Zen Technologies — a defence champion waiting on a slow government order cycle

The Short Version

Zen Technologies makes two things the Indian military increasingly needs: realistic combat training simulators (so soldiers can train without firing live rounds) and anti-drone systems (to detect and shoot down hostile drones). It’s the largest Indian player in both, builds its technology in-house, and earns unusually fat margins (~36%) with almost no debt. After a spectacular run — revenue more than quadrupled from ₹219 crore to ₹974 crore between FY23 and FY25 — the year just ended (FY26) was a down year: revenue fell 29% to ₹688 crore and profit fell 27%. The reason is the whole story in miniature: after the India-Pakistan flare-up (Operation Sindoor) in mid-2025, the government rushed emergency anti-drone purchases and put regular simulator orders on hold, so a big expected order kept slipping. Tellingly, Zen’s order book nearly doubled through the year even as revenue fell — the orders are coming, just landing later than promised. Management twice walked down its multi-year target (from ₹6,000 crore to ₹4,000 crore), and the stock still trades at ~85 times earnings on the promise of an FY27–28 rebound.

What This Company Actually Does

Training a soldier on real equipment is expensive and dangerous; a simulator lets them practise driving a tank, firing a weapon or flying a mission in a safe, repeatable, software-driven environment. Zen, founded in 1996 in Hyderabad, is India’s largest maker of these combat training simulators, having shipped 1,000-plus systems. Its second leg, growing fast, is counter-drone systems — the radars, jammers and increasingly “hard-kill” guns that detect and neutralise hostile drones, a threat that has exploded in modern conflict (Ukraine, the Israel-Iran war, India’s own border skirmishes).

Two things define how Zen makes money. First, it sells almost entirely to the government — the armed forces, police and paramilitary — which makes its revenue lumpy and order-driven: a big contract delivered in one quarter can make that quarter look spectacular and the next look weak. Second, it competes on indigenous intellectual property — under India’s “IDDM” (Indigenously Designed, Developed and Manufactured) procurement rules, a company that owns its hardware and software (down to the source code and circuit-board files) gets preferential single-vendor treatment that foreign-dependent rivals can’t match. In training simulators, Zen describes itself as effectively “the last man standing” — no real competitor on certain army tenders. That’s the moat.

It earns ~36% operating margins (very high for a manufacturer), is essentially debt-free, and is sitting on ~₹1,000 crore of cash from a FY25 share sale. The cautions on the ledger: the founders have steadily sold down their stake (from 57% to ~48.5% over three years), working capital has ballooned, a rising share of profit now comes from non-operating “other income,” and the stock is richly priced at ~85× earnings and ~8.7× book.

The Long Game

Zen sits in front of two genuine, multi-year demand waves. The first is the structural shift to simulation-based training — backed by a 2021 government Simulation Policy, endorsed by the defence establishment as a way to train faster and cheaper, and (in management’s telling) only ~7-8% penetrated so far. The second is the drone/counter-drone arms race, which management says intensified “minimum 2x” after the Israel-Iran war — a cat-and-mouse escalation (jamming → spoofing → AI-guided drones → the need to physically shoot them down) that keeps generating new product cycles.

Around these, Zen is building optionality through acquisitions and new products: naval and air-force simulators (via the ARI and Anawave buys), “hard-kill” weapon stations and smart ammunition (AI Turing), an interceptor/loitering drone called HyperStrike (pitched at ~$10,000 versus the ~$30,000 Iranian Shahed, production from FY27, which management calls potentially “the biggest product in our portfolio”), unmanned ground vehicles, and a cybersecurity suite. The mix is shifting from simulator-heavy toward roughly 50:50 simulators-and-everything-else, and exports (~7% today) are targeted at 20–30% by FY28.

But the honest core of the long game is a tension between a strong franchise and a frustrating customer. Management entered this window with a ₹6,000 crore three-year (FY26–28) revenue target, cut it to ₹4,000 crore (FY27+FY28 combined) by the third quarter, and reaffirmed that lower number at year-end — explicitly blaming the slow pace of government procurement. The demand is real and the order book is building toward it; the timing is entirely in the government’s hands, and that has repeatedly defeated Zen’s own forecasts.

