Data Patterns — a design lab that bills like a factory
Data Patterns (India) Limited
Snapshot
Data Patterns designs and builds the electronic brains of Indian weapons — the radars, the jammers, the avionics, the seekers that guide a BrahMos missile to its target. It does this almost entirely in-house: it draws the chips-and-boards itself rather than buying foreign building blocks and bolting them together. Market cap ₹26,996 cr, share price ₹4,822 (52-week range ₹2,131–₹4,956), trading at 98.7× earnings and 15.6× book value, with a return on equity (profit earned per ₹100 of owners’ money) of 16.9% and return on capital (profit per ₹100 of all money in the business) of 23.3%. In one phrase: a rare, founder-built engineering franchise riding a once-in-a-generation defence tailwind — but one that has been priced as if the tailwind never stops. As of 2026-06-20, from screener snapshot.
The verdict in two boxes — the business first, the price second
Keep them apart on purpose. The first box would read the same if the share price doubled or halved overnight. The second box is about today’s quote, and only today’s.
Box 1 — The business (durable):
| Lens | Result |
|---|---|
| Business-quality score | 17 / 23 (Quality 8.5/12 · Growth 5.5/6 · Longevity 3/5) |
| Buffett rubric | 6.5 / 10 PASS |
| Business bucket | Good, leaning Great — a high-return design house held back by a cash-hungry working-capital cycle |
| Wealth-creator type | Enduring (real IP moat, structural tailwind) · Consistent (profit has risen every year for a decade) |
| Economic Profit | ₹+85 cr (net worth ₹1,736 cr × (RoE 16.9% − CoE 12%)) — creating value, modestly |
A genuinely good business that is a real wealth creator — a design lab the country can’t easily replace — but one whose returns are dragged down by how long its money sits stuck in receivables and inventory. That verdict holds whatever the stock does today.
Box 2 — The price today (a current phenomenon):
| Reading | Result |
|---|---|
| CMP | ₹4,822 (as of 2026-06-20) |
| Price pillar | 0 / 2 (PEG ~3.3× · 5-yr payback ~10.5×) |
| Margin-of-safety band | roughly ₹1,400–₹2,200 (where PEG nears 1× / payback nears reach) |
| Mr. Market’s mood now | Greedy — a defence re-rating supercharged by the May-2025 India–Pakistan conflict; the stock roughly doubled off its low |
| CMP vs the band | Demanding — priced ~2–3× above where the framework’s price discipline would smile |
Today the market is pricing it rich — a mood driven by war-headlines and an “Atmanirbhar” (self-reliance) story, which can cool while the business above carries on unchanged.
In plain English
Picture a workshop in Chennai where about 1,200 engineers do something most Indian defence suppliers can’t: they design the electronics from scratch. When a fighter jet needs a radar, or a missile needs an eye to find its target, or an aircraft needs a box that scrambles the enemy’s signals, most companies import the hard parts and assemble them here. Data Patterns draws those hard parts itself — the chips, the radio-frequency bits, the software, the cooling — and owns the blueprint. That is the whole moat in one sentence. Owning the blueprint is why it earns gross margins north of 60% and an EBITDA margin of ~40% — numbers an analyst on the May call said “no other radar electronics company has achieved in the world.” When you own the recipe, you keep the profit; when you only assemble someone else’s recipe, the recipe-owner keeps it.
Why is this a good boat right now? Because India has decided to stop importing its weapons. The government wants home-made radars, jammers and missiles, and it is putting real money behind that wish. Data Patterns sits exactly where the money is flowing. Its order book has gone from ₹730 cr a year ago to about ₹2,062 cr today, and management says repeat orders for products it has already built and delivered could add another ₹1,900 cr over the next year. The founder, Mr. Rangarajan, talks openly about wanting to “scale into a multi-thousand-crore company” rather than crawl along at 20% a year. The tailwind is real and it is long — this is a multi-decade rearmament, not a one-year spike.
