What still sells when AI replaces competence
Knowledge now sits in every chat window. What you can still sell is time you take off other people’s hands. My learning center failed on exactly that: 0 bookings.
01
The moment
29 October 2026. The learning center with daily courses shows 16 accounts, all of them test or bot accounts, and 0 bookings. That day I shut it down. The whole story is under Repeat.
I tell you this first because it concerns you directly. If you are an employed or freelance marketer who wants out and does not yet know what to sell, the obvious thought is: I sell what I know. Courses, workshops, consulting. For years that was a good path, and right now it is the path on which it is easiest to fail.
02
What went wrong
I wanted to sell knowledge at a time when knowledge costs nothing anymore. What you used to buy at a high price now sits in every chat window, free or for a few euros a month. AI is not replacing a task here and there right now, but the factor competence itself. Whoever keeps selling only knowledge is selling something whose price is heading to zero.
And I solved a problem people did not feel. The people I wanted to reach did not want to learn something every day. They wanted their problem solved: why are no enquiries coming in? Which advertising works?
I could have checked both beforehand, and my own first rule says exactly that: research demand first. I did not keep it. How long the learning center ran and what it cost, I will add under Repeat.
03
What I tried
What I am trying since is the counter-design. Instead of teaching others how to build a marketing data warehouse with agents, I build and run it myself, for a flat fee. The client does not have to learn anything. At the same time this site is content only again: I do not write here to sell knowledge, but so that you can see how I work.
Whether the counter-design holds I cannot say yet, because it is only just beginning.
04
What I know now
What still sells is time. Not the knowledge of how to do something, but that it is done. Taking the work off the client’s hands completely, instead of explaining how to do it themselves.
Whoever sells knowledge sells appointments and consulting hours. Whoever sells time sells a result that is simply there. AI alone does not deliver that, because someone has to build, run and answer for the system.
Whom you sell this to, I am testing right now. My assumption: people who run a business close to people and have no time for marketing, data or technology. Doctors, physiotherapists, tradespeople, studios. They do not want to learn how to build an agent. They want the phone to ring. They have an urgent, conscious problem and already spend money on it today, often on agencies billing by the hour.
For your offer this means: do not look first at what you know, but at what you can take off someone’s hands. Ask yourself: what is different for the client when I am done? Not what they know, but what they no longer have to do. If you have no clear answer, your offer is still a knowledge offer in new packaging.
Honest prices get easier because of this. Whoever sells time can say: this costs this much, and for that it is done. No timesheet, no discussion about why something took longer. The efficiency AI brings then stays with you and not with the timesheet.
Knowledge does not disappear in all this. Whoever takes on work has to understand it. That is why I keep learning, only with a different aim: not to sell it, but to deliver better.
They do not want to learn how to build an agent. They want the phone to ring.
05
Open question
Does getting things done stay sellable for good? If agents at some point not only know but execute on their own, maybe only responsibility is left, meaning someone who stands behind it running. How big that business is, I do not know yet.
06
Changelog
First version. Nothing here is rewritten silently: every later change gets its own entry in this changelog.