab, and has built up roughly ₹1.8 crore across equity mutual funds, a book of shares he picks himself, and property. He is exactly the person who should know this material cold, and he does. But he'll be the first to tell you that the toolkit you're about to learn is not a machine that spits out 'buy' or 'sell'. It is a set of lenses that turn a vague feeling ('this seems expensive') into a checkable claim ('at ₹1,800 this is priced for 20% growth every year for a decade — do I believe that?'). By the end you will hold the same lenses. You will not become an oracle. You will become harder to fool — including by yourself.

The most quoted number in all of investing is the price-to-earnings ratio — the P/E. You'll hear it everywhere: 'the stock trades at 25 times', 'the market's on a P/E of 21', 'that one's expensive at 60'. Strip away the jargon and it is the simplest idea imaginable. The P/E is the share price divided by the earnings per share. It answers one plain question: how many rupees am I paying today for each ₹1 of the company's yearly profit?

Let's put it on a real example and keep it there for the whole lesson, so every number ties back to one company. Suresh has been studying Sunvale Consumer Ltd — an illustrative, well-run maker of everyday household products, the same company whose accounts he read in Lesson 27. Sunvale's shares trade at ₹1,800 each. Last year it earned ₹40 of profit per share — that's its EPS. Divide one by the other and you have its P/E.

Sit with what 45× actually means, because the '×' hides two very human readings. The first: for every single rupee of profit Sunvale earned last year, the market is asking you to pay forty-five rupees. The second, and more sobering: if the company's profit never grew a paisa from here and simply paid every rupee out to you, it would take forty-five years of profits to earn your purchase price back. Forty-five years. That is what a high multiple is — a long, patient bet that the profits will be much larger in the future than they are today. It is not automatically foolish. But it is a bet, and naming it as one is the first act of valuation discipline.

The widget above breaks the same 45× into its plain-English pieces and then does something more important: it places that number against the market's own history. Because here is the trap every beginner falls into first — thinking a P/E is 'high' or 'low' on its own. It never is. Forty-five is meaningless in a vacuum. Forty-five for a fast-growing, cash-rich consumer champion might be reasonable; forty-five for a sleepy, no-growth utility would be lunacy. A multiple only becomes 'cheap' or 'dear' the moment you set it beside something: the company's growth, its quality, its industry, the stage of the economic cycle, and — the anchor we'll use next — its own long-run average and the market's.

So is Sunvale's 45× expensive? Suresh answers a question like this the way a doctor reads a blood-pressure number — never against zero, always against what's normal for this kind of patient. He reaches for three comparisons, and you should too.

1. Against the company's own history. If Sunvale has typically traded around 35–40× over the last decade, then 45× is at the dear end of its own range — the market is more enthusiastic about it now than usual. If it normally trades at 50×, then 45× is actually a touch subdued.
2. Against its peers. A 45× multiple for a consumer company whose rivals trade at 40–55× is unremarkable; the whole sector is prized for steady, defensive earnings. The same 45× for a bank or a steelmaker, whose peers sit at 8–15×, would be extraordinary and demand a very special reason.
3. Against the market as a whole. The broad index gives you the 'going rate' for ₹1 of Indian corporate profit. A stock far above the market's multiple is one the market has singled out as special — you're paying a premium, and premiums have to be earned.

That third anchor — the market's own multiple — is worth pinning down, because it's the reference point professionals use every day. As Suresh checks this (mid-2026), the Nifty 50 — the index of India's fifty largest listed companies you met in Lesson 26 — trades at roughly 20.9 times its earnings. Its average over the last ten years has been about 23.4 times. So the going rate for ₹1 of large-company Indian profit is around 21 rupees, and the market today sits a shade below its own decade average — neither screaming cheap nor obviously frothy. Against that ~21× backdrop, Sunvale's 45× is a big premium: the market is paying more than twice the going rate for Sunvale's profits, on the belief that those profits will grow far faster than the average company's.