Now the splinter under the fingernail. A great recipe is wonderful, but this business is slow to get paid. Its customers are government agencies and they pay on their own clock — money sits stuck in unpaid bills (receivables) for about 287 days, nearly ten months, and the stuff it’s building sits as inventory for many months more. The result: in two of the last four years (FY23 and FY25) the company reported fat profits but the cash from operations was actually negative. Profit on paper is not the same as money in the bank, and here the gap is the single most important thing to keep an eye on. It is the reason this is a “Good” business and not yet a “Great” one in the Buffett sense — a Great business throws off cash; this one keeps having to feed cash back in to grow. To its credit, the cycle is improving (the cash-conversion cycle fell from 428 days to 365 days last year), and the company is debt-free with a cash cushion from past share sales. But it is the thing that separates a fountain from a sponge.
So the tension is simple. The boat is good and getting better. The seat is expensive. At 98.7× earnings and 15.6× book, the price assumes the order book keeps compounding, the cash cycle keeps healing, and exports to Europe and the U.S. actually arrive — all at once, for years. Pay that price and you’ve already handed the next several years of success to the seller. The business does not need you to believe that; the price does.
Sitting down with the management
If Buffett and Raamdeo Agrawal sat across from Srinivasagopalan Rangarajan for an afternoon, I think they’d come away mostly charmed — and with one large worry written on a napkin.
Here is who he is. Rangarajan, a chemical engineer from Madras (with a stint at IIT Madras), founded Data Patterns in 1985 and spent nearly four decades turning a small instrument-automation shop into one of India’s few genuinely design-owning defence electronics houses [MEDIUM — IIFL interview, 25 Jan 2022]. He is an engineer’s engineer, not a deal-maker. Read his concall remarks and you hear a man who cares about building the thing — “where all the IP is created in-house, we don’t import anything and integrate.” He waves the “Make in India with Pride” flag and means it. That is exactly the “passion plus integrity plus domain mastery” Buffett looks for: the business is an extension of an obsession, not a vehicle for ego.
How has he spent the owners’ money? Cleanly, by the record. The December-2021 IPO raised about ₹588 cr (part of which was the promoters selling some shares, part fresh money used to pay down debt and fund working capital and a facility upgrade) [HARD — chittorgarh.com; Business Standard, Dec 2021]. In FY23 the company raised another ₹500 cr from large institutions (a “QIP” — selling new shares to professional investors), which is why reserves jumped from ₹564 cr to ₹1,156 cr that year [HARD — Business Standard, Mar–Apr 2023]. The cash went into product development, testing facilities and capacity — and the business is debt-free as a result. Dividends have been steady and rising, around a 20% payout. The one-dollar test (did each retained rupee create at least a rupee of value?) reads pass, modestly — return on equity sits a few points above the cost of that equity, so the retained money is earning its keep, though not spectacularly.
Do they talk straight? Mostly yes, and refreshingly so. On the May call Rangarajan repeatedly refused to be pinned down on numbers he couldn’t honestly forecast — “I can’t predict for them, these are all government customers,” “I won’t predict time lines and revenue models” — and he flatly corrected analysts who tried to put words in his mouth about “delays.” A founder who says “I don’t know” several times in one call is usually telling you the truth. He guides conservatively (20–25% growth) and has tended to beat it. That is the candor the letters prize.
The governance plumbing is clean: no promoter pledging, a stable ~42.41% promoter stake with no drip-selling since the IPO, four independent directors on the audit committee, an independent chair of the pay committee, and no SEBI run-ins or accounting controversies I could find [HARD/MEDIUM — screener.in; company filings]. One nuance worth filing: the marquee private-equity backer, Mathew Cyriac’s Florintree, fully exited in February 2024 at a reported ~21× return, selling to a who’s-who of mutual funds and Singapore’s GIC [HARD — Business Standard, 16 Feb 2024]. That’s a clean, profitable exit, not a distress sale — but the sophisticated early sponsor has cashed out, and that’s a fact to note, not ignore.
Now the worry on the napkin. Two of them, really. First, the cash question from the section above — fat profits that don’t always become cash — is a management question as much as a business one: a forensic Agrawal would want several more years of operating cash flow catching up to profit before fully trusting the accounts, even though everything here looks honest rather than manipulated. Second, key-man risk. This company is Rangarajan. The bench is family-anchored (his wife is a whole-time director), there is no visible groomed successor, and he was just re-appointed Chairman & MD for another five years. A 40-year-old, founder-dependent design house with no obvious second-in-command is one health scare away from a question mark.
