How to Analyse a Defence Stock (Order Book, Execution)

In the December 2024 quarter, Bharat Electronics' management told analysts it was "confident to achieve this INR25,000 crores order inflow by the end of the financial year" — and had said the same, almost word for word, two quarters earlier (BEL Q1 FY25 concall: "highly confident in achieving its full FY25 order inflow target of INR25,000 crores"). When the year actually closed, the company reported that "the total order intake for FY25 was slightly less at ₹18,000 crores" (BEL Q4 FY25 concall). The guidance was off by ₹7,000 crore — about 28% — and management was neither lying nor careless. Bharat Electronics is one of the best-run companies in the sector. The orders did not vanish; they slipped. (Illustration of how to read the numbers, not a view on the stock.)

That single gap — confident order-inflow guidance, then a number that lands lower because a big contract signed three months late — is the whole sector in one data point. A defence company does not really sell products quarter to quarter. It sells a multi-year backlog to the government, then converts that backlog into revenue at a pace set by a customer it cannot hurry. So the headline profit you see this quarter was decided two or three years ago, and the headline you'll see in three years is being decided in today's order pipeline. Read a defence stock like a manufacturer and you will pay up for a quarter that was always going to be lumpy.

This is how to read a defence company the way someone who has watched the order-to-revenue lag does it: the four or five numbers that actually decide the outcome, the ones buried in investor decks rather than the income statement, and the one red flag that looks like growth and is actually a trap.

A boundary first. From the outside you cannot model a defence order book contract by contract — much of it is classified, and timing sits with the Ministry of Defence, not the company. What you can do is read the direction of the backlog, the execution pace, and the cash, and check whether management's account of the lumpiness survives the next two calls.

What actually drives a defence company's economics?

Picture a tailor with a two-year waiting list. His order book is enormous and his customers won't cancel — but he can only stitch so fast, the cloth (here, foreign sub-systems and engines) arrives on someone else's schedule, and his single client pays in instalments tied to delivery milestones, often slowly. His wealth is real but it is locked in the backlog and the working capital, not in this month's takings. A defence company is that tailor with a balance sheet.

Three features fall out of this. First, the order book is the asset, not the revenue. A defence company with a ₹70,000–₹1,90,000 crore backlog has years of visibility, which is exactly why the market pays a premium — but visibility is not delivery. Second, the single customer is the government, which means demand is policy-driven and bunched (a budget cycle, an election, a border flare-up), not smoothly growing. Third, cash comes late. Milestone-linked payments and government receivables mean a profitable order can sit on the balance sheet as working capital for years before it turns into cash. Miss any of these and the income statement will mislead you.

The metrics that matter — and where they hide

For each one: what it is, why it matters here, where to find it, what good looks like, and a real number.

Order book — the visibility everyone pays for

The order book (or backlog) is the value of signed contracts not yet executed. It is the single most-watched number in the sector because it sets the revenue runway. You will not find it in the income statement — it lives in the investor presentation and the concall opening remarks, and you should read it as a multiple of annual revenue, not a raw figure.

Bharat Electronics carried an order book of roughly ₹74,859 crore as of 1 July 2025 (BEL Q1 FY26 concall) against FY25 revenue of about ₹23,770 crore (Inve data, FY25, summing the four quarters) — roughly 3.1 years of work in hand. HAL is even longer-dated: its order book stood at about ₹94,000 crore in March 2024 and jumped to ₹1,89,300 crore by March 2025 (HAL Q4 FY25 concall) — against FY26 revenue of about ₹33,089 crore (Inve data, FY26), that backlog is more than five years of revenue. Long visibility is the sector's defining comfort. The trap is treating it as the same thing as growth.

Order inflow and book-to-bill — is the backlog being refilled?

Order inflow is new orders won in the period. Book-to-bill is inflow divided by revenue: above 1.0 means the backlog is growing faster than it's being consumed; below 1.0 means it's shrinking. This is the leading indicator the order book itself hides — a company can report a fat backlog for two years while quietly failing to win new work, and the order book only starts falling once execution eats into it.

Here is where the lumpiness bites. BEL's order inflow was about ₹18,000 crore in FY25 (BEL Q4 FY25 concall) on revenue of ~₹23,770 crore (Inve data, FY25) — a book-to-bill of roughly 0.76, below one, the year the backlog actually dipped (order book fell from ₹74,595 crore on 1 Oct 2024 to ₹71,650 crore on 1 April 2025, per BEL concalls). HAL's FY25 told the opposite story: new manufacturing order inflow of about ₹1,02,337 crore (HAL Q4 FY25 concall) against ~₹30,981 crore revenue (Inve data, FY25) — a book-to-bill above three. One year, two companies, the difference between a backlog draining and a backlog tripling. Never read a single year's inflow as a trend — read three or four, because one delayed mega-contract distorts everything.

