Operator Attrition in Indian EMS: What Each New Hire Really Costs

On a mid-size Indian EMS plant, an operator who quits does not cost you one month's wage. The replacement costs a recruitment fee, about a week of paid induction, and three to four weeks of slower output while she learns the station. On Noida-cluster wages that comes to roughly ₹20,000 per replacement in our worked model. At 4% monthly churn on 250 operators, that is about ₹24 lakh a year, and it appears nowhere in your accounts.

By the Mama Editorial Team. We scope camera projects for Indian factories. Every statute, scheme and wage figure below links to its primary source. Every plant-specific number is marked as an assumption so you can swap in your own.

Key takeaways

• PLI-era volumes are built with people. The PLI scheme for large-scale electronics manufacturing reports 1,85,175 direct jobs against a 2,00,000 target, and industry estimates put mobile manufacturing at around 12 lakh direct and indirect jobs (PIB, 1 April 2026).
• Short tenure is normal in blue-collar India. Deloitte's 2025 blue-collar workforce report found that 69% of blue-collar hires are temporary, with low permanent conversion and a short average tenure of 21 months (Deloitte India).
• Most of the cost is hidden in the ramp-up. In our model, the slow first month and the people supervising it cost more than the recruitment fee.
• Early leavers double-charge you. A hire who quits in week two has already used up the recruitment and induction spend, and then you pay it again.
• Churn works against the government's hiring incentive. PM-VBRY pays employers only for additional jobs sustained for at least six months (PIB, 15 August 2025). Hiring to fill ordinary exits does not raise your headcount, so it does not open a new claim.
• Cameras measure the station, not the person. They can show how long a station runs below rhythm after a change. They cannot tell you why someone left.

Why attrition is an EMS problem in 2026

Under the PLI scheme for large-scale electronics, production reached ₹11,01,813 crore, 136% of target, by February 2026 (PIB, 1 April 2026). The component scheme (ECMS) comes next: the same release expects its ₹1.15 lakh crore of investment commitments to generate around 1.4 lakh jobs.

For a mid-size EMS plant, customer volumes grow while bigger factories nearby recruit the operators you just trained. Wage pressure follows. In April 2026 Uttar Pradesh raised minimum wages as interim relief after labour unrest in Noida and Greater Noida. The state's own order records that roads were blocked and supply chains disrupted (UP order of 17 April 2026).

If the average blue-collar tenure is about 21 months (Deloitte India), and that held for your operators, a plant in steady state would be replacing very roughly half of them every year. That is our arithmetic (12 ÷ 21 ≈ 57%), not a published EMS benchmark. We found no reliable public figure for EMS operator attrition in India. Your HR register is the number that counts.

What one replacement actually involves

1. Finding the person. You pay a placement fee, a referral bonus, or recruiter time and walk-in days. Through a contractor, it sits in their margin.

2. Induction before they touch a board. An EMS operator cannot just be put on a line. ANSI/ESD S20.20, the ESD control standard many EMS customers audit against, requires initial and recurrent ESD training for all personnel who handle ESD-sensitive items, recorded in training records (Desco summary of S20.20 §6.1.2). Add safety, plant quality rules and your customer's workmanship standard, all on paid time. (For what ESD discipline looks like on the floor, see cameras in electronics/EMS plants.)

3. The ramp: slow, supervised output. This is the biggest and least visible cost. Productivity rises substantially with experience, and Argote and Epple's review in Science lists employee turnover, alongside organisational "forgetting", among the reasons learning rates vary so much between organisations (Argote & Epple, 1990). Each person who leaves takes their share of the learning with them. During the ramp, the new operator works below cycle time, a line leader or senior operator spends time beside her, and the station downstream waits.

4. The gap before the replacement arrives. Until the new hire starts, someone covers on overtime, the line runs a station short, or a floater is pulled from elsewhere.

The formal skilling system shows how deep the job really is. The Electronics Sector Skills Council's qualification for a PCB Assembly Operator (ELE/Q7804, NSQF Level 4) sets a 600-hour course (ESSCI facilitator guide). No plant runs a 600-hour induction for each replacement. In practice the shortfall gets made up on the line, on your output.

The ₹ model: one replaced operator

Here is a worked example for a mid-size EMS plant in Gautam Buddha Nagar (Noida). The wage anchor is the UP Labour Department's interim-relief order No. 374/36-2-2026 of 17 April 2026, effective 1 April 2026: Category I (Gautam Buddha Nagar and Ghaziabad) minimum wages are ₹13,690/month unskilled, ₹15,059 semi-skilled (₹14,173 basic + ₹886 VDA) and ₹16,868 skilled (UP order, English translation). The variable DA is revised periodically, so check your current rate before you use these figures.

Assumption: an operator paid a little above the semi-skilled minimum costs about ₹20,000/month fully loaded, including PF/ESI, bonus provision, canteen and transport. That is about ₹770 per day over 26 days, or ₹96 per hour over 208 hours; her wage proper (basic + DA) is about ₹15,000, or ₹72 an hour. Every line below is an assumption. Replace each one with your own figure.