Would Buffett and Agrawal shake hands on this management? Yes — with a firm handshake and a raised eyebrow. What would change their mind: a real succession plan emerging, or operating cash flow finally and durably catching up with reported profit. What would break it: any sign the receivables are being used to flatter the P&L, or the founder leaving with no one ready to take the controls.
What’s on the horizon (live-issues tracker)
Three things will decide the next one-to-three years. The first is the crux.
1. The crux — can a “design-and-deliver-fast” niche player keep its returns as the prize gets bigger and bigger? 🟡
The make-or-break sentence: This investment works if and only if Data Patterns can scale revenue several-fold while keeping both its design-owned margins AND turning that profit into actual cash. Everything else is detail. (Full interrogation below.)
2. The export and global-OEM opening 🟡 (mixed / early)
What it is: until now Data Patterns sold almost entirely to Indian agencies. Management now says European and U.S. defence majors are visiting Chennai, impressed by the prices and — repeatedly stressed — the speed (“they are just astonished that we can deliver these things in a time frame they never imagined”). The export order book is small (~₹53–63 cr) but they’ve already shipped Transportable Precision Approach Radars to a European country with successful site acceptance [HARD — May & Feb 2026 concalls]. How it’s going: real but tiny. RFPs are arriving, a marketing team is being built “early next year,” and they’re willing to co-invest in products for world markets. This is genuine optionality, not yet a revenue line. What’s next: first export contracts from European OEMs (management guides 2–4 months on the May call — watch whether that slips). Bull: a cost-and-speed edge into a rearming West. Bear: defence exports are slow, certification-heavy and politically gated; “2–4 months” from a government-pace business is optimistic.
3. The capex ramp ahead of the orders 🟡 (early, deliberately cautious)
What it is: to scale to “multi-thousand crore,” they’re building roughly nine floors of new factory space and seeker/EW capacity. How it’s going: deliberately paced — Rangarajan is explicit that he won’t “sit on capex infrastructure” if the contracts don’t come, and won’t over-build only to hand the business to “foreigners who come on the back route.” That discipline is admirable but also a tell: the capacity is being built on faith in an order book that is largely still “expected” and “negotiated,” not signed. What’s next: watch capex on the cash-flow statement against order-book conversion. Bull: capacity ready when the wave lands. Bear: capex now, cash-burning, while the big orders keep being “1 to 2 months away.”
The crux, interrogated
The plain-English mechanism. Data Patterns’ moat is design ownership — it draws the electronics itself and writes off the development cost through its profit-and-loss as it goes. That’s why margins are so high: a rival who imports the building blocks and assembles them hands the profit to the chip-maker abroad; Data Patterns keeps it. The threat isn’t that someone copies a product — these are classified, customer-specific systems. The threat is subtler: as the prizes get bigger (whole fighter-jet electronics suites, not boxes), the contracts get longer, the customers slower, and the working capital heavier — and bigger Indian corporates with deeper pockets start wanting in.
Test the analogy. Is this like a boutique architecture firm that designs beautiful houses, then gets asked to design a whole city? The skill scales, but the cash strain explodes — you carry months of work before you’re paid, and a deep-pocketed builder can simply out-wait you. That analogy holds here: the danger to Data Patterns is less “lost moat” and more “the moat is real, but the money gets stuck for longer as it grows.” The 287-day receivables and negative-cash years are that strain showing up already.
Map the competition by name.
