How to Analyse an EMS Electronics Manufacturing Company

In the December 2025 quarter, Dixon Technologies did ₹9,750 crore of revenue from its Mobile and EMS business and ₹1,050 crore of operating profit on it (Dixon Q3 FY26 concall, Mobile and EMS revenue ₹9,750 crore, operating profit ₹1,050 crore). That looks like a 10.8% operating margin — healthy. But on the same call the managing director told an analyst what the mobile business actually earns once you strip out the government subsidy: "We still feel that we should be able to have a margin of somewhere between 2.8% to 3.1%, 3.2% in our mobile phone business… here, I'm talking about the scenario without PLI" (Dixon Q3 FY26 concall, Atul Lall). (Illustration of how to read the numbers, not a view on the stock.)

Sit with that gap. The core manufacturing economics of one of India's largest contract manufacturers are a roughly 3% margin on the box it builds; the difference between that and the reported number is, in large part, a Production-Linked Incentive cheque that has an expiry date. An EMS company is not a "tech" business and it is not, mostly, a margin business — it is a logistics-and-capital business that turns other people's brands into finished goods for a thin, contractual slice of the value. Almost everything that decides whether the stock works flows from that one fact: the margin is thin, so the only ways to make real money are to grow volume relentlessly, climb toward design (ODM), or own more of the component stack — and to do all three without drowning in working capital.

This is how to read an EMS company the way someone who has actually modelled one does: the order book and the pipeline behind it, the revenue mix that decides the margin, the gross margin and raw-material pass-through, capacity and utilisation, the ODM-versus-EMS distinction that the whole thesis rides on, and the working capital that quietly eats the cash that growth is supposed to make.

A boundary first: from the outside you cannot audit a contract manufacturer's customer concentration or the true terms of its pass-through clauses. What you can do is read the direction of these numbers and check whether the order book is converting into cash, not just revenue.

Why is an EMS company a thin-margin volume business?

Strip an EMS company down and it is a job-shop at industrial scale. A brand — Samsung, a smart-meter utility, an auto OEM — hands over a design and a bill of materials, and the manufacturer buys the components, assembles them, tests them, and ships finished product. The brand owns the customer and the pricing power; the manufacturer owns a factory and a procurement desk. For that, it earns a conversion margin measured in low single digits at the pure-assembly end.

That framing fixes two beginner errors. First, a high reported margin is suspicious until you know what's in it. Indian EMS margins through FY24–FY26 are inflated by PLI incentives — cash the government pays for hitting localisation and output targets, which tapers and ends. Dixon's own management put the underlying mobile margin "without PLI" at 2.8–3.2% (Dixon Q3 FY26 concall) against a segment operating margin that screens far higher. Second, raw material is most of the cost, so revenue is a poor proxy for value added. In a phone, the chipset and memory can be 70–80% of the bill; the manufacturer's economics live in the thin layer on top. When memory prices spiked through late 2025 — Dixon flagged that memory "has moved from being a relatively small line item to one of the most sensitive parts of the bill of material" (Dixon Q3 FY26 concall) — revenue rises mechanically while the value-added margin gets squeezed. So the question is never "how fast is revenue growing?" It is "how fast is value-added growing, who pays when input costs jump, and is the cash following the revenue?"

The order book and pipeline: real demand, or a timing story?

The order book is where an EMS company's future is written, because revenue is booked against contracts won quarters earlier. Kaynes Technology reported a diversified order book of about ₹9,000 crore at the end of Q3 FY26 (Kaynes Q3 FY26 concall; Inve data, ₹9,000 cr) — roughly two and a half times its FY26 revenue, which tells you demand visibility is genuinely deep. Syrma SGS carried an order book of ₹6,400 crore in December 2025, up from ₹5,800 crore in September and ₹5,450 crore in June (Inve data, quarterly_kpi: Order Book ₹6,400 cr Dec-2025, ₹5,800 cr Sep-2025, ₹5,450 cr Jun-2025) — a book growing steadily quarter on quarter, which is the texture you want.

Here is the catch that separates a careful reader from a careless one: an order book is a measure of demand, not a delivery schedule. EMS orders get rescheduled, and the company often does not control the timing. The cleanest illustration is Kaynes's own FY26. In the June 2025 call management said "we maintain our… guidance of INR4,500 crores plus on a consol basis" (Kaynes Q1 FY26 concall); by the December call the FY26 revenue target had been brought down to about ₹4,100 crore, from the ~₹4,400 crore implied earlier (Kaynes Q3 FY26 concall). FY26 actually landed at about ₹3,626 crore (Inve data, sum of Q1–Q4 FY26: ₹673 + ₹906 + ₹804 + ₹1,243 crore). The order book was not fake — but a book that converts slower than guided turns a "60% growth" year into a 33% one (FY26 ₹3,626 crore vs FY25 ₹2,721 crore, Inve data). So read the order book two ways: its size relative to revenue (visibility) and its conversion — does book-to-revenue hold up across four quarters, or does the book keep growing while revenue keeps slipping a quarter to the right? Where you find it: the headline number is usually in the investor presentation or the concall opening remarks, not the financial statements. Inve's KPI Screener lifts it out of the deck so you don't have to.