That is about one month of the operator's loaded cost, spent before she produces a full day's work. The recruitment fee, which most owners picture, is about 15% of it.

The plant-level number

The early-leaver multiplier

Suppose (an assumption) that one in three new hires leaves within the first month. A hire who leaves around day ten has already used about ₹9,000 of the budget above: recruitment, induction, trainer share and some of the ramp. To end up with one operator who stays, you make 1.5 hires on average. That is half an early leaver (about ₹4,500) on top of each full replacement, so each retained operator costs about ₹24,300, and the annual bill in the model rises to roughly ₹29 lakh.

So the cheapest fix is often first-month retention: a named buddy, a slower first week, a station chosen for the newcomer rather than whatever gap is open.

The PM-VBRY angle: only net new jobs count

The Pradhan Mantri Viksit Bharat Rozgar Yojana (PM-VBRY) pays employers up to ₹3,000 a month for two years for each additional employee whose employment lasts at least six months. In manufacturing, it extends into the third and fourth years (PIB, 15 August 2025). The amount depends on the additional employee's EPF wage: up to ₹1,000 for EPF wages up to ₹10,000 (proportional), ₹2,000 for more than ₹10,000 and up to ₹20,000, and ₹3,000 above that, up to a salary of ₹1 lakh. An establishment with 50 or more employees must add at least five additional employees, sustained for six months, to qualify (PIB, 1 July 2025). The benefits apply to jobs created between 1 August 2025 and 31 July 2027 (PIB, 25 July 2025).

For a churning plant: filling an ordinary exit does not take you above your EPFO baseline, so it opens no new claim, and a new hire who leaves before six months does not qualify. If an employee you are claiming for leaves, the scheme's operating procedure, as summarised by IMPRI, lets the incentive continue for the remaining period once an eligible replacement joins, but you still pay the ramp in the table above. If you are hiring 40 new operators for a PLI customer, first-six-month retention decides how much you collect. Ask your payroll advisor how EPFO calculates your baseline.

What a camera can and cannot tell you

It can show the cost of a ramp at the station. A camera on a bench row or a manual insertion line can measure how long each station runs below the line's rhythm, how much the next station waits on it, and how many minutes a station sits unmanned between an exit and a replacement. Mark the date a new operator joined a station, and after a few weeks you can see how many working days it took that station to get back to rhythm. That turns the "ramp" line from a guess into a measurement. The downstream waiting is the blind-zone loss we cost in where a mid-size EMS line loses hours.

It cannot tell you why people leave. Wages, commute, hostels, a supervisor, a better offer: none of it is on video.

It cannot judge skill or workmanship. A ceiling camera cannot see a solder joint or check an ESD wrist strap test log. AOI, your test jig and your quality team do that.

It should not track individuals. Use aggregate, station-level counts (minutes below rhythm, minutes unmanned), not an operator league table. Ranking named new joiners on video would speed up the attrition you are trying to fix. Under India's DPDP regime, workers should get a clear notice of what the cameras record and why (worker CCTV notice under DPDP). For the related question of who is actually at the line versus on the muster, see contract labour headcount.

Where Mama fits

Mama reads the cameras you already have, or a few we add, and every morning sends the owner a short WhatsApp note on where yesterday's hours went. For example: which stations ran below rhythm, for how long, and how much waiting that caused. It watches machines and stations, reports aggregate counts, and does not name or rank operators. To see what that would look like on your line, send a short phone video of your floor and we will come back with where hours are leaking and where a camera would need to sit to measure them.

How to put your own number on it

• Pull twelve months of exits from the register. Split them into "left within 30 days", "left within 6 months" and "longer". The first bucket is often the cheapest to fix.
• Time one ramp honestly. For the next three joiners, record what their station produces each day until it reaches the line standard. That gives you your own ramp line instead of our 20-day assumption.
• Rebuild the table with your figures. If it comes to more than a month of loaded wages, retention is worth money, and you can show how much.

FAQ

What does it cost to replace one EMS operator in India? In our worked model on Noida-cluster wages, it costs about ₹19,800 per replacement, or about one month of fully loaded cost. Most of that is the slow first month and supervision, not the recruitment fee. Every line is an assumption, so rebuild it with your own wages and ramp time.

What is a normal attrition rate for EMS operators? We could not find a reliable published figure specific to Indian EMS operators. Deloitte's 2025 report puts average blue-collar tenure at 21 months. If that applied to your operators, roughly half the workforce would turn over each year. Your own 12-month exit register is the benchmark that matters.

Does PM-VBRY cover replacement hires? Not as new claims. The employer incentive is for additional employment above your EPFO baseline, sustained for at least six months, so filling an ordinary exit does not open a new claim, and a new hire who leaves before six months does not qualify. If an employee you are already claiming for leaves, the scheme's operating procedure allows the incentive to continue for the remaining period once an eligible replacement joins. Confirm the details with your payroll advisor or EPFO.

Can cameras reduce attrition? Not directly. They can measure what attrition costs at the station, such as ramp days, unmanned minutes and downstream waiting, so you can see which fixes pay back. Used to watch individuals, they are more likely to increase attrition than reduce it.