Revenue conversion (execution) — the number management can't fully control

Execution is how fast the backlog turns into revenue. It is the gap between winning an order and recognising it, and in defence that gap is governed by capacity, by certification, and by foreign sub-system supply you don't own. This is the metric most retail investors skip, and it's the one that decides whether a fat order book becomes earnings or just a story.

The cleanest illustration sits in HAL's LCA Tejas programme. At the FY23 calls management guided to "1 fighter and 2 trainer, total 3 aircraft during the current financial year… 16 numbers starting from '24-'25 onwards" (HAL Q4 FY23 concall). By the FY25 call the same 24-aircraft capacity had moved out: "The 24 aircraft will happen in '26-'27" (HAL Q4 FY25 concall), with the constraint named plainly — "GE has promised a supply of 12 engines in this calendar year up to December" (HAL Q4 FY25 concall). The order was never in doubt; the delivery was, because the engines come from a single foreign supplier. In Inve's record this delivery guidance is marked revised down (Inve data) — and that, not the backlog, is what moved the near-term numbers.

Indigenisation % — the margin and the moat in one number

Indigenisation is the share of a product made in India rather than imported. It matters twice over: a higher domestic-content share means fatter margins (you capture the value instead of paying a foreign OEM) and a wider moat (the government's "Make in India" policy steers nomination orders to high-indigenisation players). It's an investor-deck and annual-report number, not a financial-statement line.

Bharat Electronics runs at "70–73%" average indigenisation (BEL Q3 FY26 concall), which is a large part of why it sustains a ~30% operating margin (Inve data, Q3 FY26: OPM 30%) — when you build the box rather than assemble someone else's, the margin is yours. A company importing most of its content and bolting it together will show thinner, more volatile margins and far less pricing power, however large its order book reads.

Operating margin — and why defence margins are unusually sticky

OPM here tells you who owns the value chain. Pure-play system designers and integrators print high, stable margins; assemblers and shipbuilders print lower, lumpier ones. Bharat Electronics has held operating margin in the 22–31% band through FY25–FY26 (Inve data: 22% Q1 FY25, 31% Q4 FY25, 30% Q3 FY26), and Data Patterns — a small-cap radar and electronic-warfare designer — runs higher still, 47% in Q3 FY26 and 56% in Q4 FY26 (Inve data), because its revenue is design-and-IP rather than metal-bending. The lesson: a 50% margin and a 20% margin in the same "defence" bucket are different businesses. Read margin together with indigenisation and you'll know which one you're holding. The listed space spans the whole spread — missile-maker Bharat Dynamics, shipbuilder Garden Reach Shipbuilders and simulation specialist Zen Technologies each sit at a different point on the value chain, so the same lens has to be re-weighted for each.

Working-capital cycle and government receivables — where the cash gets stuck

This is the sector's quiet tax. Because the customer is the government and payments are milestone-linked, defence companies tie up enormous working capital, and a profitable P&L can sit on a cash-poor balance sheet for years. The number to find is working-capital days (or the cash-conversion cycle), and you find it in the concall, not the headline.

Data Patterns guided to 270 working-capital days but was running at about 340 days in Q3 FY26 (Inve data) — that guidance is marked delayed in Inve's record (Inve data). HAL's inventory days climbed from 159 in March 2024 to 263 in March 2025 (HAL concalls) as it built ahead of deliveries it couldn't yet recognise — inventory is the locked-up backlog, waiting on engines. The discipline: a defence company that grows revenue while its working-capital days keep stretching is funding its own order book, and that cash has to come from somewhere — debt, dilution, or a squeezed dividend.

How should you value a defence stock?

The right anchor is P/E on multi-year order-book visibility, with the lumpiness explicitly normalised — not P/E on a single year's earnings. Because the backlog gives genuine multi-year revenue sight, the market is willing to pay a forward multiple that capitalises several years of growth. The mistake is paying that multiple off a peak quarter or a year that happened to catch a mega-order.

The honest way to do it: take the order book as a multiple of revenue (BEL ~3.1x, HAL ~5x+ above), form a view on the execution pace that converts it (the engine-supply constraint at HAL is the whole game), and only then ask what multiple the visibility deserves. A stock trading at 40–50x trailing earnings on a backlog that's actually draining (book-to-bill below one, like BEL in FY25) is priced for an inflow recovery that hasn't shown up yet — that's a bet on the next order cycle, which is a policy and timing call, not a fundamentals call. Two-thirds of the analytical work in this sector is separating "the orders are coming, just late" from "the orders aren't coming."