| Rival | Backed by | Posture / proof point |
|---|---|---|
| Bharat Electronics (BEL) | Government (PSU), ₹3.1 lakh cr mcap | The 800-pound gorilla. 36.5% RoCE, 27.6% RoE — better return ratios than Data Patterns, and far better cash. The default prime for big radar/EW systems. |
| Astra Microwave | Listed, ₹16,000 cr mcap | The closest direct peer; an analyst on the May call asked Rangarajan directly about Astra “transitioning to complete systems like us.” He refused to comment “on open line” — telling. Similar RoE (16%), similar journey up the value chain. |
| Bharat Forge / BEML (AMCA consortium) | Industrial giants | The twist — Data Patterns is partnered with Bharat Forge (reportedly 50/30/20 with BEML) for the AMCA fighter programme, shortlisted for the RFP [SOFT — defence press, 2025–26]. So the big corporate isn’t only a rival; on the marquee programme it’s an ally. The risk is they need it more than it needs them. |
The real-world precedent. This has happened before, both ways. The bad precedent: many a brilliant defence-electronics SME worldwide hits a ceiling when programmes get large — the prime contractor (BEL’s analogue) absorbs the systems work and the SME slides back to being a sub-supplier on thin margins. The good precedent: a handful of design-IP houses (the small radar/seeker specialists in Israel and Europe) stayed independent precisely because their IP was too specialised and too fast to replace — the prime needed them more than they needed the prime. Which one Data Patterns becomes depends entirely on whether its in-house design stays genuinely ahead.
The answered follow-on questions. Is the threat to share or to margin? Mostly to cash and pace, not share — the niche is protected by classified IP; the danger is the money getting stuck longer as contracts lengthen. Which segment is protected? The pure in-house-designed products (highest margin) are safest; the system-integration contracts where they buy in trucks/cooling are where margins thin — management said exactly this. What’s the incumbent (BEL) doing? BEL is everywhere and cheaper-of-capital, but it’s a slow PSU; Data Patterns’ whole pitch is speed BEL can’t match. Has anyone actually moved? Yes — the order book tripled in a year and exports have started, so the wave is real, not anticipated.
The honest view: takeable, leaning positive — but not a slam-dunk. The moat is real and the tailwind is real. I’d take the side that the business keeps winning. But the margin-of-error sits in the cash conversion: the thesis quietly assumes the receivables cycle keeps healing as the company scales. If it doesn’t — if growth keeps demanding cash faster than it returns it — then a “Good” business stays Good and never becomes the Great cash-fountain the price is paying for. Not “too hard.” Just: the boat sails, but watch the bilge pump.
The watch-list:
- Operating cash flow turns durably positive and tracks profit — the single most important number. (FY26 OCF was ₹80 cr vs ₹271 cr PAT — still a big gap.)
- Cash-conversion cycle keeps falling toward management’s stated 320–340 day target (was 365 in FY26).
- The ~₹1,900 cr of “expected/single-vendor” repeat orders convert to signed contracts during FY27 — or keep being “1 to 2 months away.”
- First real export contract from a European/U.S. OEM lands (guided 2–4 months on the May-2026 call).
- AMCA RFP outcome for the Bharat Forge consortium.
- Any succession announcement — a named, credible number-two.
QGLP scorecard (the Motilal Oswal lens) — the receipts
| # | Question | Score | Evidence |
|---|---|---|---|
| QUALITY OF BUSINESS | |||
| 1 | Large opportunity? | 1 | India’s domestic defence opportunity US$100–120 bn over 5–6 yrs; ~13% sector CAGR; indigenisation push (_ar_FY24). Huge. |
| 2 | Favourable industry structure? | 1 | Few credible indigenous design houses; high, stable OPM 38–45% signals pricing discipline, not a price-war (profit_loss OPM row). |
| 3 | Clear, defensible moat? | 1 | In-house IP design (chips→software→full radar/EW), classified customer-specific systems, ~1,200 engineers, ~28-yr record. RoCE >20% since FY20. |