The cycle matters too, and it plays a cruel trick that the next-but-one section will expose in full. For a steady consumer company like Sunvale, earnings are fairly smooth, so its P/E is reasonably honest. But for a cyclical business — a steelmaker, a carmaker, a commodity producer — profits boom and bust with the economy. At the very top of a boom, such a company reports huge profits, so its P/E looks tiny and 'cheap'; at the bottom of a bust, profits vanish and the same P/E looks enormous or meaningless. For cyclicals, the P/E is at its most seductive exactly when it is most dangerous. Hold that thought — Koyna Steel will make it painfully concrete.

There's a hidden question buried inside every P/E, and sloppy investors never ask it: which earnings? The 'E' in the ratio can mean three quite different things, and mixing them up is how people accidentally compare a fair price to a fantasy.

Sunvale's 45× is a trailing P/E — it uses the ₹40 the company actually earned last year, a real, banked number. If analysts expect Sunvale to earn ₹47 next year, then its forward P/E is ₹1,800 ÷ ₹47 ≈ 38× — lower, because you're dividing by a bigger, hoped-for profit. Notice how the very same price can be called '45×' or '38×' depending purely on which earnings you pick. Neither is a lie; but a salesman who wants the stock to look cheap will always quote you the forward number built on the most optimistic forecast. When someone tells you a P/E, your first question should be: trailing or forward — and if forward, whose forecast?

Now flip the P/E upside down. If the P/E is price divided by earnings, then earnings divided by price is its mirror image — and this flip, called the earnings yield, is one of the most clarifying moves in all of valuation. It turns a stock into something you can compare directly to a bank deposit or a bond: an interest rate.

Read that carefully: Sunvale's earnings yield is 2.2%. That is the 'interest rate' the business itself throws off on your purchase price — not the dividend it hands you, but the total profit it earns behind each rupee you invested. A high P/E is always a low earnings yield, and vice versa; they're the same seesaw. The reason to bother with the flip is that a percentage is instantly comparable to every other percentage in your financial life — the 6-and-a-bit percent on a fixed deposit, the yield on a government bond, the rate of inflation. Suddenly you can ask a question a raw '45×' could never answer: is this stock paying me enough?

Here is the comparison that separates disciplined investors from hopeful ones. On one side, Sunvale's earnings yield: 2.2%. On the other, the closest thing to a risk-free return an Indian investor can get — the yield on a 10-year government bond, the G-sec, which the government is effectively certain to honour. As Suresh checks it in mid-2026, that 10-year G-sec yields about 6.8%. (You'll meet the G-sec itself as an instrument in Lesson 33; for now it's just 'the safe rate', the risk-free rate you first met back in Lesson 1, and it moves with the RBI's rate cycle from Lesson 9.) And behind both sits inflation, running near 4%, quietly deciding how much of any return is real.

Look hard at the bars, because the picture is uncomfortable and honest. Sunvale hands you an earnings yield of 2.2%. A government bond — no company risk, no volatility, guaranteed — hands you 6.8%. Even the whole market's earnings yield, at about 4.8% (that's the flip of the Nifty's ~20.9× P/E), sits below the bond. On the face of it, this looks absurd: why on earth would you accept 2.2% from a risky share when a safe bond pays 6.8%? A beginner who stops here concludes 'stocks are a terrible deal, buy bonds' — and gets the single most important idea in equity investing exactly backwards.

This is why the earnings-yield-versus-bond check is the discipline it is. It doesn't say 'never buy stocks that yield less than bonds' — most quality growth stocks always will. It says: know exactly how much you are paying for growth, and be honest about whether the growth is real. When the gap is comfortable — the stock yields close to or above the bond — the growth is almost a free option. When the gap is a chasm, as with Sunvale's 2.2% against 6.8%, you are paying up entirely for a future that must arrive on schedule. There is no room for disappointment in that price. That thin cushion is precisely why the next ideas — a discount to fair value, a margin of safety — exist.

If a high P/E is a bet on growth, then the obvious next question is: am I paying a fair price for that growth, or too much? The PEG ratio — price/earnings-to-growth — is the quick, rough tool built to answer exactly that. It takes the P/E and divides it by the company's earnings growth rate, written as a plain number.

