.png)
Frameworks, core principles and top case studies for SaaS pricing, learnt and refined over 28+ years of SaaS-monetization experience.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Why Is Maritime AI Agent Pricing Tied to Vessel Tonnage
Gross tonnage can look like an odd meter for AI. An agent does not necessarily consume twice as many model tokens because it supports a 3,000 GT vessel instead of a 500 GT vessel. Nor does a larger vessel always generate twice as many crew questions, certificates, or maintenance events.
Yet vessel tonnage remains one of the clearest ways to price maritime operations software with AI built in. It is already understood by shipowners, managers, class societies, ports, flag administrations, and insurers. More important, it gives a supplier a stable way to reflect the operating burden attached to the vessel the agent supports.
Monetizely’s position is clear: maritime AI agents that manage vessel operations, compliance, maintenance, or safety records should be priced primarily per vessel in gross-tonnage bands under an annual fleet subscription. Tonnage is not a proxy for tokens. It is a practical proxy for the vessel-specific regulatory exposure, operating complexity, and account value that the agent is being asked to manage.
Tonnage converts a vessel’s operating burden into a billable unit
Gross tonnage is not cargo weight. It measures the enclosed volume of a ship, while deadweight tonnage measures how much weight a vessel can carry. The distinction matters because GT is already embedded in the commercial and regulatory life of a vessel: the UK government notes that GT informs manning rules, safety requirements, registration fees, port dues, and canal charges.
Maritime buyers therefore do not need to be taught what a tonnage band means. A fleet manager already knows that a 250 GT workboat, a 2,500 GT yacht, and a 50,000 GT tanker sit in different operating environments. The difference is not simply size. It includes crew structure, equipment, documentation, reporting, inspection exposure, voyage profile, and the cost of getting a compliance decision wrong.
Several current regulatory thresholds show why GT works as a useful commercial anchor.
| Existing maritime signal | Current rule or use | What it tells an AI pricing team | Source date and source |
|---|---|---|---|
| Gross tonnage | Used as a basis for manning rules, safety rules, registration fees, port dues, and canal fees | GT is already an accepted shorthand for the scale of a vessel’s operating obligations | UK HMRC manual updated March 24, 2026 |
| AIS carriage | AIS is required for ships of 300 GT and above on international voyages, cargo ships of 500 GT and above outside international voyages, and all passenger ships | Larger vessels often face more continuous reporting and data obligations | IMO guidance accessed September 3, 2026 |
| Safety management | The ISM Code applies to many cargo ships of 500 GT and above on international voyages | At 500 GT, a vessel commonly enters a more formal safety-management environment | IMO resolution adopted December 6, 2017 |
| Electronic charts | New cargo ships of 3,000 GT and above on international voyages must carry ECDIS | A 3,000 GT band can reflect a more demanding navigational and software environment | IMO guidance accessed September 3, 2026 |
The point is not that every AI task rises in a straight line with GT. The point is that tonnage already captures meaningful changes in the vessel environment that the agent must understand, monitor, and support.
A maritime agent that tracks certificates, prepares audit evidence, flags rest-hour risks, drafts maintenance work orders, or summarizes incident reports is not serving an abstract software account. It is serving a vessel with a known operating profile. GT makes that profile visible in the contract.
Tonnage is not enough on its own. A small gas carrier can face a heavier compliance burden than a much larger general cargo vessel, because the International Gas Carrier Code applies regardless of size. The answer, however, is not to abandon GT for tokens or seats. The answer is to use GT as the core meter and package high-risk cargo, route, and workflow needs separately.
The five pricing decisions point to the vessel, not the individual user
Monetizely’s 5-Step Pricing Framework starts with five linked decisions: business goals and customer segments; packages that fit those segments; the pricing metric; price points; and the operational systems needed to meter, invoice, renew, and govern the offer. The order matters. A company that begins with a meter because engineering can count it often ends with a price that customers cannot explain internally. The sequence described in Monetizing Agentic AI forces the supplier to decide first what it is selling and to whom.
