2026-03-18 · 6 min read
What should a maritime connectivity SLA actually contain?
Most vessel connectivity contracts contain an SLA. Most of those SLAs are not worth the page they occupy.
The standard structure is familiar: a headline uptime percentage, a credit mechanism triggered by sustained outages, and a support ticket process. What is absent is everything that would make the SLA operationally meaningful — zone-specific measurement, bearer-level accountability, and remedies calibrated to actual disruption rather than cosmetic credits.
This article sets out what a maritime connectivity SLA should contain if it is going to be useful to the fleet manager accountable for performance, not the vendor who drafted it.
Why existing SLAs fail in practice
Shore-based SLA frameworks do not transfer to vessels. A terrestrial SLA can assume a single infrastructure provider, fixed geography, and a stable physical layer. Maritime connectivity involves multiple bearers — VSAT, Starlink, LTE, L-band — operating across changing maritime zones, each with different performance characteristics and different responsible parties.
When a vessel goes dark for four hours in a high-traffic shipping lane, the vendor's uptime report may record nothing at all. Beam contention does not trigger outage logs. Weather-related degradation is excluded. The SLA credit mechanism requires a single sustained outage of a defined duration — not the rolling degradation that makes a vessel functionally unreachable for half a working day.
The fleet manager knows something went wrong. The report says nothing did.
Zone-based uptime: the minimum starting point
Uptime figures mean different things in different operating environments. A vessel running port rotations has continuous LTE coverage in harbour, VSAT or LEO coverage in coastal waters, and open-ocean conditions with no terrestrial fallback. Aggregating these into a single monthly uptime number conceals the performance profile that actually matters.
A credible SLA should specify uptime commitments by operating zone: port/anchorage, coastal (within 200nm), and open ocean. Each zone carries different bearer availability and different operational criticality. Holding the provider to a single aggregate figure allows every difficult kilometre to be averaged out by easy ones.
Bearer-level responsibility splits
Multi-bearer environments create a diffused accountability problem. When VSAT performance degrades and the failover to LTE also fails, the VSAT provider blames local cell congestion and the LTE provider blames VSAT handoff logic. The fleet manager owns the problem and neither vendor owns the cause.
The SLA needs to specify which party is responsible for each bearer, what the handoff criteria are, and who is accountable for failover failure. Where a managed service provider or integrator sits above both bearers, the contract with that provider must consolidate responsibility rather than pass it down to individual bearer suppliers.
The five KPIs that matter
A monthly connectivity report should contain, per vessel, at minimum:
Uptime percentage — by bearer and by zone. Not aggregated. Not estimated from carrier-side telemetry.
Median round-trip time and P95 latency — The median tells you typical conditions; the P95 tells you whether high-latency events are frequent enough to affect operations.
Peak throughput and average throughput — by bearer. Peak capacity that is never available under load is not a useful metric. Providers advertise peak. Operators need average.
Data volume by category — operational traffic, crew welfare, and management traffic separately. This matters for cost allocation and for diagnosing whether a vessel is consistently over or under the agreed data package.
Ticket volume and resolution time — segmented by whether the fault was carrier-side, hardware-side, or configuration-side. This is the only way to track whether integration problems are improving or recurring.
Remedies that mean something
SLA credits are typically calculated as a fraction of the monthly service fee, triggered only by outages exceeding a defined duration threshold. For a vessel operating under time-sensitive cargo commitments, a credit equivalent to half a day's service fee does not compensate for the operational disruption of a four-hour blackout.
Meaningful remedies should be proportionate to operational impact, not to service cost. Where connectivity failure has a measurable downstream consequence — missed port reporting window, delayed arrival notification, crew welfare failure — the credit structure should reflect that, not the vendor's cost of service delivery.
The monthly reporting pack
The SLA should specify the format and delivery schedule of the monthly report, not just the performance commitments. A fleet manager should receive, without having to request it: per-vessel performance against each KPI, a summary of incidents with root cause and time-to-restore, and a trend line showing whether performance is improving or degrading.
If the provider cannot produce this, the SLA is unenforceable in practice regardless of what it says on paper.
The underlying problem
Connectivity vendors are not incentivised to surface the metrics that reveal their own underperformance. The current state of maritime SLA reporting reflects that incentive structure. Fleet managers who accept vendor-defined reporting frameworks will continue to receive vendor-flattering results.
The starting point is specifying what you need in the contract before it is signed — zone-level uptime, bearer-level accountability, five named KPIs, and a monthly report that you do not have to chase.
That is not an unreasonable ask. It is the minimum standard for infrastructure that vessels depend on.
Orbit measures what your SLA should.
Multi-bearer visibility, incident governance, and monthly SLA packs — independent of your bearer providers.
Book a call