Market entry evaluation, for a unit running 20 to 200 workers, is not a sector study. It is the discipline of checking whether one specific stream of work — a named customer, a named part, a named volume band — can pass through your machines, your cash cycle and your compliance file without damaging what you already run. On a shop floor the phrase has to mean things you can count: contribution per machine hour, the qualification cycle (trials, first article, PPAP), working capital locked in receivables, statutory readiness (GST, IEC, BIS, factory licence scope), utilisation on specific machines. For a Ludhiana turning shop or a Rajkot fabrication unit, this evaluation is worth more than any market report, because a small manufacturer cannot average out a bad decision across a portfolio. One wrong yes can occupy a quarter of your capacity for eighteen months and pay for it out of your overdraft.

Here is how the wrong entry usually happens. A purchase manager you met at a trade fair calls on a Tuesday. The drawing arrives on WhatsApp. You draw a sample on Saturday, quote on Sunday, and cut the price twice before the first order lands. Six months later the shop is jammed with work that earns less per hour than your old mix, the tooling was “shared” but never paid for, payments sit at ninety days, and the customer who always got his Friday dispatch is now buying from your competitor. Nobody took a bad decision. A decision was simply never taken — it drifted in. The fix is cheap: nine checks, written down, before the first quotation goes out.

The fix: nine checks before the first quotation

1. Write the market down as a product, not a sector

If you cannot write the market in one sentence — part name, customer type, geography, volume band, payment reality — you are not evaluating a market, you are daydreaming about one. “Industrial packaging” is not a market; “mono-cartons for three pharma packers around Baddi, 40,000 units a month, 45-day payments” is. Sector size figures are for large players who can chase share. Your market is bounded by your machine list, your cluster’s subcontract chain and your working capital. Put the sentence in writing, because every check that follows tests that sentence, not the idea in your head.

2. Run the contribution-per-hour test, not the per-piece test

Price per piece tells you almost nothing. What a machine hour earns after material is the only comparison that holds across enquiries. Two jobs on the same VMC, material supplied by the customer in both cases: one at ₹18 a piece on a three-minute cycle, one at ₹12 a piece on a 75-second cycle. The first earns ₹360 an hour; the second earns ₹576. Small units quote per piece or per kilo because that is how buyers ask, and that habit is how good machines end up earning less than a lathe operator’s wage.

Work out your loaded shop rate honestly — operator, power, consumables, depreciation, overhead. A mid-size turning centre in a tier-2 cluster typically runs ₹450–550 an hour loaded. If what the machine earns after material does not clear that rate with something left over for the bad months, the order is filling the shop, not feeding it.

3. Check the new work against your load, not your nameplate capacity

Capacity is not a number on a brochure. It is which machine is free in which shift, at what utilisation, with which operator. The question is not “can we make it” but “which hour will it occupy”. If the new market uses the VMC that stands idle after lunch, good. If it pushes your current best-paying part into a second shift with a 20 per cent wage premium and thinner supervision, the new market must clear that premium inside its own contribution numbers — not inside your enthusiasm.

Check the edges too: sanctioned power load, grinding and heat-treatment queues in the cluster, and every process you would borrow from a subcontractor. A market that needs a process you do not control is a market you serve at someone else’s convenience.

4. Price the qualification cycle before you price the part

Every new market has a cover charge: trials, first articles, PPAP, customer audits, tooling and — for auto OEM work — the IATF 16949 certificate most tier-1s will not discuss without. Count it as a cost with a recovery date, not as an investment you will “recover somehow”. My working rule: the full qualification spend, tooling included, should recover within twelve months of steady volumes. If the buyer will not put volumes in writing, treat that spend as a donation and decide accordingly. And ask who owns the tooling from day one. “We will pay after the first order” is not ownership; it is an option on your balance sheet.

5. Stress the receivables, not just the order value

A ₹40 lakh order on 90-day terms is, in steady state, a ₹10–12 lakh interest-free loan to a stranger, funded from your overdraft. Run peak working capital for the market at full ramp — sixty to ninety days of billing at material plus conversion, safety stock, and tooling — then check it against your actual cash line, not the one you wish you had.

For supplies to larger buyers, quote under the MSMED Act’s 45-day payment provision, keep your Udyam registration current, and learn your way around the MSME Samadhaan delayed-payment portal before you need it. For exports, the payment instrument matters more than the country: a confirmed LC reads very differently from a DA-90 with a first-time buyer. If the market only works when payments behave, the market does not work.

6. Check the input chain before you trust the customer

You do not enter a market alone. Your material supplier, heat treater and plater enter it with you, and they are usually the reason entries stall. Material grade with proper mill test certificates, availability inside the cluster, plating and heat-treatment subcontractors whose own paperwork survives an OEM audit, insert and tooling lead times — list each one against the new part. A Coimbatore pump maker I know moved into export castings and found the bottleneck was not the foundry; it was the certificates the foundry had never been asked to produce. If the inputs need certification the cluster does not have, either price that certification into the part or stay out.

7. List the statutory load before it lists you

New markets arrive with forms attached. List them before the first dispatch, not after the first notice. Interstate supply: GST registration and e-way bill discipline. Exports: the Import Export Code from DGFT, AD code registration with your bank, LUT for zero-rated supplies. An added shift: factory licence amendment, PF and ESI on the new headcount, contract labour licence thresholds in your state. Product-level: BIS marking where the product falls under mandatory certification. Each item is small; the pile is not. Assign one person, one file and one date per item, and do not send the quotation until the file exists.

