01Custom software and integrations
Business backends and BI
Some companies have outgrown their software. We build the core system that runs the business—orders, customers, inventory, margins, invoicing—and the dashboard management looks at every Monday, integrated with what you already have and ready for AI to work on it.
Node · .NET · PostgreSQL · SQL Server · AWS · Claude
What we do
The core system
A custom backend with the entities of your business and the rules that govern them. No more, no less.
Integrated with what’s already there
The ERP, the store, the warehouse, the bank. The new backend doesn’t replace anything that works: it connects it.
Dashboard
The metrics that matter, updated on their own, with today’s figure next to last year’s.
Metrics you ask for in one sentence
“Margin by product family, this quarter.” The AI builds the query; management looks at the result.
What it looks like in practice
- A company with four spreadsheets that don’t add up ends up with a single source of truth and a dashboard.
- On Monday, management opens their phone and has sales, margin and outstanding payments, without asking anyone for them.
- A software company’s client defines their own KPIs by writing what they want to see.
What you need
- A clear idea of what the business is: which entities, which rules, who decides.
- The source data, even if it’s messy. Cleaning it up is part of the job.
- Management that wants to look at the numbers every week.
A case
The same logistics software company
- What was there
- Their clients tracked shipments in an old application and asked for metrics by email.
- What we did
- The tracking application rebuilt from scratch, multi-company, with built-in reports, and a dashboard where each client defines their metrics by writing what they want to see.
What is a business management backend, and when do you need one?
A management backend is the central system that runs a business—orders, customers, inventory, margins, invoicing—with the company’s own rules, plus the dashboard (BI) that management checks. You need one when the company has outgrown its software: spreadsheets that don’t add up, the same data in three places, reports someone puts together by hand every week. Sanfeliu builds it integrated with what is already there (ERP, online store, warehouse, bank) and ready for AI to work on it: metrics can be requested in a single sentence.
- Technology: Node or .NET, PostgreSQL or SQL Server, AWS, and Claude for natural-language queries.
- It doesn’t replace the ERP or anything that works: it connects them.
- Metrics update on their own, with today’s figure next to the one from a year ago.
- First step: a free fit session; then an audit with a fixed quote.
- Case: for a logistics software company, a dashboard where each customer sets up their own KPIs by writing what they want to see.
Questions we get asked
- We already have an ERP. Why would we need a backend?
- Because the ERP does what an ERP does, and a business usually has rules that don’t fit in it: per-customer pricing, its own processes, outside data. The backend handles that and leaves the accounting to the ERP. If your ERP already covers everything, we’ll tell you in the fit session.
- What is a dashboard (BI)?
- A screen with the metrics that matter—sales, margin, receivables, inventory—updated automatically and compared with the previous period. It replaces the report someone puts together by hand every Monday.
- Can we ask for new metrics without writing code?
- Yes. With AI connected to the backend, “margin by product family this quarter” becomes a query and a chart. We did this for a logistics software company so each of its customers could build their own KPIs.
- How much does it cost?
- We quote per feature, at a fixed price, after the audit. The free fit session is for finding out whether the problem calls for a backend or for something else.
- Is messy data a problem?
- It’s part of the job. Cleaning and consolidating it is the first thing we do, and it’s usually where the company learns the most.
How we work
Let’s start by seeing if we’re a fit.
Eight questions and a call. Free, and we tell you what we think even if the answer is no.
It’s a call with the engineer who would lead the project, not with a salesperson.