| 4 | High return ratios (RoE/RoCE >15%)? | 0.5 | RoCE 23.3% (strong); RoE 16.9% (good but not great, and below RoCE — a working-capital drag). Both >15% only since ~FY20 (ratios, ratios_table). |
| 5 | Asset-light / low capital intensity? | 0.5 | Fixed assets light, BUT working-capital heavy — negative OCF in FY23 & FY25, large inventory + receivables. A “Good”, not “Great”, capital profile (cash_flow). |
| 6 | Favourable terms of trade (negative working capital)? | 0 | Debtor 287 days vs payable 82 days — it banks its customers. Cash-conversion cycle 499 days (ratios_table). The opposite of favourable. |
| QUALITY OF MANAGEMENT | |||
| 7 | Unquestionable integrity? | 1 | No pledging, no promoter selling since IPO, clean audits, no SEBI issues, single reportable segment honestly stated (shareholding; web). |
| 8 | Proven execution track record? | 1 | Beats its own 20–25% guidance; FY26 revenue +31%, order book tripled; ~40-yr operating history. |
| 9 | Growth mindset & vision? | 1 | Explicit “scale to multi-thousand crore”; investing ahead in seekers, airborne radar, exports, AI (concalls). |
| 10 | Superior capital allocation? | 0.5 | Debt-free, sensible IPO/QIP use, ~20% payout — BUT the QIP cash partly funds a hungry working-capital cycle; one-dollar test passes only modestly (EP just +₹85 cr). |
| 11 | Clear succession plan? | 0 | None visible. Founder-dependent; wife on board; just re-appointed CMD 5 yrs. Key-man risk (web). |
| 12 | Minority interests protected? | 1 | Rising dividends, no value leakage, reasonable governance, clean RPTs (web; filings). |
| GROWTH | |||
| 13 | Structural tailwind > GDP? | 1 | Defence indigenisation + global rearmament; sector growing far faster than nominal GDP (_ar_FY24). |
| 14 | Volume-led (sustainable) growth? | 1 | Growth is new programmes/products and order-book expansion, not price hikes — durable. |
| 15 | Operating leverage? | 1 | OPM rose from ~19% (FY19) to ~40–45% as sales scaled ~7× (profit_loss). Textbook operating leverage. |
| 16 | Manageable financial leverage? | 1 | Borrowings ₹5 cr vs net worth ₹1,736 cr — effectively debt-free (balance_sheet). |
| 17 | Market-share gain potential? | 0.5 | Gaining share moving subsystem→full-systems, but BEL/Astra also climbing; share-gain real but contested. |
| 18 | Earnings growth > 15% CAGR? | 1 | PAT 5-yr CAGR ~37%, 3-yr ~30%; guidance 20–25%+ (profit_loss). Comfortably clears the bar. |
| LONGEVITY | |||
| 19 | Model relevant 10–15 yrs (low disruption)? | 1 | Defence electronics is a multi-decade need; in-house IP is the opposite of fragile. |
| 20 | Can extend Competitive Advantage Period? | 0.5 | Moat real but must out-design BEL/Astra continually; widening if IP stays ahead, not guaranteed. |
| 21 | Can sustain Growth Advantage Period? | 1 | Vast addressable market, low penetration, long order visibility — runway is long. |
| 22 | Geographic / product diversification headroom? | 0.5 | Exports + new product lines (drones, seekers, airborne radar) — optionality real but unproven. |
| 23 | Adaptive, resilient culture? | 0 | Default — single-founder, no proof through a founder transition yet; strong engineering DNA but key-man-anchored. (scored conservatively) |
| PRICE (reported separately, not in the /23) | |||
| 24 | Valuation reasonable (P/E vs growth)? | 0 | PEG = 98.7 / ~30 ≈ 3.3×. Far above 1× (ratios; computed). |
| 25 | Margin of safety (PEG <1 or payback <1)? | 0 | 5-yr payback ~10.5×; PEG ~3.3×. Expensive on both. |
| Business-quality total (Q1–Q23) | 17 / 23 | Quality 8.5 · Growth 5.5 · Longevity 3 | |
| Price pillar (Q24–Q25) | 0 / 2 | reported separately | |
| (Canonical QGLP /25, for fidelity) | 17 / 25 |
The pattern is clean: Quality and Growth are the strength; Longevity is good-not-great (key-man + must-keep-out-designing); and Price is the entire problem. This is a high-scoring business — a 17/23 puts it firmly in “strong quality, a couple of real gaps” territory — whose only failing on the durable side is the cash-conversion engine (Q5/Q6). The /23 says good wealth creator; the /2 says priced like a saint.
