Buying a polishing line is a capital decision, and capital decisions are usually justified with a payback calculation. The problem is that most of those calculations are built on the purchase price and a hoped-for volume, and they leave out the three variables that decide the outcome: utilisation, yield and consumable cost.
This is a method for getting a number you can defend in front of a finance committee, and for knowing which assumption would have to be wrong for the decision to change.
The four numbers you actually need
- Current cost per square metre — what you pay today, whether that is an external finishing supplier, a subcontractor, or an internal process whose real cost you have never separated out.
- In-house cost per square metre — labour, abrasive, power, extraction, maintenance, floor space and depreciation, divided by the square metres actually produced.
- Capital cost — the full installed cost, not the machine price. Freight, foundation, power connection, extraction ducting, commissioning and training all belong in the number.
- Realistic utilisation — the share of available hours the line will actually run, after changeovers, maintenance and the weeks when there is no work for it.
Three of those four are usually estimated badly, and utilisation is the worst of them. A line quoted at 6,000 hours a year of theoretical capacity will not run 6,000 hours. It will run whatever your order book allows.
The payback formula, and why it is simpler than it looks
Payback in months equals installed capital divided by monthly net saving. Monthly net saving equals the square metres finished per month multiplied by the difference between the outsourced cost and the in-house cost per square metre, minus any fixed costs you have added.
That is the whole calculation. The difficulty is not the arithmetic; it is being honest about the inputs, particularly the ones that make the answer look worse.
A worked example with stated assumptions
The following is illustrative only — substitute your own numbers. Assume a shop that currently outsources 20,000 square metres of No.4 finishing per year, pays an external rate equivalent to 4.00 per square metre, and is quoted an installed line cost of 260,000.
Assume in-house variable cost of 1.60 per square metre — abrasive, power, labour directly on the line, and extraction. Assume 15,000 square metres of annual throughput in the first year while the order book adjusts, and 1,200 per month of fixed cost for space, maintenance and supervision.
Annual saving on variable cost is 15,000 multiplied by 2.40, which is 36,000. Subtract 14,400 of fixed cost and the net annual saving is 21,600. Payback is 260,000 divided by 21,600, which is about twelve years. On those numbers the investment does not make sense.
Now change one assumption: throughput rises to 30,000 square metres because the line also takes work currently sent to a second supplier. The variable saving becomes 72,000, net saving 57,600, and payback drops to about four and a half years. Same machine, same price, completely different decision.
That is the point of the exercise. The purchase price is fixed; utilisation is the variable that decides everything.
The three variables that kill payback
Utilisation below the assumption
Capital cost is fixed and spread over output. If the line runs at half the assumed hours, the cost per square metre roughly doubles and the payback period roughly doubles with it. Before signing, ask what share of the order book is committed rather than expected.
Yield loss
A new line does not start at the yield of a mature process. Scrap and rework in the first months are normal, and they are a real cost. A line that produces 3 percent scrap at a finished value of 20 per square metre is losing 0.60 per square metre — which can be a large fraction of the entire saving.
Consumable cost per square metre
Abrasive is the largest recurring cost on most polishing lines, and it varies more between machines than buyers expect. Belt life depends on head design, pressure control, cooling and the abrasive specification. A machine that consumes 15 percent more abrasive than a competitor can erase the price advantage in under two years.
What to model beyond the base case
Build three scenarios rather than one: a base case at your committed volume, a downside at 60 percent of it, and an upside at 130 percent. If the decision survives the downside, it is robust. If it only works at the upside, it is a bet rather than an investment.
Then test the two cost assumptions that are least reliable: the external rate you will be quoted in two years, and the in-house labour cost. Outsourced finishing prices move with capacity in the market, and in-house labour rarely gets cheaper.
Non-financial reasons that legitimately change the answer
- Lead time — in-house finishing can compress a four-week external turnaround to days, which matters if your customers compete on delivery.
- Quality control — you cannot inspect what you do not control. For cosmetic work, the ability to adjust a head and re-run a batch has value that does not appear in the payback model.
- Confidentiality and product development — new surface development is hard to do through a subcontractor.
- Capacity risk — a single external supplier is a single point of failure, and its capacity is not yours to prioritise.
These are legitimate, but they should be stated separately rather than used to paper over a payback that does not work. A finance committee will accept a strategic argument; it will not accept arithmetic that has been quietly bent.
A short checklist before you commit
- Installed cost, not machine price, in the numerator
- Committed volume, not hoped-for volume, in the denominator
- Consumable consumption per square metre from the supplier, in writing
- Yield assumption stated explicitly, with a first-year allowance for ramp-up
- Three scenarios modelled, and the downside still acceptable
- Non-financial reasons listed separately