8. Count concentration and the exit cost

Ask the uncomfortable question first: if this market goes away in year two, what exactly do you walk away with? Dedicated gauges and fixtures, dies whose ownership nobody wrote down, operators trained on tolerances nobody else pays for, a second shift you hired. If one buyer will hold more than a third of your turnover after entry, you have not entered a market — you have changed jobs and handed the employer the right to re-price you every year. Exclusivity demands without volume commitments belong in the same bin.

9. Take a pilot lot with a written review date

Never convert a market on the strength of the first order. Enter on a pilot: a defined quantity, sixty to ninety days, and a review date in writing. During the pilot, measure actual contribution per hour, actual rejections and actual payment days — not the ones you assumed. A rate contract follows only after three consecutive lots at full rate behave. If the buyer resists a review date, that resistance is itself data.

A worked example: the Ludhiana shaft enquiry

A 60-worker turning shop making aftermarket sprockets and shafts gets a call from a tier-1: quote for an OEM transmission shaft, 1.2 lakh pieces a year. On paper it is flattering. On the nine checks it looks like this.

The price is ₹118 a piece against ₹170 for a similar aftermarket part, but the cycle is tighter and volumes are steady, so contribution works out near ₹540 an hour against the current mix of ₹430 — check two passes. The volume means about 8,000 machine hours a year across three turning centres, and the grinding end needs a second shift, so the ₹540 must carry a 20 per cent shift premium and still clear ₹430 with margin — it does, barely. Qualification — PPAP, CMM, tooling, trials — comes to about ₹6.5 lakh, which full volumes repay inside three months, well inside the twelve-month rule; the check passes only if the volumes are in writing. Sixty-day payments on roughly ₹12 lakh a month of billing lock about ₹24 lakh at peak, which the unit’s cash line cannot carry without bridge finance — this check fails unless payment is quoted at 45 days under the MSME provision. Checks six and seven pass: the cluster has material with certificates, and the licence file already covers a second shift.

Decision: enter on a 5,000-piece pilot, review at ninety days, with a pre-agreed exit if payments slip past sixty days twice. The market was worth entering — at the terms, not at the flattery.

When the numbers say yes — and when to walk away

Enter when most of these hold:

• Contribution per machine hour is at least 15 per cent above your current mix at year-one prices, not promised year-two volumes.
• The work uses named idle hours — the machine that stands afternoons — rather than hours taken from your best-paying part.
• Qualification cost, tooling included, recovers within twelve months of steady volumes.
• Inputs — material with certificates, heat treatment, plating — sit inside the cluster.
• The statutory additions are ones you would have carried anyway.
• The buyer accepts a pilot with a review date without drama.

Walk away, politely and early, when you see these:

• Free samples on your tooling before any contract or rate discussion.
• “Payment as per our policy” stretched past ninety days, with the policy unwritten.
• Penalty clauses — line stoppage, air freight — with no cap and no reverse obligation.
• An annual price-down demanded on a market you just paid to enter.
• Exclusivity demanded in the cluster with no volume commitment attached.
• Trial orders that never become rate contracts, year after year.

The one-page go/no-go note

Every check above folds into one page. Market in one sentence. Annual volume band. Landed price per piece. Cycle time and the machines it occupies. Contribution per hour against the current mix. Qualification cost and recovery month. Peak working capital and the cash line it must fit inside. Payment terms in days. Statutory additions with an owner and a date. Turnover concentration after entry. Review date. Family sign.

Two failed checks out of nine, and the entry is parked for a year — not rejected, parked. Markets come back; sunk cash does not. This is not caution for its own sake. It is the difference between a decision you can defend at the bank and a story you will be retelling at the next trade fair.

In the next piece, I will break down how to cost a pilot lot without losing money on it, because most units lose money on the trial and then blame the market. If you run a 20–200 worker unit with a live enquiry in hand, fill the sheet, note where your numbers surprised you, and write in — I will take one reader’s enquiry and work it through in public, numbers included.

Frequently asked questions

How long should a trial order run before I commit capacity?

Sixty to ninety days, or three consecutive lots at full supply rate — whichever comes later. Until then, the market gets surplus hours, not dedicated ones. Convert to a rate contract only after the pilot’s actual numbers — contribution per hour, rejections, payment days — match what you quoted. Never allocate machines before that.

Is a bigger market always better than a smaller profitable one?

No. The size of the market is not your size; your share of it has to fit your capacity band. A 60-worker unit chasing a market whose minimum order quantity demands two full shifts is entering a market designed for someone else. A small market where you are the natural second source, with inputs in the cluster and payments that fit your cash line, beats a large one where you are supplier number forty-one.

What is the single number that should decide entry?

Contribution per machine hour after material and statutory costs, compared against your current mix. Order value flatters you, price per piece misleads you, and the sector’s annual turnover is irrelevant to a unit with eleven machines. One number, one comparison, honestly loaded — it settles most arguments faster than any consultant’s report.

What statutory items should I verify before quoting outside my state or country?

Interstate: GST registration, place-of-supply treatment and e-way bill discipline. Exports: Import Export Code, AD code registration with your bank, LUT for zero-rated supplies, and the buyer’s payment instrument in writing. If headcount rises: PF and ESI coverage, and contract labour licence thresholds in your state. If shifts change: factory licence amendment. Where the product falls under mandatory certification, confirm BIS scope before quoting, not after the order.