The licence fee is the only line in a marine software budget you can price accurately, because it is the only one that does not depend on the condition of your existing data or the shape of your operation. Everything else is a function of variables the vendor cannot see at quotation stage and that most fleets have never measured. A Singapore ship manager replacing a crew system in 2026 spent three months auditing the legacy database before touching production, and found that roughly thirty percent of crew certificate records were incomplete or stored in inconsistent formats — expiry dates recorded in one format in some records and a different one in others. That figure did not appear in any proposal. It appeared during a discovery phase that existed only because somebody insisted on one. Published guidance for shipowners lists the same recurring omissions: data migration covering maintenance records, certificates and historical information out of spreadsheets and paper; training for onboard crews and shore personnel, particularly where several modules go live together; and satellite communication costs, because poorly optimised cloud synchronisation generates additional VSAT and Starlink traffic. That third item appears in no generic software implementation guide, because no other industry pays for its own bandwidth by the megabyte. Start a free trial of Marine Inspection and find out what your data actually looks like before you commit to a plan.
Priced accurately
The licence
A number on a proposal, fixed at signature, and the same for you as for any comparable fleet. It is the only line where the vendor knows as much as you do.
Priced by estimate
Data migration, scaled by how good your records currently are
Training, scaled by your crew count and your rotation frequency
Satellite bandwidth, scaled by how the platform synchronises
Parallel running, scaled by how long the rollout takes
Configuration and integration, scaled by what you already run
Internal time, scaled by who owns it and what they stop doing
Six lines, none of which the vendor can size without seeing inside your operation, and several of which you cannot size either until somebody looks. That asymmetry is the entire reason implementation budgets break — not because anyone concealed anything, but because the estimate was made before the facts existed.
Thirty Percent
The most useful number in maritime implementation is not a cost. It is the proportion of records that turn out not to be migration-ready. Book a Marine Inspection demo and ask what happens if your data is worse than you told them it was.
~30%
of crew certificate records found incomplete or in inconsistent formats during a 2026 legacy system audit
How it was found, and why that matters more than the number
Before touching the production system, the project team spent three months understanding what was actually in the legacy database. The audit surfaced the problem: incomplete records, and certificate expiry dates recorded in multiple formats — one field written as day, month, year in some records and as year, month, day in others.
Neither of those is exotic. Both are what happens when a system is maintained by many people across many years without validation on entry. And both are invisible until somebody queries the data rather than reading it, which is why they are absent from proposals and present in overruns.
The lesson is not that migration is expensive. It is that migration cost is unknowable at quotation stage by either party, which means a fixed migration price quoted before a data audit is either padded or optimistic, and you cannot tell which from the number.
Ask this before signing
What happens if our data turns out to be materially worse than expected — who absorbs that, and is the position written down? A vendor who has done this before will have a defined answer, because they have met the thirty percent. One who has not will say it will be fine.
Three Costs That Exist Only in Maritime
Generic implementation guidance covers migration, training and integration. These three do not appear in it, and each is a recurring cost rather than a one-time one. Sign up for Marine Inspection and check each against any platform you evaluate.
Cost one
Satellite bandwidth
Published shipowner guidance lists satellite communication costs among typical hidden expenses, noting that poorly optimised cloud servers may generate additional VSAT and Starlink traffic during synchronisation. This is a genuine cost created by a software architecture decision, billed by a different supplier, arriving on a different budget line, and recurring on every vessel every day for as long as the platform is in use.
Ask: how much data does a vessel send in a normal day, and what does sync do when a ship has been offline for a week?
Cost two
Training against a rotating workforce
Shore staff are trained once. A crew is trained, then rotates, and the replacements arrive untrained. Published guidance flags training costs for onboard crews and shore personnel as a hidden expense, particularly where multiple modules are deployed simultaneously — and on a workforce that turns over several times a year, that is an operating cost misfiled as an implementation one.
Ask: what does training a newly joined crew member cost after go-live, and is it included or per head?
