Investment decision framework · FishMatch Group
Before you invest $1 million: fix, expand or build new?
The USD 1 million in the title is an illustration, not a minimum or a quote. This framework helps shrimp and fish farm owners decide whether to optimise what they have, expand it, or build a new RAS — before money is committed.
Short answer
Fix first when your records show a specific bottleneck — oxygen, water quality, biosecurity, feeding or infrastructure — that keeps survival, harvest weight or FCR below target. Expand when current units already reach target and demand supports more volume. Build new when the site, water or system cannot reach the target cost per kg at any reasonable retrofit cost.
Decide with numbers: baseline KPIs, the gap to target in tonnes and cost per kg, and for each option the capital cost, operating cost, downtime, added output and risk. Calculate payback only when the inputs come from your records or written supplier proposals.
FishMatch Group helps buyers with projects of at least USD 250,000 free of charge: our team reviews your request, contacts suitable suppliers for the chosen option and presents organised responses through FishMatch.
The 5-step decision framework
1. Set outcome KPIs and baseline
Survival, harvest weight, standing biomass, FCR, water-quality stability, energy per kg and cost per kg — measured over at least the last few cycles.
2. Diagnose the five constraints
Oxygen, water quality, biology and biosecurity, feeding, and infrastructure or capacity. Find which one limits results today.
3. Quantify the gap
Express the gap to target in tonnes per year and USD per kg — for example, target output minus current output, and target cost per kg minus current cost per kg.
4. Compare three options
Optimise/retrofit, expand, or build new: capital cost, operating cost, downtime, added capacity, product quality, risk and sensitivity to key inputs.
5. Decide, then specify
Choose the option, write its design basis and scope, and turn it into one supplier-neutral RFQ with acceptance tests and payment milestones.
Takeaway: the decision follows the diagnosis — never the other way round.
Diagnose the five constraints
Look for the constraint that binds first. Fixing a non-binding constraint costs money and changes little.
| Constraint | Typical symptoms | How to check | Option it usually points to |
|---|---|---|---|
| Oxygen | Night-time or post-feeding DO dips, feeding cut back, mortality at peak biomass | Continuous DO logs vs oxygen demand from daily feed | Retrofit: aeration/oxygenation, monitoring, backup power |
| Water quality | Ammonia or nitrite spikes, solids build-up, unstable pH or CO₂ | Daily tests vs species ranges; biofilter load vs feed | Retrofit: filtration, biofilter, degassing |
| Biology and biosecurity | Disease events, variable seed quality, size spread | Health records, seed source, mortality timing | Retrofit: biosecurity, nursery stage, seed sourcing |
| Feeding | High or rising FCR, uneaten feed, slow growth | Feed records vs biomass gain; feeding method | Retrofit: feeding control, feed review |
| Infrastructure / capacity | Units at target performance but no room to grow; site limits on water, power or land | Capacity of water, power, land and permits | Expand if site allows; otherwise new build |
Takeaway: name the binding constraint before pricing any solution.
Baseline questions to answer first
Answer these from your own records before comparing options.
Production
- Current output (t/yr), survival and harvest weight per cycle?
- FCR and cost per kg over recent cycles?
Water and oxygen
- Lowest recorded dissolved oxygen, and when?
- Ammonia, nitrite, CO₂ and temperature ranges?
Site
- Spare water, power and land, and permit limits?
- Energy cost per kWh and per kg produced?
Decision tree
1. Do your records show a binding constraint (oxygen, water quality, biology, feeding)?
✓ Model a retrofit of that constraint first.
✗ Go to the next question.
2. Do current units reach target survival, harvest weight and FCR?
✓ Expansion with the same proven design is a candidate.
✗ Return to diagnosis — expanding would multiply the problem.
3. Does the site have water, power, land and permits for more units?
✓ Model the expansion.
✗ Model a new build on a suitable site.
4. Can a retrofit or expansion reach the target cost per kg?
✓ Choose the lower-risk option with acceptable payback.
✗ Model a new build, for example a RAS, against the same targets.
Takeaway: most farms should answer the first two questions before talking to any builder.
Compare the three options on the same lines
Fill each cell with your own numbers or written supplier figures. The descriptions show what to compare, not typical values.
| Factor | Optimise / retrofit | Expand | New build (e.g. RAS) |
|---|---|---|---|
| Capital cost | Usually the smallest scope | Scales with added units plus shared services | Full design, civil works and equipment |
| Operating cost | Can lower cost per kg if the bottleneck is removed | Adds labour, energy and feed with output | Energy and skilled labour weigh more in RAS |
| Downtime | Possible partial shutdown during works | Existing units can keep producing | No production until commissioning and biofilter start-up |
| Added capacity | Recovered output from better survival/growth | Proportional to added units | Set by design |
| Quality and control | Improves where the constraint is removed | Same as existing units | Highest control of water and temperature, with technical risk |
| Main risk | Fixing a symptom instead of the cause | Copying existing problems | Start-up biology, schedule, financing |
| Sensitivity to check | Survival and FCR after retrofit | Market price and demand | Energy price, FCR, ramp-up time |
Takeaway: compare all three options on the same lines, including downtime.
