How to Build a Make in India Sourcing Strategy for Electronics

A Make in India sourcing strategy for an electronics product starts with the product's economics and its supply risk, not with a localisation percentage chosen in a meeting. Some lines on the bill of materials can move to Indian suppliers this quarter. Others, such as most advanced semiconductors, will stay imported for years, and pretending otherwise only produces a cosmetic local invoice.

This guide sets out a sequence that holds up in front of a finance team and a customer auditor: define how local value will be measured, build a clean baseline, segment the BOM, compare total cost, check which government schemes genuinely apply, then localise in waves with evidence at each gate. Scheme details were checked on official pages in October 2026 and change often, so treat them as a starting point for your own check.

Decide how "local" will be measured

Teams often argue about localisation because they are measuring different things. Indian supplier spend, domestic conversion value, Indian-made material, Indian design work and imported content hidden inside a local supplier's price are separate numbers. Track them separately.

If you sell to government buyers, there is a formal definition to align with. DPIIT's Public Procurement (Preference to Make in India) Order 2017, most recently revised on 19 July 2024, defines local content as the value added in India: the total value of the item (excluding net domestic indirect taxes) minus imported content (including customs duties), as a percentage of the total. Its explanatory notes exclude imported items bought locally from resellers or distributors, and exclude licence fees, royalties and technical charges paid outside India. A Class-I local supplier needs at least 50% local content and a Class-II local supplier at least 20%, unless the nodal ministry has notified a higher threshold for that item (checked October 2026).

That exclusion is the most useful idea for private buyers too. A distributor's Indian invoice for an imported IC does not add domestic value. Ask each supplier for a value-chain declaration, keep one calculation method across the programme, and get finance to sign off the method before anyone announces a target.

Build a baseline for every line

Before deciding what to move, capture the current position for each BOM line and each bought-in service:

• exact manufacturer part number or drawing, with annual demand;
• current manufacturer, supplier and production site;
• unit price, freight, duty, taxes and payment timing;
• lead time, MOQ, pack size, NCNR terms and any allocation;
• lifecycle status, quality history and whether it is sole-sourced;
• present domestic content, measured by your agreed method;
• the qualification effort and customer approvals a change would trigger;
• tooling, firmware, test or certification that depends on the part.

From this, calculate landed cost and cost-at-risk for each line. The second figure matters more than buyers expect: a small unit saving is worthless if the alternate delays a certification or causes field returns. The guide to landed cost for imported components covers the import side of that arithmetic.

Sort the BOM into segments

Put each line into one of five segments. The segment decides the approach, and it stops a team from forcing a supplier-development project into a launch schedule.

Then add spend, line-stop impact and the time each move would take. Early candidates are usually packaging, labels, cables, harnesses, simple mechanicals, selected magnetics, bare PCBs, test fixtures and final assembly. The category name is only a hint; supplier evidence decides. PCB and PCBA suppliers and passive and electromechanical part makers are reasonable places to start a longlist for wave one and two items.

Make the case on total cost

Compare like with like. A useful working formula is: total sourcing cost equals piece price, plus NRE and tooling, plus logistics and duties, plus inventory carrying, plus quality and qualification cost, plus an allowance for expected disruption.

Normalise quotes before comparing them: Incoterm and delivery point, currency and payment terms, yield and attrition, programming and test, warranty, packaging, minimum quantities, tax treatment and ramp assumptions.

An illustrative case shows why. Suppose an imported transformer lands at ₹410 and an Indian maker quotes ₹385. On 5,000 units a year the saving is ₹1.25 lakh. If qualification, safety retesting and a new fixture cost ₹3.5 lakh, the move pays back in 2.8 years. That may still be right for continuity reasons, but it is a strategic decision, not a quick win.

Local supply shortens transit and makes visits easy, yet it can still carry imported exposure. Ask each Indian supplier which materials, machines and sub-tier parts are foreign, and how currency or duty changes pass into their price.