The rule of thumb, made famous by the fund manager Peter Lynch, is that a PEG of around 1.0 is roughly fair — a company growing at 20% deserves a P/E near 20, one growing at 40% can carry a P/E near 40. Below 1, you may be getting growth cheaply; well above 1, you're paying a premium even after crediting the growth. Sunvale's PEG of 2.5 says something the raw 45× hid: even after generously accounting for its fast 18% growth, you are still paying two-and-a-half times what the simple rule would call fair. The market isn't just paying for Sunvale's growth; it's paying for growth plus quality, durability, and the comfort of a business that rarely disappoints. Whether that extra is worth it is a judgement — but PEG at least makes the premium visible.

The P/E and its family measure a company against its profits. Two more rulers measure it against different things — what it owns, and what it pays out — and each is the right tool for a particular kind of business.

Price-to-book — the ruler for asset-heavy businesses

Price-to-book (P/B) compares the share price to the company's book value per share — the net worth on its own balance sheet, per share, that you met in Lesson 27. Sunvale's book value is ₹200 a share, so its P/B is ₹1,800 ÷ ₹200 = 9.0×. At first glance that looks wild — paying nine times the company's paper net worth. But P/B has a golden rule: it is meaningless on its own and only makes sense next to return on equity (ROE).

Here's the logic. A company earning a 20% ROE turns every ₹100 of shareholders' money into ₹20 of profit a year — so that ₹100 of 'book' is a little engine throwing off ₹20 annually. Of course the market pays several times ₹100 for such an engine; the paper value badly understates it. A high P/B is justified by a high ROE. The danger is the reverse: a company trading at a low P/B — even below 1, below its own paper worth — that also earns a feeble ROE. That's not a bargain; it's the market's verdict that the assets aren't earning their keep. P/B earns its place for asset-heavy businesses where book value is real and comparable — banks and lenders above all, along with insurers, and heavy industry. For those, P/B (always read beside ROE) is often more useful than P/E; you'll see it front and centre when the fixed-income and banking lessons arrive.

Dividend yield — the cash the company hands back

Dividend yield is the plainest ruler of all: the yearly dividend per share divided by the price — the cash the company actually posts to you, as a percentage of what you paid. Sunvale pays a ₹12 dividend on its ₹1,800 share, so its dividend yield is 0.7%.

Every ratio so far has been a shortcut — a fast way to compare a price to profits, growth, assets, or payouts. Underneath all of them sits the one true idea of what a business is worth, and it is worth understanding even though you will never build the full spreadsheet as a beginner. It is called a discounted cash flow — DCF — and the whole of it fits in a sentence: a business is worth all the cash it will hand its owners over its life, added up, but with future rupees counted for less than today's.

Two everyday truths power it. First, a business is only ultimately worth the cash it can eventually pay you — everything else is a proxy for that. Second, a rupee in your hand today is worth more than a rupee promised in ten years, because today's rupee can be invested and grow, and because a distant promise is less certain. So to value a company you imagine the stream of cash it will throw off — next year, and the year after, on and on — and then you shrink each future year's cash back to what it's worth in today's money before adding it up. That shrinking is called discounting, and the rate you shrink by is the discount rate.

The left half of the widget shows the intuition as a picture: Sunvale's profit, growing year by year, with each future year's rupees drawn smaller than the last as they're pulled back to today. Add up all those shrunk-down future profits and you get an estimate of what one share is worth right now — its intrinsic value. Let's do it for Sunvale, in the simplest honest way: project its ₹40 of earnings per share growing at 18% for a decade, allow it to keep growing slowly forever after, and discount the whole stream back at 12%. The sum comes to about ₹1,550 a share.

Now to the most important truth in this whole lesson, and the one that keeps a valuation from becoming a false idol. That ₹1,550 came from two assumptions: 18% growth and a 12% discount rate. Nudge either one a little — the kind of small disagreement two sensible analysts have every day — and watch what happens to the answer in the grid on the right of the DCF widget.

Read across and down and let it unsettle you a little. The 'fair value' of the very same company, the same ₹40 of earnings, ranges from about ₹650 to about ₹3,100 — nearly a fivefold spread — depending on assumptions that all sound reasonable. Believe Sunvale grows 22% and you deserve a 10% return, and it's worth ₹3,100, a bargain at ₹1,800. Believe it grows 10% and you want 14%, and it's worth ₹650, wildly overpriced at ₹1,800. Neither person has made an arithmetic mistake. They've made slightly different guesses about an unknowable future, and the machine has amplified those small differences into enormous ones.

