Business systems
The internal software a company runs on — order handling, scheduling, reporting, whatever the spreadsheet has quietly become. Built to be operated by your staff, not by us.
Four kinds of business software, and the handover package that ships with all of them.
Business software rather than consumer apps. Unglamorous systems that a company depends on and that have to keep working.
The internal software a company runs on — order handling, scheduling, reporting, whatever the spreadsheet has quietly become. Built to be operated by your staff, not by us.
Getting two systems that were never designed for each other to exchange data reliably, including what happens when one of them is down. Most integration work is really error handling.
Removing the manual steps that exist because somebody once did them by hand. We look for the ones that are frequent and boring, not the ones that are interesting to build.
Moving data off a system that has to be retired, and maintaining what exists. Neither is glamorous; both are where the risk usually sits.
Four things we decline, so nobody wastes a call finding out.
A fixed price on an undefined scope is either padded or a future argument. We scope first, then price.
No licence-back, no escrow, no undocumented dependency on us. You can always leave.
We deliver work, not headcount by the month. Renting developers by the day is somebody else’s business model.
If existing software does the job, we say so, even though that ends the conversation.
Small increments, visible early, and no reveal at the end.
What the process actually is, watched rather than described. Specifications written from descriptions build the wrong thing accurately.
The smallest useful piece that can go live on its own. If nothing can, the scope is wrong and we say so.
Working software in short cycles, in your hands throughout. Surprises are found while they are still cheap.
Code, documentation, tests and a runbook. Then we are available, not necessary.
Four things that ship with every project. Agreed at the start, so they never become a negotiation at the end.
Not a zip file. The full repository with its commit history, so the next developer can see why something is the way it is.
A short written record of the architectural decisions and the options rejected. This is the document that saves the most time later and is skipped most often.
Automated tests and instructions to run them. Without tests, the next change is a gamble and everyone starts being afraid of the code.
How to release it, how to tell it is healthy and how to undo a bad release at two in the morning.
Standard scope — extended where a system warrants it
You do, from the first commit. The repository is yours, the history is yours, and there is no licence back to us hiding in the contract.
Then the handover package should make that unremarkable. If leaving us is painful, we did the job badly.
No. They are part of the work, not an upsell. A system without them is unfinished, not cheaper.
If you want us to. Many clients run it themselves, which is the point of the handover package.
Yes. We start with a written assessment of what is there before promising anything — inherited systems reward honesty about their state.
Registered in Cyprus, working across the EU, in English and German.
Describe the process and what currently goes wrong. If a spreadsheet or an off-the-shelf tool would do it better, that is what you will hear.