Management platforms
Internal systems that replace the spreadsheet tangle: your own database, role-based access, and a trail of who changed what.
Tailor
The software your company needs, not the one that happened to exist.
We map how your operation runs, find where the return is real, and build it. The team that designs it is the team that writes the code.
Why is this possible now?
Custom software used to be a capital project: months of development and a budget that only closed at large companies. So almost everyone ended up adapting to off-the-shelf tools, changing their process to fit the software instead of the other way round.
That changed. The cost of building dropped enough that the question is no longer "can I afford it?" but "is this process distinctive enough to be worth it?".
What can be automated changed too. It used to be fixed rules only: anything off-pattern went to a person. Today a system can read a text, decide with nuance and escalate when appropriate. That opens up processes that were not candidates two years ago.
The process, in detail
Five stages. Each has a concrete deliverable, and each can be the last one if continuing does not make sense.
Discovery
Interviews with the people doing the work, not only management. We look at the systems you already use, where copy-paste happens, and which questions nobody can answer today.
Clarity
We prioritise by impact against effort and say what should be left out. This is where the conclusion is sometimes "fix the process before automating it".
Design
Scope by stages, data model and integrations. Stage one is designed to be useful on its own even if stage two never happens.
Build
Built in production with your real data, in short increments you can see working. No single delivery at the end.
Adoption
Team training, adjustments based on real usage, and measurement of whether it moved the number we set out to move.
What we build
Everything below is work we have actually done, not a catalogue of what we might attempt.
System integrations
Connecting tools that do not talk to each other, without waiting for each vendor to schedule it on their roadmap.
Internal agents
For admin, analysis and support: they read the data, act through tools, and escalate to a person when appropriate.
Process automation
Documentation, repetitive data entry, reconciliation — anything done by hand today with volume and stable criteria.
Dashboards with natural-language querying
Ask the database a question in plain language and get the answer, instead of waiting for someone to build the report.
How we quote
By stages, with stage one designed to work on its own. If you decide to stop after the first delivery, you are not left with half a system — you are left with something useful.
The quote comes after discovery, not before. Quoting without having looked at the operation is exactly how projects end up at double the estimate.
We do not work by loose hours or blocks of hours. Each stage has written scope and a fixed price.
Concrete numbers come out of the first call, which is free — and you leave with a written plan whether we continue or not.
Architecture built for what comes next
Everything we build sits on a centralised database you own. That is not a technical detail: it is the difference between answering a new question next year and having to integrate everything again.
Functionality is exposed as well-defined operations rather than buried inside a screen. That is what lets an agent operate them later without rewriting the system.
The practical consequence: stage four is cheaper than stage one instead of more expensive. Projects not designed this way get more expensive as they grow.
Case studies
⚠ PENDING (§10.2): cases with publishable metrics are missing. A case without a number is not published, so this section stays empty until there is real, client-approved material.
Frequently asked questions
How long until it is running?
Discovery and prioritisation take two to three weeks. After that it depends on scope, but stage one is designed to be in production in weeks, not months. If someone promises you a complete system in two weeks without having seen your operation, they are selling hype.
What if we cannot maintain it afterwards?
That is the right objection and we answer it head-on. The code and the data are yours, we leave documentation and we train your team. If nobody in the company can maintain it and you do not want a support contract, we say so before starting — it is one of the reasons we sometimes recommend not building.
Do you work with the tools we already use?
Yes, and that is the norm. Most projects do not replace the existing stack: they integrate it and build custom only the layer where your differentiator lives. Replacing everything is almost never the right answer.
What happens to our data?
It stays in your infrastructure or one you control. We define at design time which data each component touches and what is out of scope. If the project involves sensitive data, that is agreed in writing before any code is written.
Do we need a technical team?
Not to start. It does help to have someone on your side who knows the process well and can make decisions: that person matters more to the project than a technical profile.
What if discovery concludes we should not build anything?
We tell you, and you keep the map anyway. It has happened and it will happen again. Charging to build something we know will not pay back is exactly what we criticise about the rest of the market.
Apply this to your company
We can look at whether this makes sense for your operation — and if it doesn't, we'll tell you.