Option payback check — try it here
For each option enter the capital cost, the extra tonnes per year it adds (or recovers), and months of lost production during works. Values shown are hypothetical placeholders.
Formulas: added profit/yr = added t × 1,000 × margin per kg. Lost margin = months ÷ 12 × current t × 1,000 × margin. Simple payback = (capital cost + lost margin) ÷ added profit/yr. Undiscounted; ignores ramp-up and financing cost.
Send this comparison with a project RFQWorked example (hypothetical values, not benchmarks)
A farm produces 400 t/yr at a hypothetical margin of USD 1.00/kg. Retrofit: USD 300,000 recovers 60 t/yr with 2 months of lost production. Added profit = 60 × 1,000 × 1.00 = USD 60,000/yr; lost margin = 2 ÷ 12 × 400 × 1,000 × 1.00 ≈ USD 66,667; payback ≈ (300,000 + 66,667) ÷ 60,000 ≈ 6.1 years. Expansion: USD 700,000 adds 200 t/yr with no downtime → payback = 700,000 ÷ 200,000 = 3.5 years. New build: USD 1,000,000 adds 250 t/yr → 4.0 years before ramp-up. In this hypothetical case expansion looks best — but only if current units already reach target, which the diagnosis must confirm. Change the margin by ±USD 0.30/kg and every payback moves substantially; test that before deciding.
Takeaway: payback is only as good as the added output and margin you can prove.
Related calculators and tools
Use these to build the baseline and the design basis. Results are planning estimates for discussion with a qualified engineer or biologist.
- Oxygen requirement calculator
Check oxygen demand against installed aeration.
- Biofilter sizing calculator
Feed load vs biofilter capacity in RAS.
- FCR calculator
Baseline FCR per cycle.
- Biomass calculator
Standing biomass from count and weight.
- Stocking density calculator
Compare density with published limits.
- RAS sizing calculator
Size a new-build RAS from target output.
- Harvest profit calculator
Margin per kg for the payback check.
- RAS cost calculator (CAPEX/OPEX ranges)
- 10-year TCO comparison (RAS, pond, cage, biofloc)
- Loan plan (payment, DSCR, grace period)
- Outcome-first guide: define results before buying
- Buy the right outcome at the right total cost
Limitations
- The example and calculator are hypothetical arithmetic. They do not predict survival, growth, FCR, prices or savings for any farm.
- Simple payback ignores discounting, ramp-up time, financing cost, tax and price changes. Use it to rank options, not to make the final investment decision.
- Diagnosis and final design need a qualified aquaculture engineer or biologist on site. Only written supplier proposals can commit to performance or price, and only a lender can approve financing.
Payment milestones and financing readiness
- Retrofit: tie payments to installation and a measured performance test (for example DO or water quality under load).
- Expansion: stage payments per phase so each block is accepted before the next.
- New build: link milestones to design approval, delivery, commissioning, biofilter start-up and first-stocking sign-off.
- Lenders look for cash flow and debt-service coverage; the decision file above is most of what they will ask for.
- FishMatch can introduce eligible projects to external financing partners. We do not lend, and approval is never guaranteed.
Takeaway: stage payments to the option's real risks.
Decision-file checklist
- Baseline KPIs from recent cycles, with how each was measured
- Binding constraint identified, with supporting data
- Gap to target in t/yr and USD/kg
- Three options compared on capital cost, operating cost, downtime, capacity, quality and risk
- Sensitivity of payback to margin, FCR and survival
- Site capacity: water, power, land and permits
- Chosen option's design basis and scope (project ≥ USD 250,000 for an RFQ)
- Financing plan, including working capital through ramp-up
Takeaway: a decision file that answers these points is also most of a loan file.
Frequently asked questions
Turn your chosen option into one RFQ
Send your baseline, the binding constraint and the option you want quoted. Our team reviews it (first human reply within 1 business day), contacts suitable suppliers for you and presents organised responses. Free for buyers; projects from USD 250,000.
Sources and method
- Timmons & Ebeling — Recirculating Aquaculture (oxygen and biofilter design rules used by our calculators)
- FAO — The State of World Fisheries and Aquaculture 2024
Last reviewed: October 2026. All example values are hypothetical; no market benchmarks, prices or savings are claimed.