There's a neat way to feel the same thing from the other direction, and Suresh loves it because it cuts through the false precision. Instead of asking 'what's Sunvale worth?', run the machine backwards and ask: 'what would I have to believe to justify paying ₹1,800?' Do that, at a 12% discount rate, and the answer is that the price already assumes Sunvale grows its earnings at around 20% a year for a decade — a touch faster even than the optimistic 18% estimate. In other words, at ₹1,800 the good news is already in the price. You're not being handed the growth; you're being asked to pay for a slightly-better-than-expected version of it in advance. That reframing — 'what's priced in?' — is often more useful than any single fair-value number.

If a fair value is really a wide range, and if your assumptions about the future are guaranteed to be wrong in ways you can't predict, then buying a stock at your best single estimate of its worth is a mug's game — because half the time your estimate will prove too high, and you'll have overpaid. The answer, and it is the oldest and best idea in careful investing, is the margin of safety.

Notice how neatly the margin of safety absorbs everything you've learned. The fair value is a wide range, not a point — so you buy near the bottom of the range, not the middle. The DCF is deliberately generous — so you knock a big discount off it. The earnings yield sits below the bond — so you demand a lower price that lifts the yield toward something that compensates you. Every source of uncertainty in the lesson is answered by the same move: pay less. And here is the quiet, liberating part — the margin of safety means you never have to be right about the exact number. You only have to be roughly right, and disciplined about the price. That is a game a careful beginner can actually play.

It also explains what to do when — as with Sunvale at ₹1,800 — a wonderful business simply isn't offering a margin of safety today. You don't have to buy it. 'A great company, no margin of safety at this price' is a complete and respectable conclusion; it means 'wonderful business, wait for a better price, or own it through a broad index instead.' There is no rule that you must own every good company right now. The overpaying you feared at the start of the lesson is defeated not by finding the perfect valuation, but by refusing to buy without a discount to a generous one.

Time to make good on a promise, and to bury the beginner's favourite mistake forever: the belief that a low P/E means cheap and a high P/E means expensive. Suresh sets two companies side by side. One is Sunvale, our quality compounder on a nosebleed 45× P/E. The other is Koyna Steel Ltd — an illustrative cyclical steelmaker — trading at a P/E of just 8×. If the multiple alone told the truth, Koyna would be the obvious bargain and Sunvale the obvious folly. The truth is close to the opposite.

Look at Koyna first, because it is a beautiful trap. At ₹120 a share on ₹15 of earnings, its P/E is 8.0× and its earnings yield a mouth-watering 12.5% — nearly double the government bond. It trades at just 0.8 times book value, below the paper worth of its own factories. On the ratios, it screams 'bargain'. And it is a trap, for reasons the multiple hides completely. That ₹15 of earnings is a cyclical peak — steel prices are high this year, so profits are fat; when the cycle turns, as it always does, earnings could halve to around ₹9, and the 'cheap' 8× P/E quietly becomes a not-cheap 13× on normalised earnings. Worse, the business is shrinking — earnings are set to fall about 5% a year — so its PEG comes out at 8 ÷ (−5) = −1.6, a negative number that isn't a bargain but a broken gauge. And it earns a feeble 9% ROE, which is why it trades below book: the market has correctly judged that its assets barely earn their keep. A low P/E on peak, shrinking earnings in a low-return business is the textbook value trap — cheap for a reason, and the reason is that it deserves to be.

Now Sunvale, the dear-looking one. Yes, 45× is a demanding price and a 2.2% earnings yield is thin. But behind it is a business growing 18% a year, earning a 20% ROE, rarely stumbling — a genuine compounding engine. Its high multiple is at least earned, even if, as we found, the current price leaves no margin of safety. So the two companies flip your intuition inside out: the 'cheap' one is the dangerous one, and the 'expensive' one is the quality one that's merely priced too high for today. The multiple alone lied about both. Only when you added growth, ROE, durability and the cycle did the real picture appear.

A good wealth manager or fund analyst does something specific with these tools — and it's a discipline you can run yourself, for free. Here it is, taken apart, so you can copy the logic and judge whether anyone charging you a fee is actually earning it.