For maritime AI, those five decisions lead in one direction:
A seat meter fails at the third step. The person using the dashboard may be a superintendent, captain, compliance officer, or shore-based manager, but the work belongs to the vessel. One compliance officer may oversee five vessels; another may oversee 50. Charging per named user would make the bill depend on the customer’s staffing design rather than on the assets receiving the service.
A token meter fails for the opposite reason. It follows the supplier’s model cost, not the buyer’s operating burden. A customer does not budget vessel compliance by asking how many tokens were used to summarize a safety report.
AMS shows that maritime agents sit beyond the seat-pricing zone
The Agentic Monetization Spectrum, or AMS, makes the case more precise. It rates an agent on three dimensions: how much work it completes without human involvement, how broad its operating domain is, and how sharply the value of its output exceeds the cost of producing it. A small score on zero-human ability favors user-based pricing because the human remains the center of the work. A large score shifts the anchor toward the agent’s output. A narrow agent looks like a tool; an agent covering a complete business workflow starts to look like a job function.
The three AMS dimensions are:
The scores below use 1 for small, 2 for medium, and 3 for large or inflecting. They are commercial judgments about the product role, not claims about any vendor’s financial results.
The important row is the first one. A maritime compliance agent is more autonomous than a traditional dashboard, but it is not a fully independent digital worker. Captains, designated persons ashore, technical managers, and masters still review information and carry legal responsibility. The IMO’s current guidance on maritime security is explicit that the master retains ultimate responsibility for ship safety and security.
That medium autonomy might appear to favor seats. AMS points elsewhere because the operational domain is also medium: the agent supports a continuous vessel workflow, not a single employee task. It may monitor documents every day, prepare evidence ahead of audits, surface missed maintenance steps, and keep a vessel-ready record across crew rotations. The right commercial anchor is the vessel’s standing operating burden.
The output/cost ratio also matters. A single model call that flags an expired certificate may cost pennies, while the avoided disruption, audit delay, or manual follow-up can be much larger. That is an inflecting relationship, not a linear one. A tonnage-banded vessel price captures part of that value without forcing the buyer to debate whether each alert “resolved” a problem.
Public agent pricing validates the meter only when it matches the job
The market already offers a useful reference set. Public pricing shows four distinct agent models: per resolution, platform-plus-consumption, per seat, and flat subscription. Each works when it follows the buyer’s unit of value. None should be copied into maritime software without asking whether the underlying job is similar.
The pattern is consistent: Intercom can charge per resolution because a customer-support resolution is discrete, observable, and close to the buyer’s value. GitHub can charge per seat because a developer still directs and reviews the coding work. Salesforce needs more than one option because it sells a broad platform across many enterprise workflows. Magnor’s GT-banded price is the closest fit for maritime agents because the vessel, rather than the conversation or employee, remains the enduring unit of responsibility.
Maritime suppliers should resist a superficial reading of Intercom’s success. A support agent can define a resolution as a customer leaving without further help. A vessel-compliance agent cannot define success so cleanly. Was a certificate “resolved” when the AI flagged it, when a superintendent reviewed it, when the class society accepted it, or when the vessel sailed? A meter that creates argument at every invoice is not value-based pricing. It is a collections problem.
The right test is not whether a meter is technically exact. It is whether a fleet buyer can forecast the bill, connect it to the asset receiving value, and defend it in a budget meeting. GT bands outperform the alternatives for core vessel operations agents.
The table does not argue that vessel count is irrelevant. Every vessel should be counted. It argues that an undifferentiated per-vessel fee leaves money on the table at the high end and overcharges smaller vessels at the low end. GT bands give the supplier a simple way to recognize real differences while keeping the invoice stable.
A sound offer has two layers:
- A fleet platform fee can cover common onboarding, single sign-on, core integrations, and fleet-level analytics.
