There is a server somewhere in a lot of maritime offices. Under a desk, in a cupboard behind the coats, in a small room that also holds the spare printer.
It runs the system the business depends on. Nobody is quite sure which version. The upgrade has been discussed twice and postponed twice, because upgrading means a weekend, and a weekend means somebody has to be there, and the person who understands it has one holiday a year.
None of that appears on an invoice. It is a cost all the same.
What the server actually costs
Not the hardware. The hardware is cheap and it mostly works.
The upgrade you keep not doing. Every version you fall behind makes the next jump bigger and the next weekend longer. Eventually the gap is large enough that the upgrade becomes a project, with a quote attached.
The one person who knows. Backups, certificates, the disk that filled up in March, the reason a particular service has to be restarted in a particular order. That knowledge is real and it is unwritten, and it walks out of the building with them.
The features you do not have yet. This is the expensive one, and the least visible. When new functionality requires a migration, it arrives late or it does not arrive at all. You are not paying for what you have; you are paying by not getting what has been built since.
The security you are quietly responsible for. Patches, access, the laptop of the colleague who left. In a port agency of nine people, nobody's job title says any of this.
What changes when it is not yours to run
SHIPM8 runs as a service. There is no server in your office, no upgrade weekend, and no version number for you to know. New functionality appears because it has been released, not because somebody scheduled a migration.
That has a consequence worth being explicit about, because it is the part customers tell us they notice: improvements arrive in the subscription rather than as a quote. We build what fits the platform, so what one customer asks for and gets becomes part of what everyone has. Nobody is billed for a change that was worth making.
It also means the numbers behave. A monthly subscription per user is a figure you know before the month starts. A busy month does not cost more, a quiet one does not cost less, and there is no project invoice arriving in the third quarter for work that turned out to be necessary.
The honest trade
You give up control of the timing. If we release on a Tuesday, it is released on a Tuesday. If you have a very particular reason to stay on a version for another six months, that is not something a shared platform offers.
In practice that trade has gone one way for the customers who have made it, but it is a real trade and it deserves stating rather than glossing over. The question is not whether you like the idea of running your own system. It is whether running it is work you want to keep owning, and whether the person doing it has anything better to do.
Two questions to ask yourself
Not of a vendor. Of your own operation:
- When did we last upgrade, and what did it cost in hours rather than in euros? Include the postponements. Twice-postponed is still twice-considered.
- If the person who looks after our system left next month, what would we find out? If the answer takes a while, you already know the size of the risk.
If you are weighing the two, ask for a demo and we will talk about what moving actually looks like from where you are, including the parts that are inconvenient.