Industrial CAPEX decision guide
Before You Invest $1 Million: Fix, Expand, or Build New?
A practical framework for plant owners and project teams: measure the outcome gap first, find the real constraint, then compare three options on the same numbers — before any supplier is asked for a price.
Fix (optimise or retrofit) when one constraint limits output and the rest of the operation has headroom. Expand when the operation is balanced but at its ceiling and the site, utilities and market can carry more. Build new when the product, process, site or required capacity has changed beyond what the existing asset can deliver. Decide by comparing all three on the same KPI gap, CAPEX, downtime, operating cost, risk and sensitivity — then write one RFQ for the chosen route.
“$1 million” is an illustrative headline figure only — not a minimum, a quote or a typical price. Global B2B Group reviews industrial projects from USD 250,000, and the buyer service is free.
1. Define the outcome KPIs and baseline
Write down what must change, how you measure it today and where it has to be. Use your own data from a representative period that includes changeovers and seasonal peaks. Leave cells blank rather than guess.
| KPI | Baseline (your data) | Target | How to measure |
|---|---|---|---|
| Saleable output per week | — | — | Packed, released units or tonnes, not machine counts |
| OEE of the constraining stage | — | — | Availability × performance × quality (ISO 22400-2 definitions) |
| Scrap, rework or giveaway | — | — | % of input by weight or units, per shift |
| Energy per unit | — | — | kWh (or fuel) per tonne / unit from sub-meters |
| Labour hours per unit | — | — | Paid hours ÷ saleable output |
| Contribution per production week | — | — | Revenue minus variable cost, from finance |
Takeaway: No baseline, no business case — every later number depends on these six cells.
2. Diagnose five constraints before choosing a route
An investment only pays if it removes the constraint that actually limits the outcome. Check all five; most plants have one dominant constraint at a time.
| Constraint | Question to answer | Usually points to |
|---|---|---|
| Equipment / bottleneck stage | Which single stage limits output, and is it speed, availability or quality loss? | Fix / retrofit |
| Utilities | Can power, steam, refrigeration, air, water and effluent carry more output? | Expand (with uprating) or build new |
| Space and site | Is there physical room, floor loading and permitted use for more equipment? | Expand or build new on another site |
| People and skills | Are losses caused by staffing, training or maintenance capacity rather than machines? | Fix (often non-CAPEX) |
| Market and product | Is demand proven for the extra volume, and does the product specification change? | Expand if proven; build new if the product changes |
Takeaway: If the binding constraint is people or maintenance, spending on equipment rarely closes the gap.
3. Quantify the gap to target
Convert the KPI gap into money per year, so the three options can be compared against the same requirement. Use contribution (revenue minus variable cost), not revenue.
An option that does not close the gap is not a cheaper version of the right answer — it is a different, smaller project.
Required annual gain = (target output − baseline output) × contribution per unit + avoided scrap, energy and labour cost per year
Takeaway: State the required annual gain before you look at any option's price.
4. Compare the three options on the same criteria
The pattern below is qualitative and general. Your own numbers replace it in the worked example and calculator.
| Criterion | Fix / retrofit | Expand | Build new |
|---|---|---|---|
| CAPEX | Lowest of the three | Moderate | Highest |
| Capacity ceiling | Limited by headroom of other stages | Step change within site limits | Set by the new design |
| Downtime on current operation | Short planned shutdowns | Phased tie-ins, often the longest | Mostly off-line until cut-over |
| Quality / specification change | Little | Some | Full re-specification possible |
| Main risk | Interfaces with ageing equipment | Line balance and utility uprating | Execution, permits, commissioning, ramp-up |
| Time to benefit | Shortest | Medium | Longest |
Takeaway: Downtime and ramp-up belong in the cost — they are where retrofit and expansion cases most often go wrong.