- A per-vessel GT subscription should remain the primary commercial meter for the agent’s ongoing work.
That is not an attempt to avoid choosing a meter. The GT-banded vessel subscription is the central meter. The fleet fee simply recovers shared account work that does not belong to any single ship.
Public list pricing from Magnor offers a useful illustration of how GT bands change what a customer pays. Consider a 10-vessel fleet with eight vessels below 500 GT and two vessels above 3,000 GT. The calculation uses Magnor’s current monthly list prices and its stated annual billing method of 10 times the monthly rate.
The example shows why a flat per-vessel rate creates avoidable distortion: it would cost this mixed fleet 28% more than the GT-banded structure over the same three-year period.
The supplier’s benefit is equally important. A fleet made up of larger ships should not receive enterprise-grade support, complex integrations, deeper compliance workflows, and more demanding reporting for the same price as a fleet of smaller vessels. GT bands create a defensible path to higher contract value without inventing opaque activity charges.
A token meter may look prudent because inference costs can be volatile. It offers a direct line from supplier cost to customer price. That apparent discipline is also its weakness.
When underlying inference gets cheaper, a supplier that sells tokens has made its own cost curve the customer’s reference price. OpenAI reported on April 14, 2025 that GPT-4.1 mini matched or exceeded GPT-4o on its cited intelligence evaluations while reducing cost by 83%, and that GPT-4.1 was 26% less expensive than GPT-4o for median queries.
Those changes do not mean maritime software prices should never fall. A supplier should revisit price points as competition, willingness to pay, and margins change. They do mean that the meter should not be tied to a cost input that can decline sharply while the buyer’s operational exposure remains intact.
A vessel still needs audit-ready evidence, crew records, maintenance planning, and compliance oversight when model prices fall. The supplier should capture lower inference costs through better margin, a broader product, lower entry prices, or more competitive GT bands. Passing every model-cost change through a token rate teaches customers that the agent is infrastructure rather than operational software.
Internal controls should still protect margins:
- Route simple classifications and document extraction to lower-cost models.
- Set reasonable use policies for bulk historical document processing.
- Price unusually large data migrations, custom integrations, and external data feeds as fixed-scope work.
None of those measures requires the customer’s monthly fleet subscription to fluctuate with AI usage.
GT is the primary meter, not a claim that all vessels of the same size create the same work. A 2,000 GT offshore support vessel, cruise vessel, and product tanker can have different data flows, risk profiles, and regulatory needs. Those differences belong in packages.
A supplier should separate the offer into a core vessel package and a small number of fixed add-ons:
This structure preserves a coherent commercial story. The buyer pays for the operating environment of each vessel, then adds capabilities that reflect a specific business need. The supplier does not need a new meter every time an edge case appears.
Maritime AI vendors have an opportunity to avoid the pricing confusion now appearing in many horizontal agent markets. The category does not need to choose between a legacy maritime metric and modern AI economics. It needs to use the existing maritime metric for the commercial relationship and modern AI economics for internal cost control.
The strongest offers will make three promises at once: the fleet can forecast what it will pay, the supplier can earn more as it supports more demanding vessels, and the invoice reflects the vessel that carries the operational and regulatory burden. GT bands can meet all three conditions.
Build the commercial data model around the fleet register. Treat vessel ID, GT, vessel type, flag, and activation date as pricing data from the first implementation, not as fields added later by finance.
Make the 500 GT and 3,000 GT thresholds starting hypotheses, not permanent doctrine. Test whether those bands predict implementation effort, support load, renewal value, and willingness to pay across the fleet segments you serve.
Sell a vessel operating system, not access to an AI feature. Sales teams should lead with audit readiness, faster superintendent work, cleaner handovers, and reduced manual follow-up rather than token counts or model names.
Create a separate commercial path for hazardous cargo, offshore operations, and unusual trade patterns. Those customers should receive a more valuable package, not a confusing exception to the core GT logic.