The Story So Far

The thread through FY26: every quarter management called the year “muted,” kept pushing a big simulator order further out, watched anti-drone demand surge post-Sindoor — and watched the order book climb even as revenue fell.

Q1 FY26 (reported August) — “disappointing”

Management opened by calling the quarter “kind of disappointing.” About ₹60–70 crore of revenue slipped out of the quarter on last-minute design changes (a deferral, not a lost order). The order book stood at ~₹754 crore. Crucially, a ~₹650 crore army simulator order was expected “by end of September” — Zen being the only qualified vendor — and FY26 was framed as muted with FY27–28 to be “very, very exciting.” The ₹6,000 crore three-year target was reaffirmed.

“This year, we expect it to be muted, but the next two years… are going to be very exciting.” — Ashok Atluri, CMD (Q1)

Q2 FY26 (reported October) — the order slips

A weak quarter again. The ₹650 crore simulator order didn’t come — pushed back because the government was absorbed in post-Sindoor emergency anti-drone buying. Management insisted it would land in H2 and that “2027 and 2028 are going to be great years,” while conceding it had been “very wrong about the slowness with which things would happen.” Anti-drone demand, by contrast, was strengthening fast.

“We were actually very sure it will come by September 30th. But we were very wrong about the slowness… now, they will definitely come in.” — Ashok Atluri (Q2)

Q3 FY26 (reported February) — the target is cut

Revenue ticked up and the order book jumped to ₹1,427 crore (from ~₹675 crore two quarters earlier) — real evidence orders were finally flowing. But management formally scaled the ambition down: from ₹6,000 crore to about ₹4,000 crore of execution across FY27+FY28, blaming the slow procurement pace. Anti-drone, including “emergency-basis” hard-kill demand, was now leading the pipeline.

“We are actually scaling down our targets at this point in time… we should be executing [about ₹4,000 crore] between the two years.” — Ashok Atluri (Q3)

Q4 / full-year FY26 (reported May) — a down year, a building book

The full year landed as guided-muted: revenue ₹688 crore (down 29%), profit ₹218 crore (down 27%), margins easing to ~36%, and the March quarter showing the weakest operating margin in the window (~29%) on operating deleverage. But the order book stood at ₹1,336 crore — roughly ₹1,000 crore of it product to be delivered in FY27 (mostly Q2–Q3). Management reaffirmed the ₹4,000 crore FY27+FY28 target, the 35% EBITDA / 25% PAT margin guidance, and pointed to anti-drone as “the mainstay two to three years from now.”

“We stick to our projection of ₹4,000 crore turnover in FY2027 and 2028 put together.” — Ashok Atluri (Q4)

The pattern a long-term investor should read: the franchise is intact and arguably strengthening — order book nearly doubled, anti-drone demand structurally up, new products (HyperStrike, smart ammunition, naval) maturing, margins still rich. What broke in FY26 was timing, not the business: a government that pivoted to emergencies and slow-walked regular orders. The uncomfortable corollary is that management’s forecasts have a poor recent hit-rate — the big order kept slipping and the three-year target was cut by a third — so the FY27–28 rebound, while supported by a real order book, rests on a procurement cadence Zen doesn’t control.

Where Things Stand

Zen enters FY27 with the order book it spent FY26 waiting for — ~₹1,000 crore of product to deliver, mostly in the second and third quarters — and two demand waves (simulation-based training, counter-drone warfare) that are structurally expanding. It’s debt-free, cash-rich, high-margin, founder-run, and differentiated on indigenous IP in niches where it often faces no real competitor. Management expects FY27 to be the highest-revenue year ever, building toward the ₹4,000 crore two-year target, with anti-drone and new products (HyperStrike interceptor, smart ammunition, naval simulators) as the next legs.

The countervailing reality is equally clear and worth holding without flinching. FY26 was a genuine down year. The multi-year target has already been cut once (₹6,000 → ₹4,000 crore) and the marquee order slipped repeatedly — so management’s timing credibility is dented, and the whole rebound hinges on a government order cycle that has been frustratingly slow. Working capital has stretched alarmingly (the broad screener metric shows working-capital days ballooning past 500, though management tracks a tighter ~196 and guides to 140–150 once execution-driven inventory normalises). A growing slice of profit is non-operating other income. Promoters keep trimming their stake. And the stock trades at ~85× earnings and ~8.7× book — a valuation that already prices in the FY27–28 surge actually arriving on schedule. For a patient investor, the question isn’t whether Zen has a great franchise — it does — but whether the order book in hand converts on the timeline the price assumes, after a year in which the timeline kept slipping.