Cost three
The operational calendar
A successful 2026 fleet migration was structured around the operational calendar rather than an IT release schedule, phasing the rollout to drydocking windows and planned crew rotations — the natural points where vessel operations pause or slow. That is the right approach and it has a cost: your rollout timeline is set by the fleet's movements rather than by the project plan, and the timeline is itself a cost line.
Ask: how does the rollout phase across our actual docking and rotation schedule, and what does that do to the duration?
Six to twelve weeks
The deployment window is a cost line, not a schedule line
A platform taking six to fifteen months to go live delays every benefit and carries months of parallel-running cost. One live in six to twelve weeks starts returning value almost immediately. On a phased maritime rollout that means registry setup and template loading in the first weeks, data migration in the middle, training and integration next, and live operations inside a fiscal quarter rather than a fiscal year.
What Parallel Running Actually Costs
The line nobody budgets, because it is not paid to anybody. It is absorbed, which makes it invisible on an invoice and entirely real in the operation. Schedule a walkthrough and ask for the go-live date rather than the contract date.
Two systems, one team
Until every vessel has transitioned, the old system and the new one both hold live data, and somebody is responsible for keeping them consistent. On a fleet phased to docking windows, that period is measured in months rather than weeks.
Two sets of licence fees
Legacy contracts rarely terminate cleanly mid-term, so the old subscription frequently runs alongside the new one for a period. This is a real cash cost and it is routinely omitted from the comparison because it belongs to neither budget cleanly.
Divided crew attention
Crews on transitioned vessels are learning a new system while crews elsewhere continue on the old, and the shore team supports both. That split is where most of the productivity cost of a slow rollout actually sits.
And the compounding effect
A longer rollout crosses more crew rotations. Cross enough of them and you are training the same vessel's people twice on the same system before the fleet is even live — which converts an implementation cost into a recurring one purely as a function of duration.
The Full Line Item List
Ask for a written figure or a defined position on each of these before a proposal is finalised. Most are answerable; the ones that are not are the ones worth negotiating. Start a free trial and generate a data sample to inform the migration question specifically.
Table 1: Every Line That Is Not the Licence
| Line | What drives it | Timing | What to ask for |
|---|---|---|---|
| Data audit | The condition of records nobody has queried in years | Before migration, or you discover it during | Whether an audit is included, and what it covers |
| Data migration | Maintenance records, certificates and history out of spreadsheets and paper; described in maritime guidance as frequently the single biggest one-time cost | One-time, scaled by data condition | A ceiling, plus a defined position if the data is worse than expected |
| Implementation services | Fleet size, module count and configuration depth; frequently a separate line item | One-time | Fixed fee rather than day rate, covering go-live across the fleet |
| Configuration | How far the platform bends to your processes | One-time, then on change | How many configuration decisions are required before go-live |
| Integration | Connections to crewing, procurement, finance, class portals | One-time, then on either system changing | Whether API access is included at your tier or priced separately |
| Initial training | Crew and shore headcount, and whether modules go live together | One-time per person | Delivery method, and cost per vessel rather than per session |
| Rotation retraining | Crew turnover frequency, which is several times a year | Recurring, permanently | Whether it is included or charged, and what the material is |
| Satellite bandwidth | Sync architecture and how efficiently it handles reconnection | Recurring, per vessel per day | Typical daily volume, and behaviour after a long offline period |
| Parallel running | Rollout duration, which is set by docking and rotation windows | For the duration of the rollout | A realistic go-live date per vessel, not a contract date |
| Legacy overlap | Whether the old contract can be exited cleanly | Until the old contract ends | Your own notice terms, checked before you commit to a timeline |
| Internal ownership | Who runs the project and what they stop doing | Throughout, and beyond | An honest day estimate costed at a real rate |
| Change requests | What was not specified because nobody knew to specify it | After go-live, forever | The change request process and its pricing, agreed in advance |
Twelve lines. One of them is the licence.
Which is why the cheapest quote and the cheapest project are frequently different proposals.
Where the Proportions Land
Published figures from a maritime modernisation business case give a sense of relative scale, and the ratio is more useful than the absolute numbers. Book a walkthrough and apply the ratio to whatever licence figure you have been quoted.