5. Worked illustrative example (hypothetical values)
A hypothetical plant earns USD 20,000 contribution per production week and needs USD 350,000 more annual gain to reach its target. All values are invented for illustration — they are not market benchmarks, quotes or expected savings. Horizon: 10 years, undiscounted.
| Fix / retrofit | Expand | Build new | |
|---|---|---|---|
| CAPEX (USD) | 300,000 | 1,000,000 | 2,500,000 |
| Downtime weeks × 20,000 | 2 → 40,000 | 6 → 120,000 | 1 → 20,000 |
| Annual gain (USD) | 150,000 | 400,000 | 700,000 |
| Added annual OPEX (USD) | 10,000 | 80,000 | 150,000 |
| Net annual (gain − OPEX) | 140,000 | 320,000 | 550,000 |
| Payback = (CAPEX + downtime) ÷ net | 2.4 years | 3.5 years | 4.6 years |
| 10-year net = net × 10 − CAPEX − downtime | 1,060,000 | 2,080,000 | 2,980,000 |
| Gain −20 %: payback / 10-year net | 3.1 y / 760,000 | 4.7 y / 1,280,000 | 6.1 y / 1,580,000 |
| Closes the 350,000 gap? | No | Yes | Yes |
- Fix has the fastest payback but does not reach the target — it is only the right answer if the target can be lowered.
- Build new has the largest 10-year net but the highest outlay, the longest payback and the largest exposure if the gain falls 20 %.
- Expand closes the gap with a moderate outlay; whether it beats build new depends on risk tolerance, capital availability and how certain the extra demand is.
- A discounted (NPV/IRR) view would narrow the gap between options with early and late benefits; use the payback/IRR tool for that.
Takeaway: The fastest payback and the biggest total are rarely the same option — check gap closure first, then economics, then risk.
Try it: compare your three options
Enter your own estimates. Calculation runs in your browser only; nothing is sent or stored.
Downtime cost = weeks × weekly contribution. Net annual = gain − added OPEX. Payback = (CAPEX + downtime cost) ÷ net annual. Horizon net = net annual × years − CAPEX − downtime cost. Undiscounted screening only — no tax, inflation, financing cost or residual value.
6. Decision tree
- 1. Is the binding constraint people, maintenance or method? → Fix it first (often little or no CAPEX), then re-measure.
- 2. Is one equipment stage the constraint and do the others have headroom that covers the gap? → Fix / retrofit.
- 3. Is the operation balanced at its ceiling, with site, utilities and proven demand for more? → Expand.
- 4. Has the product, process, hygiene/safety standard or required capacity changed beyond the asset? → Build new.
- 5. Do two options both close the gap? → Choose on stressed payback, downtime exposure and risk score, not on CAPEX alone.
- 6. Is the chosen scope below USD 250,000? → It may still be a sound investment, but it is outside Global B2B Group's project scope.
Takeaway: Follow the constraint, not the budget you hoped to spend.
7. Decision scorecard
Score each option 1–5 per criterion, multiply by the weight and add up. The weights below are a starting point; agree your own before scoring.
| Criterion | Weight | Fix | Expand | Build new |
|---|---|---|---|---|
| Closes the KPI gap | 25 % | |||
| Stressed payback / horizon net | 20 % | |||
| Downtime and ramp-up exposure | 15 % | |||
| Technical and execution risk | 15 % | |||
| Quality / specification fit | 10 % | |||
| Utilities and site fit | 10 % | |||
| Financing readiness | 5 % |
Any option that fails “closes the KPI gap” should be rejected or re-scoped, whatever its total.
Takeaway: Agree weights with operations, finance and quality before the numbers come in.
8. Buyer checklist before the RFQ
- Six KPI baselines filled from your own data, with the period stated.
- Binding constraint identified and evidenced (stage data, utility logs, staffing).
- Required annual gain calculated in contribution, not revenue.
- All three options costed including downtime, ramp-up and added OPEX.
- Sensitivity run at −20 % gain (and higher CAPEX if quotes are uncertain).
- Utilities, permits and site constraints checked for the chosen route.
- Acceptance criteria (FAT/SAT) written as measurable KPIs.
- Payment milestones and financing pathway outlined.
Takeaway: If you cannot tick all eight, the RFQ will return quotes you cannot compare.
9. Payment milestones and financing readiness
Tie payments to verifiable progress: design approval, factory acceptance test (FAT), delivery, site acceptance test (SAT) and a performance test after ramp-up. A retention held until the performance test aligns payment with the outcome you are buying. Actual terms are negotiated with the chosen supplier.