Buffett lens (the Berkshire-letters read)
| # | Test | Verdict | Evidence |
|---|---|---|---|
| 1 | Good boat? (business > management) | PARTIAL | High RoCE (23%) but capital-hungry working cycle → “Good”, not “Great”. A fountain that occasionally runs dry. |
| 2 | Moat + franchise + pricing power | PASS | Stable/rising 40%+ EBITDA margin through scaling; classified in-house IP = a real franchise. RoE > CoE since ~FY20. |
| 3 | See’s test — high returns on little capital | PARTIAL | Earns well, but growth eats cash (negative OCF FY23, FY25). Must keep feeding capital — the asset-heavy-in-disguise case. |
| 4 | Capital allocation — one-dollar test | PASS | Retained + raised capital lifted both book and market value; debt-free; no value-destroying M&A; modest but positive EP. |
| 5 | Owner-oriented, candid management | PASS | Founder says “I don’t know” honestly; under-promises, over-delivers; no spin. “Eat our own cooking” — 42% stable stake. |
| 6 | Integrity / forensic (no “credit P&L, debit BS”) | PARTIAL | Honest accounts, BUT profit doesn’t reliably become cash and receivables are very high — the legitimate version of the worry; watch it. |
| 7 | Circle of competence / predictability | PARTIAL | You know what it does in 10 yrs (defence electronics); but contract timing and order conversion are lumpy and government-paced. |
| 8 | Mr. Market — gift or trap now? | FAIL | 98.7× P/E, 15.6× book, doubled on war-sentiment. Priced for perfection — the patsy at this price pays for years of success upfront. |
| 9 | Patience / compounding runway | PASS | Long runway at decent RoE; large addressable market, low penetration; can compound for a decade if the cash cycle heals. |
| 10 | Honest red flag (mandatory, unscored) | — | See below. |
Score: 6.5 / 10 PASS — a real business with two real gaps (cash conversion and price).
The See’s test, spelled out. See’s Candies was magic because it threw off mountains of cash while needing almost none reinvested. Data Patterns is the partial version: it earns See’s-like margins, but unlike See’s it has to keep pouring cash into inventory and unpaid customer bills to grow. FY26 operating cash flow was ₹80 cr against ₹271 cr of reported profit — only about 30 paise of every reported rupee showed up as cash. A true See’s converts most of it. This is the line between a Great business and a merely Good one, and Data Patterns sits right on it — improving, but not across yet.
The one-dollar test, spelled out. Has each rupee retained created a rupee of value? Net worth is ₹1,736 cr earning a 16.9% return against a ~12% cost of capital — so the retained money earns about 4.9 percentage points above its cost, generating roughly ₹85 cr a year of economic profit (value above the cost of owners’ money). That’s a pass — the rupees are creating value — but a thin one. A Great wealth creator (BEL, at 27.6% RoE) generates a vastly bigger spread on the same equity. So management is allocating capital well, not brilliantly; the working-capital drag is what holds the spread down.
The framework metrics
- Economic Profit = Net Worth ₹1,736 cr × (RoE 16.9% − CoE 12%) = ₹+85 cr. Creating value, modestly. (CoE 12% used — Indian benchmark mid-point.)
- Terms of Trade = Debtor days 287 ÷ Payable days 82 ≈ 350%. Deeply unfavourable — it banks its customers for ~10 months. (Cash-conversion cycle 499 days, improving from a worse past.)
- 5-yr Payback = Mcap ₹26,996 cr ÷ projected cumulative 5-yr PAT (~₹2,560 cr, assuming ~22% PAT CAGR off ₹271 cr) ≈ 10.5×. Far from the <1× multi-bagger signal.
- PEG = P/E 98.7 ÷ PAT growth ~30% ≈ 3.3×. Well above 1×.
- RoE − CoE spread = ~+4.9 pts. RoE > 15% in roughly 5 of the last 10 years (the high-return era only began ~FY20; earlier years were single-digit) — falls short of the “≥7 of 10” moat-proof bar on history alone, though clearly above it now.
- Consistent/Volatile test = Consistent (pass) — net profit rose essentially every year FY18→FY26 (1→8→21→56→94→124→182→222→271 ₹cr); no fall >10%, terminal far above initial. A Consistent compounder → value on earnings (P/E), which is exactly why the market dares to use a big multiple.