The move, decoded: the professional flips every stock into its earnings yield and sets it against the risk-free G-sec, so 'expensive' and 'cheap' become concrete instead of vibes; then they estimate a rough intrinsic value and refuse to pay more than a discount to it — a margin of safety; and, crucially, they never let the phrase 'but it's a great company' justify any price at all. The logic is everything you've just learned: price and value differ, the future is a range, and a wonderful business bought without a discount can still lose you money. The do-it-yourself substitute is genuinely within reach — you can flip a P/E into an earnings yield in your head, compare it to the ~6.8% bond, and ask 'what growth does this price assume, and do I believe it?' You will not do it as fast or as deeply as a professional. But you can absolutely do it well enough to smell danger.

Valuation language is catnip for fraudsters, because a ratio quoted with confidence sounds like analysis even when it's bait. The scam this lesson arms you against wears the costume of a bargain: the 'undervalued' penny stock or tiny SME 'set to 10× because its P/E is so low', pushed by a tipster on Telegram, YouTube, or a WhatsApp 'research' group.

The mechanics are old and simple. Operators quietly accumulate a thinly-traded small stock, then pump it — flooding tip channels with 'multibagger' calls, often citing a single seductive number: 'trading at just 3× earnings, criminally undervalued, target 10× in six months.' Retail buyers pile in, the price balloons, the operators dump their shares onto the crowd, and the price collapses — leaving the latecomers holding a worthless stock. The low P/E was never a bargain; it was either built on a fabricated or one-off 'earnings' figure, or it was a genuinely low multiple on a genuinely dying or fraudulent business — the value trap of the last section, weaponised. Remember Koyna: a low P/E on bad earnings is not cheap, it's a warning. On an outright scam stock, it's the hook.

Maybe reading this stung a little, because you've already done the thing it warns against — bought a company at a nosebleed multiple near the top of a hype cycle, watched it fall, and felt the specific shame of realising you overpaid for something everyone was excited about. Set the blame down. This is the most common investing mistake there is, and it is made by professionals with Bloomberg terminals, not just beginners with an app. You were not stupid. You were human, in a crowd, during a story that felt certain at the time.

Here is what the tools in this lesson tell you to do now, calmly. First, remember mean reversion: a stretched multiple tends, over time, to come back toward its average — sometimes by the price falling further, but often by the earnings quietly growing into the price while it marks time. If the underlying business is genuinely good, time is working to repair the overpayment for you; the company grows into the valuation you paid. Second, if you still believe in the business and it's a fair price now (or below), you can average in — buy more at today's lower price, which pulls your overall average cost down and lifts your future earnings yield. Third, and most important, size it: a position that hurts this much is probably too large. The lesson isn't 'never buy quality'; it's 'never let one exciting story become too big a share of your money.'

A fair question hangs over everything you've just learned. If thousands of professional analysts, armed with more data and time than you'll ever have, are already running these exact calculations on every stock every day, then the price on the screen already reflects their collective best guess. The good news and the bad news are baked in. So what, honestly, can a part-time beginner hope to add? Isn't it all priced in?

Mostly, yes — and that is not a defeat, it's the single most useful thing valuation teaches you. The reason a stock's price is so hard to beat is precisely that all this analysis has already happened; the market's price is the melted-together judgement of everyone who's done the work. Which means the honest conclusion of learning valuation is not 'now I can find the hidden bargains the pros missed.' It is: 'now I understand why finding those bargains is so hard, and why paying for growth is a bet even the experts get wrong.' That humility is the point. It is exactly why, for almost every beginner, the wisest move is the one Lesson 23, Why Beginners Index, spells out in full: don't try to out-value the market — buy the whole market, cheaply, through an index fund, and let its ~11–12% long-run growth compound while you get on with your life.

So hold both truths at once, the way Suresh does. The tools are real and worth knowing — he uses them on the slice of his money he picks himself, and they've saved him from many an exciting, overpriced story. And the humility is also real: even he keeps the core of his ₹1.8 crore in broad, low-cost index and diversified funds, because he knows how thin the edge is and how wide the range of 'fair value' really runs. Learn valuation to see clearly and to protect yourself. Then, for most of your money, let the index do the heavy lifting — and let the next lessons build that boring, powerful core.

