Bespoke software

We build the thing no product covers.

Inherited systems, integrations with no ready link, processes specific to your company alone, and the servers everything runs on.

Diagram: the inherited system stays in place, a layer around it connects it to the shop, the warehouse and accounting, and underneath is the infrastructure we run.

Two systems we kept.

Wholesale distributionGermany · ERP in production since 2013

What we found

The ERP exposed no interface of any kind. The vendor had closed years earlier.
Orders were retyped from the shop into the ERP, and stock was retyped back the other way.
Two people spent about half their day on the retyping, and mistakes surfaced at invoicing.

What we left

A layer around the ERP reads its data model directly and exposes a clean interface.
The shop, the warehouse and accounting now read the same numbers from one place.
The ERP was never replaced. It is still the system of record.

Manual re-entry

Years the ERP has run

Weeks to first release

Logistics servicesNetherlands · codebase inherited from a previous team

What we found

The team that wrote the system had left. Nobody in the company could say how it worked.
It ran on one server that had not been updated in three years, with backups nobody had restored.
Every change was postponed because no one knew what it would break.

What we left

The system was read, documented and covered with tests before a single line was changed.
The environment moved to infrastructure that is rebuilt from code, with restores tested monthly.
Changes ship again, in small releases, with a way back from each one.

Weeks to first documented map

100%

Restores tested monthly

Days offline in the move

What the work includes

What we take on

The vendor of the system no longer exists

It stays where it is and keeps running. We build the layer around it that connects it to everything else.

Two systems nobody has ever connected

We write the link between them, including to software with no documentation and no support.

A project abandoned by another team

We take it over, document what it does and carry on. Understand first, change after.

The old system cannot be saved

We build the new one and move the data across with no interruption to the work.

What remains after go-live

Support stays with us. If the system stops, we take up the case within the agreed response time and stay on it until service is restored.

The environment is described in code and rebuilt from nothing. The work does not depend on one person’s memory.

The code, the access and the documentation are yours from day one.

If you have nowhere to run the system, we take that on too. The response times and the scope of support are set out separately.

How we do not work

We do not open with a quote for replacement. If one of our finished products will do the job, we will say so, even when the bespoke project is the larger deal. An inherited system that works is an asset, not a problem.

The first step is a short conversation.

We reply within 24 hours. A detailed brief is not needed beforehand.