Revenue mix: the metric that decides the margin

Two EMS companies can do identical revenue and earn double the margin one over the other, entirely because of what they make. Mix is the margin. Consumer electronics (phones, TVs) is the highest-volume, lowest-margin end — fierce competition, commoditised assembly. Industrial, medical, aerospace and defence sit at the other end: lower volume, longer qualification cycles, stickier customers, fatter margins.

Watch what the mix does to the same income statement. Kaynes, weighted toward industrial, automotive, aerospace and defence, ran a ~16% operating margin in Q4 FY26 (Inve data, opm 16%, Q4 FY26). Dixon, dominated by consumer mobile and appliances, sat at a 4% operating margin in the same quarter (Inve data, opm 4%, Q4 FY26). Same sector, four-times-different margin — and the reason is mix, not management quality. Syrma sits in between and is openly engineering its mix: it guided to bring its consumer segment "down to 30% of total revenue" (Syrma Q4 FY25 concall) and grow industrial, medical and automotive, where in Q3 FY26 the auto segment grew 30% YoY, industrial 29% and med-tech 31% (Inve data, quarterly_kpi, Q3 FY26). Where to find it: the segment split lives in the investor PPT and the concall, almost never in the standalone P&L. Read it as a trend — a company drifting toward consumer to chase volume is trading margin for revenue, and the screen-level margin will tell you a quarter or two later.

Gross margin and raw-material pass-through

Below the operating line, the number that reveals the real value-add is gross (or material) margin — revenue minus the cost of components. Syrma reported a gross material margin of 24% in H1 FY26 (Inve data, quarterly_kpi: H1 FY26 Gross Margin 24.3%; Q2 FY26 23.8%), up sharply from a year earlier — management noted it "rising to 24% from 15% YoY" (Syrma Q4 FY25 concall). That jump is the whole game in miniature: a richer mix and more design content lifted the slice the company keeps off the same bill of materials.

The second thing to check is who eats raw-material inflation. Good EMS contracts pass component-price changes through to the customer, usually with a lag of a quarter. The lag matters: when memory and chipset prices spiked in late 2025, a manufacturer on pass-through terms sees margin dip for a quarter, then recover; one absorbing the cost sees it stay down. Listen on the concall for whether management describes a contractual pass-through or talks vaguely about "managing input costs" — the difference is the difference between a logistics business and a speculator on commodity prices. Here is the homely version: an EMS firm should run like a toll booth, taking its fixed cut whatever the price of the cars passing through; the moment it starts betting on the price of the cars, it has stopped being a toll booth.

Capacity, utilisation and capex: the engine and its fuel bill

Because the margin is thin, volume is the only lever — and volume needs plants. So an EMS company is permanently spending on capacity, and the question is whether that capex earns a return before the next round arrives. Two numbers tell you: utilisation (are existing lines full?) and the capex-to-order-book ratio (is the spend matched to demand?). Syrma ran total gross capacity utilisation of about 65% in FY25 and guided FY26 to ~67.5% (Inve data, quarterly_kpi, FY25/FY26) — meaning real headroom to grow revenue without fresh capex, which is exactly when EMS economics look best.

The trap is the opposite case: heavy capex into capacity that order-book conversion isn't yet filling. Dixon is building a 1-million-square-foot Noida facility plus display-module and component plants (Dixon Q3 FY26 concall), and Kaynes spent through FY25–FY26 on an OSAT semiconductor plant and a PCB facility — the latter guided "ready November/December 2025" and then marked delayed (Inve data, Promise Tracker, birth quarter Q4 FY25, PCB Facility Readiness — delayed). Capex slipping is not fatal; capex slipping while the revenue it was meant to serve also slips is the combination that turns a growth story into a cash sink. Track guided capacity-online dates against what management says next quarter — the concall record is where the slippage shows.

ODM versus EMS: the metric the whole thesis rides on

This is the distinction that separates a contract assembler from a value creator. EMS (pure assembly) builds to the customer's design — lowest margin, lowest switching cost. ODM (original design manufacturing) means the company designs the product itself and the customer just brands it — higher margin, far stickier, because the customer can't easily move a design it doesn't own. The single most important forward number in an EMS thesis is the trajectory of ODM share, because that is the company climbing the value chain to escape the 3% trap.

