When no tool on the market fits your process
Building your own software is expensive to maintain and easy to justify badly. So the first thing we do is look for a reason not to build it. When we find none, then it is worth it.
Requests
Reports
- Request 1NewIn progress
- Request 2In progress
- Request 3Waiting for customer
- Request 4Completed
Built for your process
What you get
A system built for your process
Internal systems, portals, back offices or dashboards that follow the process the company really has.
The code and the documentation
The code is yours, it sits in a repository of yours, and the documentation is delivered with it.
Role-based access and a log
Role-based access, a log of who did what and the data in one place, with verified backups.
How it unfolds
Look for a reason not to build
We first see whether buying off the shelf and adapting solves it. Only when it does not is it worth building.
Assessment and written scope
With the process mapped and the scope written down, the price is fixed before starting.
Build in phases
One part working in production as early as possible, and the rest afterwards.
What we do not promise
- We do not build when buying off the shelf and adapting works out cheaper. We say so before proposing a quote.
- Custom software has an ongoing maintenance cost. It is not a one-off purchase.
- If the scope changes midway, that is a new and explicit conversation, not a surprise invoice.
More detail
The right question is not ‘how much does it cost’
It is this: what does your company do differently, and is that an advantage or an accident?
If it is a historical accident, the cheap route is to change the process and use an off-the-shelf tool. If it is what sets you apart, adapting to generic software means paying to become the same as everyone else.
Most companies are in the first case. That is why most companies should not commission software. It is worth saying this on a page that sells custom software.
When building really pays off
The process is the product. The way you handle requests, prioritise work or calculate prices is what makes customers choose you.
Nobody on the market has this problem. Specific sectors, their own regulation, unusual business models. There is software for clinics and for garages; there is none for what your company does.
The cost of licences has overtaken the cost of building. It happens with large teams and tools charged per user. Beyond a certain size the sums reverse.
You need to give access to people outside. Customers who want to see the status of the process, partners who need to submit something. Portals are hard to buy ready-made and relatively simple to build.
How we build
In phases, with one part working in production as early as possible. Not because it is fashionable, but because the first version of any system is wrong somewhere, and it is better to find that out with a process in use than with the whole system finished.
We use boring technology on purpose. Relational databases, languages that have been around for decades, simple hosting. Custom software lives for ten years, and what matters is that five years from now someone can open the code and understand what is there.
It is from the experience of keeping systems running, and not only of launching them, that we say what we say about maintenance cost.
What we are always asked
When is it NOT worth building?
When your process is common and there is off-the-shelf software that covers 80% of it. In that case, the right thing is to buy and adapt the remaining 20% with automation or integration. Building from scratch just to have the last 20% made to measure is the most expensive decision you can make, and we say so before proposing a quote.
Will we become dependent on you?
The code is yours, it sits in a repository of yours, and the documentation is delivered with it. If one day you want to hand maintenance over to another team, what they need is there. Technical dependency is a way of tying clients down that does not interest us.
How much does it cost to maintain?
Custom software has an ongoing cost: hosting, security updates, fixes and further development. It is a budget line, not a one-off purchase, and we talk about it in the proposal and not six months later.
Tell us what is holding back your operation.
Describe the problem in your own words. You do not need to know what has to be built.
- We read your request.
- We reply within one hour, by email or telephone, in English or in Portuguese.
- If it makes sense, we arrange a conversation.