Check which government schemes actually apply

Incentives can improve a sound business case. They should never be the reason a weak one goes ahead. As of October 2026 the main MeitY programmes look like this:

• ECMS. The Electronics Component Manufacturing Scheme was notified on 8 April 2025 with an original outlay of ₹22,919 crore, which the Union Budget 2026-27 raised to ₹40,000 crore, according to a PIB backgrounder of February 2026. Its target segments are sub-assemblies such as display and camera modules (A), bare components including multi-layer PCBs and electromechanicals (B), selected bare components such as HDI and flexible PCBs and SMD passives (C), supply-chain and capital equipment (D) and telecom optical transceivers (E). The ECMS portal states that windows for segments A, B, C and E closed on 30 September 2025, while segment D stays open until 30 April 2027.
• SPECS. The Scheme for Promotion of Manufacturing of Electronic Components and Semiconductors offers 25% of eligible capital expenditure. Its portal states that new applications were accepted only up to 31 March 2024, so it matters now mainly for suppliers already approved under it.
• PLI for large-scale electronics manufacturing. Notified in 2020 around mobile phones and specified components, it rewards incremental sales of goods made in India. Confirm its current status with MeitY before assuming any benefit for your product.
• EMC 2.0. The modified Electronics Manufacturing Clusters scheme, notified in April 2020, funds clusters with plots, ready-built sheds and common facilities.
• India Semiconductor Mission. Semicon 1.0 carried an outlay of ₹76,000 crore. The ISM site records Cabinet approval of Semicon 2.0 in July 2026 with an outlay of ₹1,27,500 crore across six pillars, with notifications for all six issued in August 2026.

For a buyer, the practical question is usually whether a supplier's capacity or price depends on a scheme. If it does, ask what happens to the price if the incentive is delayed or disallowed. Take specialist legal and tax advice before building an award on an assumed benefit.

Choose locations on evidence

A cluster address does not prove capability. Score candidate locations on supplier and sub-tier density, engineering and operator skills, power quality, port and airport access, nearby test labs and calibration, distance to your customers and service teams, state approvals, and the room to add floor space and labour. Add disaster and concentration risk: putting every critical supplier in one district swaps one exposure for another.

Trade shows help here. A visit to electronica India in Bengaluru lets you see several regional suppliers in two days, but treat every stand conversation as an unverified claim until the site visit.

Pre-qualify, then develop through gates

Send every candidate the same short set of questions:

1. Which operations happen in-house at the proposed site?
2. Which critical materials and sub-processes are imported or subcontracted?
3. What equipment, capacity and yield data support the quote?
4. Which certifications cover this exact site and scope?
5. How are sources, substitutions and traceability controlled?
6. What tooling, NRE, MOQ and ramp assumptions sit behind the price?
7. How are product changes notified, and what is the continuity plan?

Then move each supplier through a fixed sequence: identity and capability screen, NDA and a controlled technical package, a comparable RFQ with a should-cost review, a process audit, first-article approval, a pilot build, limited production release and an ongoing scorecard. Agree drawing revision, acceptance criteria, approved sub-tiers, tooling ownership and remedies before the pilot, not after the first rejection. The electronics supplier due diligence guide goes deeper on the identity and site checks.

Localise in waves

A wave plan keeps the launch safe while local share grows. Each wave closes only when its evidence exists.

The ICEA and ICRIER study Globalise to Localise makes a related argument at national level: domestic value addition grows when manufacturers first reach export scale and a competitive base of ancillary suppliers develops, rather than when localisation is forced early. At product level the lesson is similar. Keep a qualified global source wherever risk justifies it; the multi-supplier strategy guide explains how to decide.

Run it as a programme for twelve months

Months one and two clean the data, confirm demand and imported content, and pick a balanced first portfolio. Months three and four are for discovery, RFIs and cluster visits, ending in a capability-gap map covering equipment, materials, test, certification and working capital. Months five to seven issue comparable RFQs, complete audits and agree who funds and owns tooling. Months eight to ten run pilots under realistic conditions and close corrective actions. Months eleven and twelve release controlled volume, recalculate domestic value from actual transactions and add the next wave.

Review monthly: domestic value by the agreed method, landed cost and purchase price variance, yield and on-time delivery, source and site concentration, imported sub-tier exposure, inventory, and incentive benefit realised against what was assumed. Report projects by stage (screened, quoted, audited, sampled, piloted, released, benefit realised) so projected savings are never reported as achieved localisation.

Where localisation programmes usually go wrong

The common failures are predictable. A trader is selected when the goal was an Indian manufacturer. A part moves on unit price before qualification. A safety-critical change goes in without validation. Tooling is transferred with no agreement on ownership or maintenance. All supply ends up in one domestic cluster. And a percentage is announced that nobody can reproduce from transaction data. Each of these is avoided by the same habit: decide on evidence, and write down how every number was calculated.