Lenders and lessors typically ask for the same evidence this framework produces: baseline, gap, options compared, sensitivity and an acceptance plan. Global B2B Group can describe possible pathways through independent financing partners; it does not lend, and no approval or terms are promised.
- Avoid large unsecured advance payments; ask about advance-payment and performance guarantees.
- Retrofit and expansion: schedule payments around shutdown windows.
- Build new: include permits and utility connection in the milestone plan, not only equipment.
- Keep 5–15 % contingency in the budget until quotes are compared (a planning range, not a benchmark).
Takeaway: Milestones tied to FAT, SAT and performance protect both cash and outcome.
Tools to run your own numbers
These existing calculators give planning estimates, not quotations.
- CAPEX calculatorBuild up installed cost with installation, logistics and contingency.
- TCO calculatorLifecycle cost of an option over your horizon.
- Throughput calculatorCheck line capacity and the bottleneck stage.
- Project delay cost calculatorPrice downtime and late commissioning.
- Payback and IRR toolDiscounted view of the chosen option.
- Upgrade vs expand vs new lineQualitative guide for production lines specifically.
- Right outcome, right total costComparing supplier offers once the route is chosen.
- Procurement Project OSRFQ checklist, quote normalisation and FAT/SAT milestones.
Sector-specific calculations
This guide covers cross-sector decision criteria. Sector equipment and sizing are handled by the specialist group platforms:
Limitations
- All example values are hypothetical; they are not benchmarks, quotes or expected results.
- The calculator is undiscounted and excludes tax, inflation, financing cost and residual value.
- The output is a screening aid, not engineering, legal, tax or investment advice.
- This guide is AI-assisted editorial content prepared by the Global B2B Group project team; it has not been individually reviewed by an external expert.
Frequently asked questions
How do I decide between fixing, expanding or building new?
Identify the binding constraint and the required annual gain first. Fix when one stage limits output and others have headroom; expand when the operation is balanced at its ceiling with site and utility capacity; build new when product, process or capacity needs exceed the asset. Compare all three on the same gap, CAPEX, downtime, OPEX and risk.
Does a project need to reach the $1M headline figure?
No. The $1M figure in the title is only an illustrative headline. Global B2B Group's minimum industrial project size is USD 250,000.
Should I include downtime in the investment cost?
Yes. Lost contribution during shutdowns and ramp-up is a real cost and often decides between retrofit and expansion. Multiply weeks of lost production by contribution per week.
Is the option with the shortest payback the best?
Not necessarily. A short payback is irrelevant if the option does not close the KPI gap. Check gap closure first, then stressed payback and horizon net, then risk.
When is ROI or break-even meaningful?
Only when the baseline, gain and costs come from measured data or real quotations. With guessed inputs, use the result to rank options and identify what to measure, not to promise a return.
Can Global B2B Group tell me which option is right?
The project team can review your submitted data, help structure the comparison and prepare a supplier-neutral RFQ, then contact relevant suppliers on your behalf. The decision remains yours; no outcome is guaranteed.
Does the buyer pay for this service?
No. The buyer service is free.
Can you arrange financing?
We can outline possible pathways through independent financing partners. We do not lend, and no approval or terms are promised.
Sources and further reading
- ISO 22400-2:2014 — KPIs for manufacturing operations management (OEE definitions)
- E. M. Goldratt, The Goal (1984) — theory of constraints
- IEC 62381 — factory and site acceptance tests (FAT/SAT) for process automation systems
- ICC Uniform Rules for Demand Guarantees (URDG 758) — advance-payment and performance guarantees
Turn the chosen option into one comparable RFQ
Submit your baseline, gap and preferred route. Our team reviews and completes the RFQ, contacts relevant suppliers on your behalf and returns an organised comparison — free for buyers, for projects from USD 250,000.
Global B2B Group is a supplier-neutral managed procurement service. We do not manufacture equipment, guarantee supplier performance, publish real-time prices or approve financing. Estimates on this page are for planning only.
Read this guide in