you have about two years to show up
machines are becoming the front door, and a machine cannot recommend what it cannot read.
two things a founder said to me back to back, and it took me a day to notice they were the same thing.
the first was about websites. he sells the infrastructure that large multi location brands run their websites on. he says people are still buying a redesign. they look at the site through the eyes of a human visitor, they care how it feels, and there is nothing wrong with that except that it is no longer the main event. his bet is on the structure underneath, because the primary reader now is a machine deciding whether to recommend you to somebody who asked it a question.
that changes what counts as a mistake. the popup you find mildly annoying makes an agent give up and leave the site. the booking widget you rented from a third party is a wall. the duplicate pages from an seo campaign three agencies ago are noise. he says every brand he takes on arrives carrying years of hacks and a lot of trauma, and that most of the job is deciding what to delete.
the second thing was about founders. he is unapologetic about personal brand and he thinks you have less than two years to either prove what you have done or talk about what you are doing. hiding behind the camera and pulling strings, in his words, is making it harder on yourself for no reason at all.
same claim, twice.
machines are becoming the front door, for brands and for people, and a machine cannot recommend what it cannot read. for most of history, being quiet was neutral. you were simply not in the room, and the room was small anyway. now silence is a data problem. when somebody asks a model who they should talk to about a thing, you are either in the answer or you are not, and the thing deciding is reading whatever you left lying around.
this is exactly why i publish every conversation i have. not because the view counts are impressive, they often are not, and my dad reminds me of this. because five hundred conversations sitting somewhere a machine can read is a different kind of asset than five hundred conversations that happened on a video call and evaporated.
the work you did quietly this year, that nobody wrote down anywhere, might as well not have happened by next spring.
if somebody asked a model about your field tomorrow, would your name come up?
the machine economy brief
one email when it matters: bitcoin, ai, robotics, and what founders should do about it. unsubscribe anytime.