Licence and implementation
Quoted in one published Singapore modernisation case at eight hundred thousand to two million Singapore dollars, covering the software itself and the professional services to deploy it.
Data migration and testing
Two hundred to five hundred thousand in the same case — a substantial fraction of the licence line, and the one most sensitive to the thirty percent problem described above.
Training and change management
One hundred to two hundred and fifty thousand, grouped together because they are the same activity viewed from two angles: getting people to use the thing you bought.
The ratio worth carrying
Migration, testing, training and change management together run at roughly a quarter to two fifths of licence and implementation combined. If your quote covers only the first line, you are looking at something between sixty and eighty percent of the project.
Those figures come from a large enterprise modernisation in a specific market and should not be scaled directly to a smaller fleet or a modern subscription platform, where implementation is measured in weeks and priced in thousands rather than hundreds of thousands. What transfers is the proportion, not the magnitude — and the proportion says that the lines outside the licence are not a rounding error.
Why Vendors Cannot Simply Quote This
It is worth being fair about the mechanism, because the reflexive assumption is concealment and the reality is usually structural. Start a free trial and remove the largest unknown before anybody quotes.
They cannot see your data
Migration cost scales with record condition, and record condition is only knowable by querying the database. A vendor quoting before that has either padded the number or guessed low, and the buyer cannot distinguish between the two from the figure alone.
They cannot see your calendar
Rollout duration is set by docking windows and crew rotations, which the vendor does not hold. A generic twelve-week plan meets a fleet whose vessels dock across nine months and becomes a nine-month plan.
The market is not transparent
Industry commentary notes that the maritime software market is not transparent, and that the difference between a vendor who understands how data flows across vessels in a staggered rotation cycle and one who does not becomes apparent only after the contract is signed. That is a procurement problem as much as a pricing one.
What follows from all three
Do the discovery before the contract rather than after it. A data audit, a pilot on one vessel and a rollout plan mapped to your own docking schedule convert three unknowns into three numbers — and they are cheap to produce compared with the cost of finding out during implementation.
Questions That Change an Implementation Quote
Each of these has moved a number in real negotiations, and a vendor who has implemented in maritime before will have answers ready. Schedule a demo and put them in writing before the proposal is drafted.
Table 2: Questions to Ask Before the Proposal Is Written
| Area | The Question | A Real Answer | What Should Worry You |
|---|---|---|---|
| Data condition | What happens if thirty percent of our records are incomplete or inconsistent? | A defined position, because they have met that before | Reassurance that it will be fine, which means they have not |
| Migration ceiling | Is the migration price a fixed fee or an estimate? | A ceiling in the contract with a defined scope | A day rate, which transfers all data risk to you |
| Go-live date | When is the last vessel live, not when does the contract start? | A per-vessel schedule mapped to your docking windows | A single go-live date for a fleet that does not move together |
| Bandwidth | What does a vessel transmit in a normal day, and after a week offline? | A figure, because they have measured it | Never having been asked, which tells you about their other customers |
| Retraining | What does training a newly joined crew member cost after go-live? | Included, or a per-head figure you can budget by rotation | Training treated as a one-time implementation activity |
| Parallel period | How long do we run both systems, and who keeps them consistent? | An explicit period with a named owner on each side | The question not having occurred to them |
| Configuration count | How many decisions do we have to make before go-live? | A number, and a template to start from | Unlimited configurability offered as a benefit |
| Change requests | What does a change cost after go-live, and how is it priced? | A stated process and rate agreed in advance | Deferred to the relationship, which is where budgets go |
| Reference | Can I speak to a customer about what their implementation actually cost? | A contactable reference of comparable fleet size | A case study PDF with no reachable person behind it |
2026 IMPLEMENTATION REALITY
The figures here are drawn from published maritime case studies, consultancy analysis and vendor guidance rather than from an audited dataset. The thirty percent record-condition finding comes from a single documented fleet migration and describes that operation rather than an industry average; your own data may be considerably better or worse, which is precisely the argument for auditing it. The modernisation cost figures come from one large enterprise project in one market and should not be scaled to a smaller fleet or a modern subscription deployment — what transfers is the proportion between lines, not the magnitude. This page is published by a software vendor. Several of the questions above are ones we would have to answer too, and a fair evaluation asks them of every candidate including us. Satellite bandwidth is a real cost created by architecture. Published shipowner guidance identifies poorly optimised cloud synchronisation as generating additional VSAT and Starlink traffic; the magnitude depends entirely on how a specific platform is built, so ask for measured figures rather than assurances. Nothing here removes the statutory obligations that apply during a transition — records must remain available throughout, which is itself an argument for a phased rather than an abrupt cutover.
