We didn't start with code. We started with Excel.
Before building tools for others, we built them for ourselves. We worked as employees at different companies, and building tools wasn't our job: it was what we did to get rid of the repetitive work of our day-to-day. That is how Provecto started.
What we saw
We always saw inefficiencies. Everywhere we have worked, we saw colleagues who did their job well but lost a lot of time doing it. They didn't have the training to do it more efficiently, and it never occurred to anyone how to improve it: it was simply done that way. The most advanced tool they used was copy and paste. Technology could do a good part of that work and nobody had shown them how to take advantage of it.
For example, at a travel agency, the internal system was good for invoicing, but not for managing processes. Everything else lived in each person's memory and was spread across loose Google Drive sheets: you had to remember what was missing, note it down and check it again.
First, Excel
We started with what we had at hand: micro-automations in Excel, without a line of code, chaining formulas with logic. You changed the values in a few cells and a full paragraph was generated with the variable words filled in, ready to copy, paste and send. We matched up spreadsheets to spot discrepancies in invoicing. We ended up building an Excel file that centralised much of the day-to-day work, with colours and an interface with buttons. It worked and looked good, but it was still an Excel file: the company needed a real system.
We changed company, not the problem
In time we changed company. Here there was a system, but it wasn't efficient enough: there were still many things left to do by hand. And in many of those tasks there was no human decision to make, because everything followed a well-defined logic. So we connected the processes to automate those tasks. That is where we made the leap from Excel to code.
Judgement first, code second
With AI, the leap came naturally, and that is precisely why we take how it is used seriously. It isn't about asking it for something and waiting to see what comes out. First we decide what has to be done, how, what the flow is, every detail of the work and the edge cases that may appear. Then AI executes: it writes the code following that judgement. We validate it, design the interface and review the result. It is a tool in our hands, not an autopilot.
From code we moved on to desktop tools and then to full portals. Always with the same logic: scattered processes turned into custom-built tools.
Provecto wasn't born from a business idea. It was born from a real need, and from the certainty that what worked once could work for others.
We take what we deliver seriously
A tool that is used every day can't be left half-finished. This is what working well means to us.
First we understand, then we build
We study how your process really works before writing anything: what has to be done, in what order and what can go wrong.
Every decision has a reason
We define the flow, the details and the edge cases. AI executes, but the judgement is ours and we review everything.
We keep going until it really works
We don't consider anything finished because it "looks fine". We wait and keep polishing until we get the result we were aiming for.
We look at every detail
We review everything we produce, including what almost nobody will notice. Details are what make a tool inspire trust.
"Good" isn't enough
If something is good but could be better, we make it better. We hold ourselves to more than what is asked of us, because we work to be proud of the result.
Deadlines that are met
We agree on a delivery date and commit to it. If something changes along the way, we tell you in time.