Custom software development
CRMs, ERPs, portals and internal apps built around how you work, with AI where it helps. You own the code.
When is custom software development worth it?
It pays off when your process is what sets you apart and no product on the market covers it without forcing it. The typical sign: you pay for three or four subscriptions, stitch them together with spreadsheets and someone still spends hours every week reconciling data by hand.
It does not pay off if a SaaS product solves 90% of what you need. In that case we recommend configuring it well and closing the gap with a business process automation, which costs far less. We make that call with you during AI consulting, with the numbers on the table.
02What kind of custom software do we build?
We focus on management applications for small and mid-sized companies, where the value lies in fitting your process and saving your team time:
- Custom CRM: records, opportunities and follow-up with the fields and stages of your sales process, not someone else's.
- Light ERP: orders, stock, production or projects, connected to your accounting software.
- Client portals: your customers check orders, download documents or open tickets without calling.
- Internal apps: work reports, inspections, checklists or resource booking, mobile included.
- MVP: a first version of a digital product to validate with real users before you invest more.
How do we add AI to a custom application?
AI enters as one more feature of the application, at the points where your team reads, writes or searches. For example: upload a contract and have the record fields fill themselves, search years of case files with plain-language questions, summarize a customer's history before a call, or let an AI agent create orders from an email.
We design these features so the AI model can be swapped without rewriting the application, and so sensitive data stays inside your infrastructure when it does not need to leave.
04Who owns the code?
You do, from day one. The repository, domain, database and cloud provider account are created in your company's name, and we work through delegated access.
We use standard, widely adopted technologies, with documentation and tests, so any other development team can continue the project if you ever decide to. We do not use proprietary platforms that would force you to stay with us.
05What technology do we build with?
We work with TypeScript, React and PostgreSQL, a widespread stack that any web developer knows. We host with providers that have EU data centers when the application handles personal data, and we add automated tests on the critical parts so future changes do not break what already works.
For AI features we use the OpenAI, Anthropic or Google APIs through our own abstraction layer, so switching provider is a configuration decision and not a new project.
How we do it, step by step
- Discovery
We interview the people who will use the application and map the real process.
- Clickable prototype
You validate screens and flow before we write a line of code.
- First version
We build the core that already saves work and put it in your team's hands.
- Iterations
We add features by priority, with a delivery every two weeks.
- Support and evolution
Maintenance, backups and improvements under an optional monthly contract.
| Off-the-shelf SaaS | Custom software | |
|---|---|---|
| Upfront cost | Low | Fixed budget per phase |
| Monthly cost | Per user, grows with the team | Server and maintenance, fixed |
| Fit with your process | You adapt | It adapts to you |
| Integrations | Whatever it offers | Whatever you need |
| Ownership | You rent access | The code is yours |
| Time to use | Days | 8-12 weeks |
Frequently asked questions
How much does custom software development cost?
It depends on the number of areas, user roles, integrations and the scope of the first version. We give you a fixed price per phase, agreed before we start, after a free first call and discovery. Hosting and third-party services are paid directly to the providers.
How long does it take to build a custom application?
A first useful version is usually in use within 8 to 12 weeks. We prefer to ship something your team uses early and improve it every two weeks over spending six months building without feedback.
Is custom software better than an off-the-shelf ERP?
If your business fits a standard ERP, the standard product is almost always better. Custom software wins when your process is different, when you pay for many licenses covering features you do not use or when you need integrations the standard product does not allow.
What happens if we stop working together?
You keep everything: code, data, servers and documentation. Because we use standard technologies, another team can continue development. We help with the handover if you need it.
How do I choose a custom software development company?
Check who owns the code, whether the stack is standard, how often they deliver and whether they show you a prototype before building. Ask for a fixed price per phase. We put the code, data and servers in your name, and deliver every two weeks.
Is maintenance included?
The first months after delivery include bug fixes. After that we offer an optional monthly contract with security updates, backups, monitoring and hours for improvements.
Keep exploring
AI consulting services
AI consulting that doesn't stop at a slide deck: strategy, pilot, scale and operation with one team.
Business process automation with AI
AI workflows that read, classify and log data across your tools, with human review on exceptions.
AI agent development
Assistants that answer on WhatsApp, web or phone and act inside your CRM, calendar or ERP.
AI document processing
AI extraction, classification, summaries and validation for contracts, delivery notes, forms and ID documents.
AI sales and CRM automation
Lead capture, enrichment and scoring, follow-ups, meetings and an up-to-date CRM, with AI.
AI for small business
Where to start with AI in a small business, what it costs and how to move in phases without risk.
RAG (Retrieval-Augmented Generation)
RAG (Retrieval-Augmented Generation) is a technique where an AI system first searches your documents or databases for relevant information, then generates its answer using only that content instead of relying on what the model memorized.
LLM (large language model)
A large language model (LLM) is an AI model trained on enormous amounts of text to predict the next word, which lets it understand questions and write, summarize, translate, classify and reason about text.
Pick a time yourself
30 minutes by video call. Free, no commitment.
Prefer us to get in touch? Leave your details.