Frequently Asked Questions
What is usually the biggest hidden cost?
Data migration, described in maritime software guidance as frequently the single biggest one-time cost — moving maintenance records, certificates, historical information and contracts out of spreadsheets, paper and legacy systems. Its size scales with the condition of your existing records, which neither party knows at quotation stage. A documented 2026 fleet migration found roughly thirty percent of crew certificate records incomplete or stored in inconsistent formats, including expiry dates written in different date formats across the same field, and that only surfaced because the team spent three months auditing the legacy database before touching production.
Why does satellite bandwidth appear on this list?
Because it is a genuine recurring cost created by a software architecture decision, and it exists in almost no other industry. Published shipowner guidance lists satellite communication costs among typical hidden expenses, noting that poorly optimised cloud servers may generate additional VSAT and Starlink traffic during synchronisation. The cost arrives on a communications budget rather than an IT one, recurs on every vessel every day, and is entirely dependent on how the platform handles synchronisation — particularly after a vessel has been offline for an extended period, when a poorly designed sync will attempt to move everything at once.
Why is training not a one-time cost?
Because the workforce rotates. Shore personnel are trained once and stay; crews are trained, rotate off, and are replaced by people who have never seen the system. Published guidance flags training for onboard crews and shore personnel as a hidden expense, particularly where multiple modules are deployed simultaneously — but the maritime specific is that crew training recurs several times a year, permanently. Budgeting it as an implementation line means it disappears from the model after year one and continues appearing in the operation. Ask whether retraining is included or charged per head, and what self-service material exists for a newly joined engineer at two in the morning.
What does a long deployment actually cost?
More than the delay in benefits, though that alone is significant — a platform taking six to fifteen months to go live delays every benefit and carries months of parallel-running cost, whereas one live in six to twelve weeks starts returning value almost immediately. Parallel running means two systems holding live data with somebody responsible for consistency, frequently two sets of licence fees because legacy contracts rarely terminate cleanly mid-term, and divided crew and shore attention. And a longer rollout crosses more crew rotations, which can mean training the same vessel's people twice on the same system before the fleet is even fully live.
How should a fleet rollout be phased?
Around the operational calendar rather than an IT release schedule. A documented 2026 fleet migration structured its rollout around drydocking windows and planned crew rotations — the natural points where vessel operations pause or slow — rather than around a project plan. That is the right approach and it has a consequence worth budgeting: your timeline is set by fleet movements you do not control. A vendor quoting a generic twelve-week plan without asking about your docking schedule has not accounted for the constraint that will actually determine the duration.
Are vendors hiding these costs deliberately?
Usually not, and the structural explanation is more useful than the cynical one. Migration cost depends on record condition the vendor cannot see; rollout duration depends on a docking and rotation calendar they do not hold; and industry commentary notes the maritime software market is not transparent, with the difference between a vendor who understands how data flows across staggered rotation cycles and one who does not becoming apparent only after the contract is signed. The practical response is to do the discovery before the contract rather than after — a data audit, a pilot on one vessel, and a rollout plan mapped to your own schedule turn three unknowns into three numbers.
Find Out What Your Data Looks Like Before You Buy the Plan
A free trial on one vessel produces three things a proposal cannot: a genuine view of what your existing records look like when they meet a structured system, a measured capture time from the people who will actually use it, and a rollout estimate based on your own docking and rotation calendar rather than a generic twelve-week template. Those three convert the largest unknowns in an implementation budget into numbers, at no cost, before anybody has to sign anything.
One vessel. No commitment. Then quote it properly.
