Logistics systems, built by people who've done the job

You don't need an enterprise supply chain platform. You do need better control.

Most operations sit between a spreadsheet that's outgrown itself and enterprise software that costs more than the problem. We build the thing in between.

A logistics system shaped around how your operation actually runs, owned and run by us, at a price that used to only buy something off the shelf. Enter the data once and every document comes out of that one record.

Enter once  →  MEL  ·  M.R.  ·  load list  ·  crane list  ·  manifest  ·  DG declaration  ·  dispatch note  ·  inventory

Custom built used to be unaffordable. That changed.

For twenty years the choice was a platform that half-fitted or a six-figure development project. So everyone bent their operation to fit the platform, and the people doing the work built spreadsheets to cover the gaps.

Two things changed. AI has collapsed what it costs to build and maintain software. And integration platforms mean nobody writes plumbing any more — connecting a database, a document engine and a mail system is configuration now, not a project.

So the maths is different. A system built around your methods now costs less than bending your operation around someone else's, and it doesn't come with seats you don't use or modules you'll never turn on.

We build to your process, but not from a blank page. Crystalline V1 already does the hard part, and adapting it is a fraction of the work of starting over.

One well. The same field, re-entered in every document that needs it.

21data fields
23documents
0separate entries

Supplier, description, quantity and unit of measure are each typed out more than twenty times per well.

What that buys you

The time back, first of all

Count what happens before a vessel sails: the same item typed into a requisition, then a load list, then a manifest, then a crane operations list, then a dangerous goods declaration. Five times, by hand, under time pressure.

Enter it once and every document comes out of that one record. What used to be a morning of retyping is a few clicks, and the version that goes out is the version in the system.

Changes are recorded

A spreadsheet gives you a file called final_v3_REVISED and somebody's memory. Here every edit carries the person and the time, deleted records are recoverable, and every document ever issued stays filed against the movement that produced it.

One source of truth

A figure changes in one place and everything generated afterwards reflects it. No reconciling five files that disagree.

Documents generated, not typed

The paperwork comes out of the data. Nobody retypes a weight from one form onto another, so nobody transposes one either.

It feels like a spreadsheet

Deliberately. Nobody has to be persuaded to adopt it, sent on a course, or talked out of their own way of working.

Starts with whatever you have

An item can begin as a line and a supplier name and fill out as the detail arrives. You don't need clean data before you can start. Requirements get added at the point of rumour.

Fully managed

We own and run the whole stack. You hold no vendor contracts, chase no licences and administer no integrations.

Crystalline V1 — the starting point, not the product

V1 is our logistics system. It was built for offshore supply movements and it's the foundation every client build starts from.

Nothing in it assumes a particular region. Locations, vessels, units, document formats and terminology are all set up per client, whether you're running out of Peterhead, Esbjerg or Jebel Ali.

Watch what happens before a vessel sails. The same item gets typed into a requisition, then a load list, then a manifest, then a crane operations list, then a dangerous goods declaration. Five times, by hand, under time pressure, from five slightly different spreadsheets. Every retype is a chance for a weight to drift, a UN number to transpose, or a piece count to stop matching the deck.

Then the vessel comes back and it runs in reverse. Something was landed, something stayed on deck, something was consumed, something is coming ashore as waste. By the time anyone asks where a particular item went, the trail is four spreadsheets and someone's memory.

V1 takes that data once and generates all of it, with every figure traceable to the record it came from. It was designed around how a shorebase genuinely works: movements, requisitions, instances, containers, splits, backload, remaining on board. If those words already mean something to you, it'll make sense within the hour.

Finished documents land in your own storage, pre-numbered and consistently named, and go out by email at the same time if you want them to. Nobody is naming files by hand or chasing the current version.

What a new client gets is a system that already works on day one, and a starting point that gets shaped to their methods rather than imposed on them. The structure holds across most logistics operations. What changes is the vocabulary, the documents and the rules — and that's the part we build with you.

Entered once, at source. Everything downstream reads the same data.

No rekeying. No drift. One version of the truth.

Beyond cargo

The same pattern extends further. Crew change and flights, drilling operations schedules, certification and inspection registers, waste tracking.

The no-nonsense guarantee

Five commitments. Each one is specific enough to hold us to, which is the point of making them.

01Automation, not AI

AI is why we can build this affordably — it has changed what software costs to make. It does not go anywhere near your operational data.

Your documents are produced by deterministic scripts working from records your people entered. Nothing on them is inferred, predicted, summarised or written by a language model. A figure on a document is the figure in the record, every time, reproducibly. If we ever want to change that, we will ask you first and you can say no.

02We are upfront about what we cannot do

Our user guides carry an honest status against every action, including the ones that do not work yet and the ones with known defects. You will hear what is missing from us, in writing, before you need it — not from a document that came out wrong.

03We never make your operational decisions

Where we display information that bears on a decision — weather, sea state, lifting data, certification status — we show the provider and the time it was issued, and we leave it at that. We do not calculate whether something is within limits and we never will.

That judgement belongs to the qualified people making it, with their own tools and their own accountability. We do not remove accountability. We enable it.

04You have a say in how it is built

The inputs, the outputs and the tools underneath are all shaped during scoping. Bring your document formats, your reporting needs, your IT team's view on where data sits — the more that lands before we build, the less gets changed while you're waiting to use it.

05Your data leaves with you

It is your operational record, not ours. It is exportable in full, in a usable format, at any point — including the day you decide to stop working with us.

And some things we will not do

  • We will never reach out.
  • We will never circle back.
  • We will not look for synergies.
  • Headspace and bandwidth will not be an issue.
  • No photographs of notebooks.
  • No thoughts from an airport lounge.
  • No AI-generated filler.

Documentation people actually use

Most system training walks you through screens. Ours starts from the job.

The V1 user guide is organised the way the work is: pick what you want to do, pick what you want to do it to, read the steps. It also runs as a walkthrough — the whole job in order, from reference data through to close-out — so a new starter can follow the shape of a sailing before touching a record.

Every action carries its current status and, where it applies, what tends to go wrong with it. That candour is deliberate. Anyone learning a system should know which parts to check twice.

COMING SOON

The guide is being finished alongside the current build. It will be published here once the actions it documents have been through end-to-end testing — we would rather it was accurate than early.

If you would like to see it before then, ask and we will send you the current version, defect list and all.

Governance documents

Our information security and data protection policies are published in full.

CRYS-POL-AM-01Access Management Policy

How access is granted, controlled, reviewed and revoked.  Download PDF

CRYS-POL-DP-01Data Protection Policy

How personal data is handled, retained and protected.  Download PDF

CRYS-POL-IR-01Incident Response Policy

How incidents are detected, contained, investigated and reported.  Download PDF

CRYS-POL-BC-01Business Continuity Policy

How service is maintained and restored through disruption.  Download PDF

CRYS-POL-CM-01Change Management Policy

How changes to the platform are proposed, tested and released.  Download PDF

CRYS-POL-VM-01Vendor Management Policy

How suppliers underpinning our platforms are selected and reviewed.  Download PDF

CRYS-DPA-01Data Processing Agreement

Available on request to organisations in discussion with us.  Request a copy

Talk to us

If you recognise the problem in your own operation, we would like to hear about your setup.

Tell us what the work involves, which paperwork causes the most rework, and what you are using today. We will tell you honestly whether we can help.

Registered office

Summit House, 4-5 Mitchell Street
Edinburgh EH6 7BD

Company number

SC900213, registered in Scotland