India assembles far more electronics than it makes: production reached ₹13.11 lakh crore in FY26, roughly $148 billion, but 80–90% of a device’s component value is still imported, so the chain below is really the story of a country working its way down from final assembly toward silicon. Read it with the margin map in mind — the EMS contractors at stage 3 carry the market’s highest multiples on the thinnest margins, the design firms at stage 4 earn the fattest margins on rented talent, stage 2 is the hollow middle three subsidy schemes are trying to fill, and the best single business in the chain is a defence brand whose customer cannot leave. Tap any stage to open it, then any company for detail.
At the head of this chain
Highest return on capital and fastest sustained sales growth among the 38 listed companies in this chain above ₹1,000 Cr market cap and ₹500 Cr revenue. Sustained growth is the lower of the 3-year and 5-year sales CAGR, so a single strong year cannot win it — Netweb Technologies compounded 69.9% over three years and 72.5% over five. Same rule on every map. Snapshot 8 Sep 2026.
How the Indian electronics & EMS sector works
Six stages from silicon to the shop shelf. The market pays the most for stage 3; the durable margins live elsewhere.
- Margin pool
- Split three ways, none of them where the headlines point. Assembly (stage 3) earns 4–11% operating margins redeemed by asset turns — Dixon converts 3.6% margins into 29.2% ROCE. Design (stage 4) earns the fattest margins: Tata Elxsi runs 22.5% OPM at 30% on capital. And the single best business in the chain is a brand with a captive customer — BEL at 28.6% margins and 36.4% ROCE. The consumer brands at stage 5, squeezed between imported components and platform retail, mostly earn less than their own contractors.
- Bargaining power
- Upstream, with global component and chip suppliers — 80–90% of a device's bill of materials is imported, so input prices are set in Shenzhen, Taiwan and Seoul. Downstream, with platforms and large retail, which set the promotional calendar. The Indian stages in between — EMS, even brands — are price-takers on both sides; only defence electronics and genuine design IP escape.
- Demand or supply led
- Policy-led on supply, cyclical on demand. The mobile PLI conjured an assembly industry in five years and is now winding down; ECMS is attempting the same for components; the ISM for fabs. Underneath, demand is consumer-cyclical — one rained-out summer in FY26 cut profits at Voltas, PG Electroplast, EPACK and Symphony simultaneously. The chain's growth is real; its smoothness is not.
- Who owns the customer
- For phones — half the market — unlisted brands (Samsung, Apple, Xiaomi) and two e-commerce platforms. For appliances, the listed brands own distribution but rent manufacturing from stage 3. The EMS companies own no customers at all: their revenue is a contract, which is why DCX Systems' revenue fell 27% in FY26 and another 53% year on year in the June 2026 quarter on order timing alone and why customer concentration is the number to check before any multiple.
- Barriers to entry
- Low at assembly — capital and PLI approval, which is why EMS margins are thin and competition keeps arriving. High at components and silicon (capex, process IP, yield learning — the ECMS exists because subsidy alone hasn't cracked it) and at brand distribution (a million-outlet channel takes decades). The stage with the lowest entry barrier currently carries the highest multiples; that tension resolves one way or the other.
- Threat of substitutes
- For the EMS layer the substitute is the customer's own factory — brands can in-source (IFB and V-Guard already manufacture in-house) or global EMS can undercut. For components the substitute is the status quo: imports. For retail, direct-to-consumer. The defensible positions are design IP, the defence customer, and distribution density — everything else competes on price against someone bigger.
This is a generalist read of how the industry typically works, not a rule. Anomalies exist at every stage — a well-run company in a poor part of the chain routinely beats a badly-run one in a good part, and structure changes with the cycle. Use it as a starting frame, not a conclusion.
No company matches that search.