Peer comparison
| Company | Mkt cap (₹cr) | CMP (₹) | P/E | P/B | RoE | RoCE | FY-latest sales (₹cr) |
|---|---|---|---|---|---|---|---|
| Data Patterns | 26,996 | 4,822 | 98.7 | 15.6 | 16.9% | 23.3% | 925 |
| Bharat Electronics (BEL) | 3,12,054 | 427 | 51.5 | 13.0 | 27.6% | 36.5% | (very large) |
| Astra Microwave | 15,990 | 1,684 | 82.9 | 12.2 | 16.0% | 20.2% | — |
| Paras Defence | 11,352 | 1,409 | 132 | 15.7 | 12.6% | 16.9% | — |
The relative read sharpens the absolute one without rescuing it. Data Patterns is the richest-priced of the genuine peers on P/E except Paras — and Paras has worse returns, so its higher multiple is not a defence. The standout is BEL: the giant PSU earns better returns (27.6% RoE, 36.5% RoCE) at barely half the P/E — a sobering comparison that says “the best-run, best-capitalised name in this space is cheaper and higher-quality on the numbers.” Data Patterns’ distinctive edge over BEL is speed and pure in-house design depth (BEL is a slow PSU); its edge over Astra is a few years’ head-start up the full-systems value chain. But on price, the whole defence-electronics shelf is expensive, and Data Patterns sits near the top of it. The absolute band (₹1,400–₹2,200) and this relative read agree for once: nothing here is cheap.
Latest quarter & what’s happening now
Q4 FY26 (reported 15 May 2026): revenue ₹345 cr (down ~13% YoY on programme timing, but nearly doubled sequentially), EBITDA margin a remarkable 56%, PAT ₹138 cr. Full-year FY26: revenue ₹925 cr (+31%), EBITDA ₹371 cr (+35%), PAT ₹271 cr (+22%). [HARD — May-2026 concall; screener quarters/profit_loss.]
Two concall takeaways: (1) Order inflow exploded — ₹1,121 cr in FY26, +216% YoY; order book ~₹2,062 cr (incl. negotiated), plus ~₹1,900 cr of expected single-vendor repeat orders [MEDIUM — May 2026]. (2) Guidance held at 20–25% revenue growth with 38–40% EBITDA margins and net-cash status, but management hinted real upside (“substantially higher than guidance”) if repeat orders land early [MEDIUM]. Live catalysts: BrahMos seeker production orders (expected “4–5 months”, SOFT/MEDIUM); EW jammer pods moving to flight testing (MEDIUM); European export contracts (guided 2–4 months, SOFT); AMCA RFP for the Bharat Forge consortium (SOFT).
Where the two lenses agree — and disagree
They agree on the spine: this is a genuine, IP-moated, founder-led wealth creator (QGLP 17/23; Buffett tests 2, 4, 5, 9 all PASS) trading at a price both frameworks flag as demanding (QGLP Price 0/2; Buffett test 8 FAIL).
They part company in one revealing place. The QGLP checklist, by counting tailwind + margins + growth, scores the business a confident 17/23 — it reads almost Great. The Buffett lens is harder-nosed on two points the checklist under-weights: the See’s / cash-conversion test (test 3, PARTIAL) and forensic cash-backing of profit (test 6, PARTIAL). That’s the signal. The numbers say Great; Buffett’s insistence that profit must become cash drags it back to Good. Trust the Buffett flag here — it’s pointing at the exact thing (₹80 cr OCF on ₹271 cr PAT) that separates this from a true cash-fountain. The second, smaller divergence is succession: the checklist barely dents it; Buffett’s “is the company bigger than one person?” lands a clean key-man worry.
Final classifications: Good business (leaning Great if the cash cycle heals) · Enduring wealth creator (real moat, structural demand) · Consistent compounder (profit up every year) · sitting in the upper-middle of the Economic Profit curve — creating value, but not yet a super-profit gusher.
The price as a current phenomenon
This section judges the price, not the business. The 17/23 above is settled; here we only ask what Mr. Market is charging today.