The Four Checks

1. Quality and moat. A genuinely good business with a real but niche-bound moat. The moat is indigenous intellectual property meeting a procurement rule: under India’s IDDM framework, a vendor that owns its hardware and software down to source code gets preferential — sometimes single-vendor — treatment, and Zen is the largest Indian player in both combat simulators and anti-drone systems, “the last man standing” on certain army tenders. That is closer to a regulatory licence than to mere execution, and it has held through a thousand shipped systems and a clean run in Operation Sindoor. What could erode it: a single customer (the government) controls all the timing, the counter-drone field is a fast-moving arms race where today’s frequency-dominance edge invites tomorrow’s challenger, and a defence boom attracts entrants. Durable in simulators, contested in drones.

2. Returns on incremental capital and runway. The operating engine earns a lot; the balance sheet currently doesn’t. ROCE ran 23%, 46% and 37% across FY23–25 while revenue quadrupled to ₹974 crore — that is what a rupee deployed in the actual business returns. The headline then fell to 16.2% (ROE 10.7%) in FY26, but mostly because a ~₹1,000 crore equity raise left ₹1,308 crore of cash earning deposit rates while revenue dipped 29%. The runway is real: simulation training is per management only ~7–8% penetrated, anti-drone demand has roughly doubled since the Israel-Iran war, and the order book ended FY26 at ₹1,336 crore against ₹688 crore of revenue. The caveat is the working-capital appetite — screener’s broad measure shows 543 working-capital days and a 425-day cash conversion cycle — so each rupee of growth ties up a lot of capital on the way through.

3. Capital allocation for the stage. Mostly rational, with one large unanswered question. The FY25 QIP raised ~₹1,000 crore at a rich valuation — selling expensive paper is the right way to fund a build-out — and the bolt-on acquisitions (ARI, Anawave, AI Turing) have so far been small and productive: subsidiaries contributed ₹260 crore of FY26 revenue at 26–30% PAT margins. R&D spend is being deliberately raised as a structural cost, debt is near zero, and the dividend payout is a token 4–7% — appropriate while reinvestment returns are high. The question is the ₹1,308 crore still sitting in cash, now generating ₹86 crore of other income that flatters reported profit; it has not yet been deployed at anything like operating returns. No buyback history is visible in the data. And the promoters have sold their own stake down from 57% to 48.5% in three years — not a company allocation decision, but not the signal you’d choose either.

4. Price. Demanding, and explicitly priced for the rebound arriving on time. As of the June 2026 snapshot the stock trades at ₹1,780 — 83.4 times earnings and 8.4 times book on a ₹16,080 crore market cap, with a 0.11% dividend yield — after a year in which revenue fell 29%, the weakest quarterly operating margin in the window (28%), and a multi-year target already cut from ₹6,000 to ₹4,000 crore. The order book supports a strong FY27, but at this multiple the market is paying today for the ₹4,000 crore two-year promise from a management whose recent forecasts have repeatedly slipped on a procurement cadence it doesn’t control. A fine business at a price that leaves no room for the government to be slow again.

Sources

• Concall transcripts (4): Q1 FY26 (Jul 28, 2025), Q2 FY26 (Oct 27, 2025), Q3 FY26 (Feb 2, 2026), Q4 FY26 + full-year (May 2026) — BSE filings, converted to markdown. These carried the order-book trajectory, the evolving guidance, and the segment/demand detail.
• Annual reports (3): FY23, FY24, FY25 sections — extracts were thin on chairman’s-letter/MD&A narrative (flagged in the digests); the financial arc leans on the screener tables and concalls.
• Screener.in snapshot: quarterly and annual tables, ratios, shareholding — fetched 2026-06-07 (logged-out session); source of the working-capital and other-income flags.
• Research files: vault/Sources/Earnings/Zen Technologies Ltd/ — raw transcripts, AR sections, snapshot, per-document digests (not published).