How to choose what to manufacture

Most people decide what to manufacture by finding a product that interests them and then working out whether the numbers support it. The order that works is the reverse: establish what you can sell, to whom, at what margin, with the capital you have — and let that narrow the list of products rather than starting from one.

Getting this wrong is expensive in a particular way. By the time it becomes obvious, the machinery is bought.

What actually constrains the decision?

Four things, and they narrow the field faster than any amount of market research.

Capital. Not what you have, but what you have after keeping enough working capital to survive the first year. Promoters routinely commit everything to plant and then discover they cannot fund raw material and receivables until the first payments arrive.

Access to buyers. The single most underweighted factor. If you already sell to a customer base, products they buy are dramatically easier than products you would have to find new buyers for. This is why traders moving into manufacturing succeed more often than newcomers — they start with demand.

Location. Power availability and cost, effluent treatment where the process needs it, distance to raw material, distance to market. For heavy or bulky products, freight can decide viability before anything else does.

What you can supervise. A process you cannot personally assess means depending entirely on a technical person, which is workable and raises the risk considerably.

How should candidates be compared?

On the same handful of figures, at the same level of detail, before falling in love with any of them.

What to establish	Why it decides things
Total project cost	Whether it is within reach at all
Working capital cycle in days	How much cash is tied up before you are paid
Gross margin at realistic prices	Whether there is room for anything to go wrong
Break-even utilisation	How full the plant must run to stop losing money
Who the incumbents are	Whether you are entering a crowded market
Raw material availability and volatility	Whether your margin is yours or your supplier’s
Regulatory approvals required	How long before you can produce anything

Break-even utilisation is the most useful single number and the least often calculated. A project that breaks even at 40% of capacity has room to be wrong about demand. One that breaks even at 75% has almost none — and demand forecasts for a new unit are usually wrong.

Where do people go wrong?

Choosing a product because the margin looks high. High margins attract entrants, so either there is a barrier you have not identified, or the margin is not what the published figure suggests, or it is about to fall.

Believing they will sell at list price. A new entrant with no track record generally sells below the established price, at least at first. A model built on prevailing prices has no allowance for the discount required to win a first customer.

Underestimating the working capital cycle. Buying raw material for cash and selling on 60-day credit means funding two months of turnover indefinitely. In several trades that is a larger number than the machinery.

Assuming a subsidy will arrive. Incentive schemes are real and worth pursuing, and they are also slow, conditional and subject to change. A project that only works with the subsidy is a project that does not work.

Choosing a process that needs one person. If the whole operation depends on a technician you have just hired, you have a staffing problem disguised as a manufacturing business.

Is it better to make something you already sell?

Usually, and by a wide margin.

A dealer or distributor moving into manufacturing starts with three advantages that a newcomer spends years acquiring: they know what actually sells and in what quantity, they have customers before the plant is commissioned, and they understand the price the market really pays rather than the list price.

The risk to be honest about is channel conflict. Manufacturing a product you also distribute for others puts you in competition with your suppliers, and they will notice. That is a manageable problem, but it should be a decision rather than a surprise.

What should you do before committing?

1. Shortlist three or four products, not one. Comparison surfaces things a single option never does.
2. Cost each to the same depth — machinery, infrastructure, working capital, approvals.
3. Talk to actual buyers before believing any market estimate. Five conversations beats any report.
4. Visit an existing unit making something similar. An hour on a shop floor corrects more assumptions than a month of desk research.
5. Calculate break-even utilisation for each and discard anything above a level you would be comfortable missing.
6. Then write the project report for the survivor.

StartYourIndustry was built for the first two steps — project profiles with investment breakdowns, demand data and supplier directories, so comparing options does not mean starting each from nothing.

Common questions

How much of my own money should go in?

Banks expect a meaningful promoter contribution, and the proportion varies by scheme and by bank. More useful as a planning question: keep enough outside the project to fund the first year of working capital and a delay, because both are likely.

Should I buy used machinery?

It lowers project cost and raises operating risk, and some lenders and schemes restrict it. If you cannot personally assess the condition of a used machine, the saving is not worth what it exposes you to.

How long from decision to production?

Longer than planned, in almost every case. Approvals, power connections and machinery delivery all slip. Build the delay into the financial model rather than discovering it after interest has started accruing.

Is a smaller plant safer?

Lower capital at risk, but often a worse cost per unit — so it can break even at a higher utilisation than a larger plant. Compare break-even utilisation rather than assuming small is safe.

More on this: preparing a DPR and what a CMA report is. See also StartYourIndustry.