The numbers show how early this journey is. Kaynes reported ODM at only ~5% of its order book in Q3 FY26 against a "Pure EMS Business ODM Target" of 40% (Inve data, quarterly_kpi: ODM Share in Order Book 5%, Pure EMS ODM Target 40%, Q3 FY26) — a long road, and a target to hold management to. Syrma's ODM was 16% of revenue in Q3 FY26, up from 12% the prior quarter (Inve data, quarterly_kpi, Q3 FY26 / Q1 FY26). Kaynes's older investor deck shows ODM and product-engineering solutions already a large share of value even when small in order count (Kaynes Q3 FY25 investor presentation, ODM & Prod. Eng. and IoT 56%) — a reminder that ODM revenue carries more margin per rupee than EMS revenue. Where to find it: ODM share is almost always a PPT or concall disclosure, never a line item. A rising ODM share with rising gross margin is the cleanest proof an EMS company is becoming a better business; a flat ODM share with rising revenue is just more of the same thin work.

Working capital: where the cash hides — or disappears

Here is the metric that quietly decides whether all that growth ever becomes cash for an owner. EMS is working-capital-intensive: you buy components, build, hold inventory, ship, and wait to be paid — and growth consumes cash before it produces it, because every new order needs inventory funded up front. The number to watch is net working capital days, and the spread across the sector is enormous. Syrma ran net working capital days of about 68 in December 2025 (Inve data, quarterly_kpi: Net Working Capital Days ex-Elcome 68, Q3 FY26). Kaynes ran about 85 days on the same basis (Inve data, quarterly_kpi: Net Working Capital Days 85, Q3 FY26). Dixon ran a negative working capital cycle of around −7 to −8 days (Inve data, quarterly_kpi: Working capital cycle −7 days Q3 FY26, −8 days FY26).

That negative number is the prize and the explanation for Dixon's scale-on-thin-margins model: management said its "negative working capital cycle… gives us sufficient headroom to invest in capacity, components and new categories" (Dixon Q3 FY26 concall). When suppliers and customers effectively fund your inventory, a 3% margin can still throw off cash and earn a high return on capital. The inverse is the killer: a fast-growing EMS company with working capital days expanding is funding its own growth from the balance sheet, and the reported profit will not show up as cash. Syrma itself guided to bring working capital "below 65 days for the full year" and that target was marked delayed (Inve data, Promise Tracker, birth quarter Q4 FY25 — delayed) — not alarming on its own, but exactly the kind of guidance worth holding management to, because in this sector the gap between profit and cash is the working-capital line.

How to read valuation for an EMS company

EMS stocks trade on a price-to-earnings multiple, and the multiple is usually high because the market is paying for years of guided volume growth and the order book behind it. That is defensible — but only if you adjust for two things the headline P/E hides.

First, strip PLI from the earnings you're capitalising. A material chunk of current EMS profit is incentive income that tapers and ends. A P/E that looks reasonable on reported earnings can look very different on earnings ex-PLI — and the right question is what the multiple is on sustainable operating profit at the underlying 3–5% margin, not on the subsidised number. Second, a thin-margin business has violent operating leverage, so earnings are far more sensitive to a revenue miss than the margin suggests: at a 4% margin, a 10% revenue shortfall can wipe out a quarter of profit. Kaynes is the live example — when FY26 revenue slipped from a ₹4,500-crore-plus guide to ₹3,626 crore actual, profit growth came in well short of what a "60% growth" year had implied, and a stock priced for the guidance had to re-rate to the delivery. The honest way to value an EMS company is to ask what you're paying per rupee of order book that actually converts to cash — high P/E plus a fast-converting book plus negative working capital is a different animal from high P/E plus a book that keeps sliding right.

A worked case: Kaynes, said versus did

Take Kaynes through FY26 with the numbers and one management exchange. In June 2025 management reaffirmed FY26 revenue "guidance of INR4,500 crores plus" (Kaynes Q1 FY26 concall). By December the target had been brought down to about ₹4,100 crore, from the ~₹4,400 crore implied earlier (Kaynes Q3 FY26 concall). The full year landed at about ₹3,626 crore (Inve data, sum of FY26 quarters). On that December call an analyst put the pattern to management directly: "last year also, we have guided INR3,000 crores of revenue and end up INR2,700 crores. And in the beginning of the year, we guided at INR4,400 crores, now we are targeting INR4,100 crores… what are the risks?" (Kaynes Q3 FY26 concall, analyst Manish Ostwal).