Here's the whole lesson in one live tool. Type in a share price, its earnings per share, a growth guess and the return you'd demand, and it does everything you've learned at once: the P/E, the earnings yield set against the 6.8% bond, the PEG, and a rough discounted-cash-flow fair-value range with a plain green/amber/red read on the price. It opens pre-filled with Sunvale's exact figures, so you can watch the lesson's numbers appear — then clear it and value anything you like.

Play with it until the big lesson is in your fingers, not just your head: change the growth from 18% to 14% and watch the fair value collapse; drop the discount rate from 12% to 10% and watch it leap. Same company, same earnings — a completely different 'worth', from a nudge to a guess. That instability isn't a bug in the tool; it's the truth about valuation, which is exactly why you demand a margin of safety and, for most of your money, let an index carry the load. When the verdict light glows amber on Sunvale at ₹1,800 — 'above the central estimate, thin margin of safety' — you'll know precisely what it means, and precisely what to do: wait for a better price, or own it through the whole market instead.

The questions beginners ask most about valuation, answered plainly — drawn from the kinds of things people wonder aloud once the P/E stops being a mystery and starts being a tool.

• Is a low P/E always cheap? No — and this is the most expensive myth in investing. A low P/E can mean a genuine bargain, or a value trap: a business that's shrinking, cyclical at its peak, or in trouble, where the low multiple is the market's correct verdict that the earnings won't last. Always ask why the P/E is low before treating it as cheap.
• What's a 'good' P/E? There isn't one — it depends entirely on growth, quality, sector and the cycle. A 45× can be fair for a fast, durable compounder; an 8× can be dangerous for a dying cyclical. The market's ~21× is the going rate for an average large company; treat that as your reference point, not a universal 'fair' number.
• How is the earnings yield like an interest rate? Flip the P/E: earnings ÷ price. A stock on 45× has a 2.2% earnings yield — the profit the business earns behind each rupee you invest, as a percentage, so you can compare it directly to a 6.8% bond or 4% inflation. It turns 'expensive' from a feeling into a number.
• Why would I accept a 2.2% earnings yield when a bond pays 6.8%? Because the bond's 6.8% is frozen forever, while the stock's 2.2% sits on earnings growing ~18% a year — a yield that climbs. You're paying for a growing return instead of a fixed one. Just be sure the growth is real, and demand a margin of safety, because there's no cushion if it disappoints.
• Do I actually need to build a DCF? No. As a beginner you never need to build the spreadsheet. You need the intuition — a business is worth its future cash discounted to today, so growth and the discount rate move value — and the humility the sensitivity grid teaches: the answer is a wide range. That understanding, plus a margin of safety, is enough.
• Why does the same company have two different 'fair values'? Because fair value depends on guesses about an unknowable future — growth and the discount rate — and small, reasonable differences in those guesses produce hugely different answers (Sunvale ran from ₹650 to ₹3,100). Two honest analysts disagreeing isn't a mistake; it's proof that valuation is a range, not a verdict.
• Trailing or forward P/E — which should I use? Know which you're being quoted. Trailing uses last year's real profit (honest but backward-looking); forward uses next year's forecast (relevant but a guess, usually optimistic). If someone quotes a low forward P/E to make a stock sound cheap, ask whose forecast it rests on.
• Is a high dividend yield better than a low one? Not necessarily. A low yield from a high-ROE company reinvesting its profits (like Sunvale keeping ₹28 of every ₹40) can build far more wealth than a high yield from a business with nowhere good to grow. Your total return is the dividend plus the growth; for a compounder, the growth is the whole story.
• Can I use these ratios on a bank or an NBFC? Lean on price-to-book (read beside ROE) more than P/E for banks, lenders and other asset-heavy businesses, where book value is real and comparable. P/E still helps, but P/B is often the sharper tool there — you'll see this properly in the fixed-income and lending lessons.
• If it's all priced in, is valuation pointless for me? The opposite — it's just not a tool for beating the market. Valuation literacy tells you what you own, unmasks scams and value traps instantly, and keeps you calm in manias and crashes. For most of your money the right use of it is to confirm that buying the whole index and staying invested is the sane default.

A quick refresher on the words this lesson introduced — each in one plain line. If any feels shaky, that's the section to reread before moving on.