How we evaluate and rank aquaculture AI tools
A transparent look at how tools enter the database, how profiles are written, and how the "Get My AI Tool Recommendation" flow scores and orders tools for your farm.
Core principles
The non-negotiables that shape every profile and every ranking.
Editorial independence
Every tool profile is written and maintained by our editorial team, not by the supplier. Suppliers can submit commercial content, but it is clearly labelled and never changes the profile, ranking, or advisory report.
No pay-to-rank
Rankings are generated algorithmically from farm inputs (species, production system, goals, budget, current tooling) against the tool's capabilities and evidence level. Advertising budget, referral fees, or commercial relationships do not move a tool up the list — ever.
Evidence-weighted
We distinguish between marketing-only claims, pilot deployments, and commercial deployments. Tools with public customer references and operational deployments are weighted higher than tools with only marketing pages.
Reality checks, not brochures
Each profile states where a tool is genuinely strong, where it may be limited, and what it does not directly address. We would rather lose a supplier than mislead a farm manager.
How a tool gets into the database
Four editorial steps, in order, before a tool reaches a farm manager.
1. Curation
We maintain a short, intentional database — currently 42 AI tools relevant to aquaculture. Tools are only added when we can verify the supplier, product scope, and at least one real deployment or detailed product specification.
2. Profiling
Every tool gets a standardized 5-section operator profile: what it is, what it's best at, where it fits, what it needs to work well, and where it may be limited — plus a typical reason to consider it and what it does not directly address.
3. Matching
When a farm profile is submitted, each tool is scored against the inputs. The top five are returned with explicit match reasons, and the advisory report highlights best fit, alternatives, and tools to avoid for that context.
4. Review & refresh
Profiles are reviewed when suppliers release new versions, when customer evidence changes, or at least during our scheduled editorial passes. Every profile shows its last-reviewed date.
How the recommendation engine scores tools
The scoring signals applied when you submit a farm profile through "Get My AI Tool Recommendation". Top 5 are returned with explicit match reasons.
| Signal | Weight |
|---|---|
| Species supported | +30 |
| Production system | +25 |
| Goal match | +15 each |
| Commercial evidence | +10 |
Scores are deterministic and reproducible. We do not re-rank manually after generation, and we do not add handcrafted boosts for individual tools (including our own).
Conflicts of interest — full disclosure
The Aquaculture AI Database is operated by Total Production Solutions (TPS), which also develops its own tool, Precision Fish Detection (PFD). PFD is included in the database on the same terms as every other tool.
To keep the database trustworthy:
- PFD is not pinned, boosted, or promoted above other tools in any list or recommendation.
- PFD is scored by the same algorithm, with the same weights, as every other tool.
- The editorial profile for PFD follows the same 5-section structure and reality-check rules as all other profiles.
- Commercial / supplier-provided content (for any tool, including PFD) is clearly labelled on tool pages and never changes the profile or ranking.
If you believe a profile or ranking looks biased, please tell us via the contact form. Editorial corrections are public.
Ready to see it in action?
Submit a short farm profile and get a decision-grade recommendation with explicit match reasons and an advisory report.