From 150 users
Below a hundred and fifty users the question is what a platform costs. Above it, the question is who operates it, who decides what, and what happens at seven in the morning when something is wrong. This page answers those.
The operating model
A platform at this size is not a delivery, it is a service with an owner. Four things decide whether that works, and none of them is a feature.
Your own data centre or European cloud – the platform does not require a vendor's hosting. Virtualisation, PostgreSQL, object storage, backups and monitoring belong to the operating model, not to a separate contract nobody reads.
Monitoring of processes, jobs and integrations, agreed response times, a support path with an escalation route. What that costs at your size is scoped individually, because at this size a published tier would be a guess.
Your own modules are tested against each release, not rediscovered during it. That is the return on an architecture where extensions live in the same process as the data instead of beside it.
The same team runs this platform in our own group every working day. Not a demo and not a reference customer – our own books close on it.
Companies and countries
Multi-company is not a switch, it is a set of decisions about what is shared and what stays apart. This is where those lines fall, and what each one costs you in daily operations.
| Area | How it sits on one platform | What that means in operations |
|---|---|---|
| Master data | Shared where it should be, company-specific where it must be | A product is maintained once, not thirteen times – and a price still differs per company |
| Books and closing | Separate ledgers per legal entity | Each company closes on its own; the group view is a report, not a second system |
| Tax logic | Per company and per country, including chains across borders | Triangular transactions and reverse charge are configuration, not a manual correction after the fact |
| Documents across entities | A document can cross a company boundary inside the system | Intercompany sales and purchases stay one process instead of two systems reconciling |
| Languages | Per user, per customer and per document | A customer receives the order confirmation in his language, not in the language of the seller |
| Currencies | Per company, per customer and per document, with rate history | Valuation and reporting stay comparable without an export into a spreadsheet |
| Permissions | Per company, per role and per record | Someone can work in two companies without seeing the third |
Thirteen companies in six countries run this way in our own group today. What that looks like in practice
Governance
At this size the interesting question is not whether the system can do something. It is whether you can still explain, a year later, why it does it that way.
Separated
By people
Complete
Reviewed
How the layers below this are separated from the public internet is part of the architecture. The security architecture
The responsibility model
Not two, and not "jointly". Where responsibility is shared, nobody is called at seven in the morning. This is the split we write down before a project starts.
| Layer | Who owns it | What that means when something breaks |
|---|---|---|
| Platform and modules | Corelane | We are called, we analyse, we fix – including where the cause is in our own code |
| Processes and configuration | Corelane, with your key users | Changes are specified together and built by us; your people decide what the process has to do |
| Data and its quality | You, supported by us | We can show where data blocks a process; maintaining it stays a business task |
| Server and virtualisation | Whoever operates it | Ours if we host, yours if you do – named before the project, not after an incident |
| Network, firewall, workplaces | Your IT or a specialist | Not our trade; we write the requirement and judge the delivery |
| The seam between them | Corelane | Coordinating across it is our job – including while it is still unclear whose side a fault is on |
An installation with eight hundred users as a reference. Our own operation is thirteen companies in six countries, and its weight is in companies, countries, catalogue and automation rather than in seat count. If your number is what matters to you, ask us in the first conversation and you will get that answer, not a case study that stretches to fit.
Where we have not done something before, it is scoped as work rather than sold as experience. That is also why there is no price on this page: at this size the honest figure comes after we have seen your landscape, and the assessment is what produces it.
Where the trades outside the platform end and who we bring in for them is set out on the comparison page.
Working together
How the relationship actually runs once the project is over. Deliberately short, because the value is in it being kept rather than in it being described.
You get the people who build and operate the platform, with name, address and direct dial. At this size one of them is designated for you, and the escalation path above them is written down rather than discovered.
A fixed rhythm with your key users and IT: what happened, what is open, what comes next, what it costs. Decisions are taken there, so the backlog does not become a wish list.
Requests, incidents and projects enter the same intake and are visible to both sides. What is developed is specified before it is built and tested before it is live – the same route for a field label and for a new country.
What we build for you belongs to you: your own modules in readable Python, your database, your server. Changing supplier costs effort, not a rebuild. We would rather be chosen again than be hard to leave.
The next step