Review the price point every two quarters while keeping the GT meter stable. Model costs, competitive offers, and adoption data may justify rate changes. They do not justify changing the unit customers use to understand the product.
Assumptions
The AMS scores are analytical judgments based on each product’s public description and pricing model, not audited performance measures. The three-year example assumes eight vessels below 500 GT, two vessels above 3,000 GT, unchanged list prices, annual billing at 10 times the monthly rate, and no implementation, integration, discount, tax, or currency effects.
Footnotes
- Monetizing Agentic AI, Amazon listing: https://www.amazon.com/Monetizing-Agentic-AI-Handbook-Transformation/dp/B0H7Z13VKJ/
- Monetizely, “Step 1: Goals and Segmentation,” accessed September 3, 2026: https://www.getmonetizely.com/monetizing-agentic-ai-book-saas/step-1-goals-and-segmentation
- Monetizely, “Step 2: Packaging - Designing Offers That Fit,” accessed September 3, 2026: https://www.getmonetizely.com/monetizing-agentic-ai-book-saas/step-2-packaging-designing-offers-that-fit
- Monetizely, “Step 3: Choosing the Right Pricing Metric,” accessed September 3, 2026: https://www.getmonetizely.com/monetizing-agentic-ai-book-saas/step-3-choosing-the-right-pricing-metric
- Monetizely, “The Agentic Monetization Spectrum,” accessed September 3, 2026: https://www.getmonetizely.com/monetizing-agentic-ai-book-saas/the-agentic-monetization-spectrum
- Monetizely, “Step 4: Finding the Right Price Points,” accessed September 3, 2026: https://www.getmonetizely.com/monetizing-agentic-ai-book-saas/step-4-finding-the-right-price-points
- Monetizely, “Step 5: Operationalizing Agentic AI Pricing,” accessed September 3, 2026: https://www.getmonetizely.com/monetizing-agentic-ai-book-saas/step-5-operationalizing-agentic-ai-pricing
- UK HM Revenue & Customs, “TTM15020 - Background Material: Ships,” updated March 24, 2026: https://www.gov.uk/hmrc-internal-manuals/tonnage-tax-manual/ttm15020
- International Maritime Organization, “AIS Transponders,” accessed September 3, 2026: https://www.imo.org/en/ourwork/safety/pages/ais.aspx
- International Maritime Organization, Resolution A.1119(30), adopted December 6, 2017: https://wwwcdn.imo.org/localresources/en/KnowledgeCentre/IndexofIMOResolutions/AssemblyDocuments/A.1119%2830%29.pdf
- International Maritime Organization, “Electronic Nautical Charts and ECDIS,” accessed September 3, 2026: https://www.imo.org/en/OurWork/Safety/Pages/ElectronicCharts.aspx
- International Maritime Organization, “IGC Code,” accessed September 3, 2026: https://www.imo.org/en/ourwork/safety/pages/igc-code.aspx
- International Maritime Organization, “Frequently Asked Questions on Maritime Security,” accessed September 3, 2026: https://www.imo.org/en/ourwork/security/pages/faq.aspx
- Magnor Vessels, “Yacht Management & Compliance Software,” pricing accessed September 3, 2026: https://magnorvessels.com/
- Intercom, “Fin AI Agent Outcomes,” July 30, 2026: https://www.intercom.com/help/en/articles/8205718-fin-ai-agent-outcomes
- Salesforce, “Agentforce Pricing,” accessed September 3, 2026: https://www.salesforce.com/agentforce/pricing/
- GitHub, “About Billing for GitHub Copilot in Organizations and Enterprises,” accessed September 3, 2026: https://docs.github.com/en/copilot/concepts/billing/organizations-and-enterprises
- Delyt, “AI Agents for Customer Support, Sales, and Social,” pricing accessed September 3, 2026: https://www.delyt.ai/
- OpenAI, “Introducing GPT-4.1 in the API,” April 14, 2025: https://openai.com/index/gpt-4-1/