Management's answer is the most important thing in this article. The CFO explained that the company plans orders "anywhere between 6 months to 5 years time frame," that "the orders get rescheduled. And on that, we don't have too much of control," and concluded "the guidance for quarterly number is not the appropriate way" to judge the business (Kaynes Q3 FY26 concall, Jairam Sampath). Read it without cynicism: this is almost certainly true, and Kaynes is a genuinely strong, well-regarded manufacturer with a deep ₹9,000-crore order book. But it is also the exact reason an EMS order book deserves a discount to its face value — the demand is real and the timing is not the company's to promise. The lesson is not "Kaynes missed." It is that in EMS, the order book tells you the direction and the ceiling, never the schedule — and an investor who capitalised the ₹4,500-crore guide at a growth multiple paid for a delivery the company itself couldn't control. (Illustration, not a view on the stock — and a read on how guidance converted through one year, not a lifetime verdict.)

This is the pattern Inve's Promise Tracker is built to surface: each forward guide pinned to the quarter and quote it was made in, then marked as later calls confirm, revise, or quietly drop it — so you see the sequence of an EMS book converting (or not) rather than re-reading every transcript by hand.

Red flags specific to EMS

• Reported margin propped up by PLI, with thin or falling margin underneath. When a manufacturer puts its own ex-incentive margin at ~3% (Dixon Q3 FY26 concall) but the screen shows double that, you are capitalising a subsidy with an end date. Always ask for the number without PLI.
• Order book growing while revenue keeps slipping a quarter right. A rising book is good; a book that grows as guidance gets cut (Kaynes FY26) means conversion, not demand, is the bottleneck — and conversion is what pays you.
• Working-capital days expanding through a growth phase. Profit that doesn't become cash is the EMS death by a thousand cuts. Negative or stable WC days during growth is a quality signal; ballooning WC days is the balance sheet funding the income statement.
• Customer concentration you can't see. A book dominated by one anchor brand can vanish faster than a plant can be repurposed. Management that won't discuss concentration is telling you where the fragility lives.
• Capex outrunning order-book conversion. Capacity built ahead of demand that doesn't arrive on schedule turns operating leverage upside-down — fixed costs against revenue that slipped.

Where this lens can be wrong. The strongest case against everything above is Dixon itself. By the reasoning here — thinnest margins, biggest consumer mix, most PLI dependence — Dixon should be the least attractive of the three. Yet its negative working capital cycle means a 3% margin compounds capital faster than a 16%-margin peer that has to fund its own inventory, and its sheer scale and backward-integration push (camera modules, display modules, components) is precisely the climb up the value chain this article praises elsewhere. A reader who mechanically prefers the high-margin name over the thin-margin one can pick exactly wrong, because in a working-capital business return on capital, not margin, is the scoreboard — and the thinnest-margin operator can earn the highest return on capital. The margin is the most visible number and the most misleading one taken alone. (Illustration, not a view on any of these stocks.)

A hard limit worth restating: from a transcript and a PPT you cannot verify customer concentration, the true terms of pass-through clauses, or whether a "won" order is firm or a letter of intent. EMS analysis lowers your odds of overpaying for a timing story; it does not let you audit the order book.

A repeatable workflow

1. Decompose the order book. Size relative to revenue (visibility) and conversion across four quarters (does it become revenue on schedule, or slide right?).
2. Read the mix. Consumer vs industrial/medical/auto/defence — the split that decides the margin, found in the PPT, not the P&L.
3. Strip the margin. Gross/material margin for real value-add, and the operating margin without PLI for the sustainable number.
4. Check the engine. Utilisation and capex against the order book — headroom is good, capex outrunning conversion is the risk.
5. Watch ODM share. A rising ODM share with rising gross margin is the company escaping the 3% trap; a flat share with rising revenue is just more thin work.
6. Follow the cash. Net working capital days — the line where EMS profit either becomes cash or quietly doesn't.

Frequently asked questions

If you've followed the spine — order book, mix, gross margin, capacity, ODM, working capital — you'll notice they all collapse into one question. Invert the usual one. Don't ask "is this EMS company growing fast?" Ask: if this manufacturer were quietly buying growth it can't convert to cash — over-promising on an order book it can't schedule, funding inventory from its own balance sheet, leaning on a PLI cheque that's about to expire — what would the numbers look like, and do these rule it out? A book that grows while guidance is cut, working-capital days that climb through the growth, and a margin that collapses without the subsidy do not rule it out; they are the pattern itself.

And the owner's question to sit with before you buy a share of any contract manufacturer: five years out, when the PLI has tapered and the easy consumer volume has commoditised further, what must I believe — about ODM share, about owning more of the component stack, about converting that order book to cash — for this to still be a business that earns its capital back? If the answer leans on this year's subsidised margin rather than the slow climb up the value chain, you've priced the headline, not the business. Compare this with the very different engine inside a lender in how to analyse an NBFC — there the balance sheet is the product; here it's the fuel tank.

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