A worked case: said vs did at Bharat Electronics

Walk the order-inflow guidance through one year and the sector's character shows up cleanly (illustration, not a view on the stock; figures from Inve data and BEL concalls):

Read the three rows together. Through two calls management held a confident, round-number guide — ₹25,000 crore — while the year-to-date inflow at Q3 sat near ₹11,000 crore, meaning the entire ₹14,000 crore balance depended on a clutch of large contracts (NGC, QRSAM) signing in the final quarter. They didn't sign in time. The actual came in at ₹18,000 crore, the order book ticked down over the year, and in Inve's Promise Tracker that FY25 order-inflow guidance is marked missed (Inve data). Notice what this is and isn't. It is not evidence of weak management — BEL still grew revenue ~17% and held a ~30% margin (Inve data, FY25). It is evidence that confident inflow guidance in this sector is a forecast about a government's signing calendar, and deserves less weight than the book-to-bill trend underneath it. A single big contract slipping one quarter the wrong side of 31 March turns "achieved" into "missed" — which is exactly why a sequence of guidance tells you more than any one call. This is the pattern Inve's Promise Tracker is built to surface: each forward commitment pinned to the quarter it was made, with a verdict as later calls come in.

Red flags specific to defence

• A fat order book that hasn't grown in years. Visibility without inflow is a backlog being eaten. Watch book-to-bill, not the raw backlog — BEL's dip below 1.0 in FY25 is the early version of this.
• Execution guidance that keeps sliding right. When delivery dates move out call after call — "16 from FY25" becoming "24 in FY27" — the order book is real but the earnings are further away than the multiple assumes.
• Working-capital days stretching while revenue grows. Profit on paper, cash stuck in receivables and inventory. If management can't give you a clean working-capital number, that is the answer.
• Indigenisation stalling. A company that can't lift domestic content stays a margin-taker dependent on foreign supply — and on the foreign supplier's schedule, as the HAL engine constraint shows.
• Round-number order-inflow guidance with no contract-level basis. "₹25,000 crore this year" with the bulk loaded into Q4 is a hope about a signing calendar. Ask what's already concluded versus what's "expected."

A note on incentives before you trust the optimism: a defence-PSU management is steered by the Ministry of Defence and rewarded on order wins, so its instinct is to guide to the full pipeline and trust the contracts to land on time. Don't ask the barber whether you need a haircut — read the book-to-bill yourself.

Frequently asked questions

Where this lens can be wrong. The strongest case against everything above is that the lumpiness is the opportunity. A reader who sells every time order inflow misses by a quarter will keep selling the best companies right before the delayed mega-contract lands — BEL's "missed" FY25 inflow was followed by guidance for ₹27,000 crore-plus in FY26 (Inve data), the very recovery a backlog-draining reading would have feared. The honest claim is narrower than it looks: book-to-bill and execution pace tell you whether the visibility is being refilled and converted, not when the next order cycle turns. They lower your odds of overpaying for a peak quarter; they cannot time a policy-driven customer. And we haven't modelled the geopolitics — a border event or an export breakthrough can rewrite the order pipeline faster than any concall would warn you.

Inve's KPI Screener lines up order book, inflow, margin and working-capital days across defence companies — value, trend, and a data-confidence flag per number — so the buried metrics take minutes, not an afternoon of deck-reading. For the spread-and-credit cousin of this discipline, see how to analyse an NBFC.

The discipline comes down to refusing to be impressed by the backlog alone. The order book is the asset, but it speaks through book-to-bill, execution pace, indigenisation, and the cash trapped in working capital — and any one of those can be quietly turning while the headline backlog still reads large. So invert the question you bring to a defence company's results. Don't ask "how big is the order book?" Ask: if this company were winning fewer new orders and converting the old ones slower while the backlog still looked impressive, what would the numbers show — and does this book rule that out? A flat backlog with book-to-bill below one and stretching working capital does not rule it out; it is the pattern itself.

And the owner's question to sit with before buying any defence stock: what must I believe about the next order cycle and the company's ability to execute it — not this quarter's profit — for this backlog to actually become cash in my pocket five years out? If the honest answer leans on a round-number inflow guide rather than the conversion record underneath it, you've read the headline, not the business.

Related reading

• How to Analyse a Bank Stock (NIM, CASA, GNPA, ROA)
• How to Analyse a Building Products Company (Pipes, Tiles)
• How to Analyse a Cables and Wires Stock