The margin-of-safety band. The framework’s price discipline wants PEG ≤ 1× and/or a 5-yr payback approaching reach. At 30% PAT growth, PEG = 1× implies a P/E near 30× — versus 98.7× today. Run that against FY26 EPS (₹48.5) and you get a “price-discipline-satisfied” zone of roughly ₹1,400–₹2,200 (PEG ~1–1.3× / payback edging toward sane). Note this band brackets the stock’s own 52-week low of ₹2,131 — i.e. even at its bottom this year, it was only just entering the framework’s smile. At ₹4,822 the market is paying ~2–3× the band. This is arithmetic, not a forecast: it’s the price at which a patient owner stops pre-paying for future success.
The Mr.-Market read. The crowd is greedy on this name, and you can name the reason. Indian defence stocks re-rated violently after the May-2025 India–Pakistan conflict (“Operation Sindoor”) put indigenisation on every front page; Data Patterns roughly doubled off its ₹2,131 low. The fundamentals genuinely improved (order book tripled), but the multiple expanded on sentiment as much as substance. A 98.7× P/E embeds: order book keeps compounding, cash cycle keeps healing, and exports arrive — all at once, for years. That’s not fearful pricing; that’s a market that has already given the company credit for winning.
The tension, plainly. A wonderful business can sit at an unwonderful price — and that is precisely this case. The boat is good and improving; the seat is expensive. The 17/23 doesn’t move if the stock halves tomorrow; the buyer at 98.7× simply does better or worse depending on whether reality matches a near-flawless script. And remember: this reading can flip next quarter — a single soft order-conversion print, or defence sentiment cooling, could re-price the seat sharply, without one bolt changing in the workshop.
Conviction texture
The bull case, strongest form: You’re buying the design lab India can’t replace, at the start of a multi-decade rearmament where the government is forcing buyers to “Make in India.” In-house IP means 40% EBITDA margins no integrator can match; the order book tripled in a year; exports to a rearming West are just beginning; the founder is honest and under-promises; the balance sheet is debt-free. If revenue compounds 25%+ and the cash cycle heals as it’s already doing, today’s nosebleed multiple becomes a memory and the compounding runway is a decade long.
The bear case, strongest form (test 10’s red flag): You are paying 98.7× earnings and 15.6× book for a business that, in two of the last four years, generated no cash — and a comparable, better-returning peer (BEL) trades at half the multiple. The single strongest reason this is not yet the wealth creator the price implies: reported profit isn’t reliably becoming cash (₹80 cr OCF on ₹271 cr PAT in FY26), receivables sit at 287 days, and growth keeps demanding capital. Add key-man dependence on a founder with no visible successor, government-paced lumpy orders that make any quarter a coin-flip, and a price that has already discounted years of flawless execution. The numbers partly refute the bear (the cash cycle is improving, profit growth is real) — but they don’t refute the valuation half at all.
What the numbers actually support: a genuinely good, improving, well-run business — and a price that has run far ahead of where the framework’s discipline would buy. Both can be true at once; here they are.
The two-or-three things to watch that tip it: (1) operating cash flow durably catching profit; (2) the ₹1,900 cr of expected repeat orders converting to signed contracts; (3) a real export win and/or a succession answer. No buy/sell/hold — just: the boat is sound, and Mr. Market is, for now, charging a premium fare.
Sources
- Screener snapshot: https://www.screener.in/company/DATAPATTNS/ (fetched 2026-06-20)
- Q4 FY26 concall, 15 May 2026 (
_concall_May-2026.md); Q3 FY26 concall, 6 Feb 2026 (_concall_Feb-2026.md) - Annual Reports FY25, FY24 (
_ar_FY25_sections.md,_ar_FY24_sections.md) - Founder background/philosophy: IIFL interview, 25 Jan 2022
- IPO (Dec 2021) & QIP (Mar–Apr 2023): chittorgarh.com; Business Standard
- Florintree (Mathew Cyriac) full exit Feb 2024: Business Standard, 16 Feb 2024
- AMCA / Bharat Forge consortium: defence press (idrw.org / defensemirror.com), 2025–26 [SOFT]
- Defence re-rating context: Operation Sindoor, May 2025
- Peers (screener snapshots, 2026-06-20): Bharat Electronics, Astra Microwave, Paras Defence
- Assumptions: Cost of Equity 12%; PAT CAGR ~22% (guidance) to ~30% (3-yr actual) used where stated; ~5.6 cr shares (equity ₹11 cr at ₹2 face value).