I run four companies in sectors that have nothing to do with each other: marketing, weddings, real estate and software.
I'm not a guru. I don't sell smoke. I don't have a rented Lamborghini.
What I have is four businesses running without me being there. And a method that repeats in all four.
My edge: I don't look at your business from inside your industry. I look at it from the outside, with patterns I've already seen work in others. And that's where the decisions you weren't seeing appear.
With SEB Creativos I've worked on more than 400 projects over the last 10 years. Each one has taught me something. Each sector has its logic, its client, its way of making money. I've seen many.
Five qualifications between MBAs, master's degrees and specializations. But what really teaches is invoicing the following month.
And for the last two years, everything I build runs on artificial intelligence. Agents that reply, qualify and get work done inside my own companies, every day, with real clients on the other side. Software that used to take six months of development and now takes a couple of weeks.
If your company doesn't work when you're not there, you have a job. Not a business. That's what I come to fix.
You come in with a problem. You leave with a plan.
You execute. I review, correct, and push.
If you don't want to work with me directly, you might need what my teams do.
Marketing and AI agency. Ads, branding, copy, audiovisual production, custom development and AI put to work inside the business. For brands that want to sell, not just look pretty.
We plan weddings in Marbella. Couples who don't want a generic wedding. Neither do we.
Wholesale real estate for investors. No decorative middlemen. No Instagram tours. Real deals.
Software for event professionals. CRM, projects, community and an AI working inside: it replies, qualifies and moves work forward. Because running an events business with 6 different tools is an extreme sport.
What I apply in mine. The same thing you'll apply in yours.
If a process depends on you, it's broken. Document, delegate, and let the system do the heavy lifting for you.
You don't hire hands. You hire heads. If you have to approve every decision, you don't have a team: you have expensive executors.
What can't be measured can't be improved. And what can't be improved gets eaten by the competition sooner or later.
If a machine can do it, stop doing it yourself. A well-placed agent works at night, never forgets the follow-up and doesn't ask for holidays. That's the difference between scaling or burning out.
«It's not about working more. It's about building something that doesn't need you to keep running.»
No mile-long forms. No silly questions. Just enough to know if we fit and get back to you fast.
I reply personally within 48 hours. If you don't hear from me in that time, something broke.
With the most repetitive and least important process. That is where you learn how a model behaves with your data and your people without risking a client. Starting with the critical process “because that is where it shows” is also where the first failure shows, and that failure kills internal adoption for a year. I go into it in AI for business: what works and what is hype.
Anything that is boring, repetitive and verifiable at the same time. Boring so nobody misses it, repetitive because if it happens twice a month the saving will not pay for the setup, and verifiable because you have to be able to tell in one second whether it went well. The third condition is the one almost everyone skips, and it is the one that decides whether you sleep at night. The method is in what to automate first and what not.
It comes down to a single number: how many hours a month it takes today. If a process eats sixteen hours a month, almost any setup pays for itself in weeks. If it eats two, it probably is not worth it and what you need is a calendar reminder, not a project.
Not your team — the tasks your team does reluctantly and without judgement, yes. In practice the same team handles more volume without new hires, and that is only good if you had already decided what you wanted them to do with the time freed up.
If the problem is that nobody fills in the CRM, no AI bolted on the outside will fix it: you need one that receives the data on its own. If the problem is that you want a summary, do not change CRM. The question that separates a product from a marketing layer is whether the AI writes to the CRM or only reads it. I go into it in CRM with AI: what actually changes.
It is a published scale, V0 to V6, for stating precisely how much autonomy an AI agent acting on your systems actually has. It is useful for two very practical things: demanding a specific level from a vendor in writing instead of a vague “it has AI”, and not fooling yourself about your own system. It lives at the Vergara Scale, with a DOI.
If your question is not here, write to me and I will answer it.
The tool was never the problem. The problem is that nobody has decided what the machine does and what you keep doing yourself.
That gets decided in one afternoon. And it changes the next two years.