Ongoing · Bolivia
A rented audience, and how to start owning one
An established FM station with an enormous social following and comparatively little audience of its own — and a web platform built so the station has somewhere of its own to send people.
The problem
The station's reach is real and it is large. Almost none of it belongs to the station. It is measured in followers on platforms that own the relationship, decide who sees a post, and can change the rules on a Tuesday without telling anyone. A rented audience converts poorly into an owned one, and it converts worst at exactly the moment you need it to.
That matters commercially, not just philosophically. Advertisers buy against numbers a station can show them, and the numbers this station could show were somebody else's numbers. Meanwhile the thing a listener searching for the station would most want — the live stream, the schedule, the presenters — sat behind a site that search engines could barely read.
The approach
The work came in two halves, and the first was not code. I produced an analysis of the station's public presence and then three competing landing-page directions rather than one recommendation: one built around the live player, which is what people already arrive for; one built around the on-air personalities, the most underused brand asset the station has; and one built around a visible commercial door for advertisers, because that is where the revenue actually is.
Three, deliberately. Each implies a different site, a different measure of success and a different amount of work from the station's own teams — and that choice belongs to the station, not to the person building it.
The build is Next.js on the App Router, server-rendered, with Supabase behind the content and a streaming embed for the live player. Server rendering is the load-bearing decision here: a station whose whole presence lives on social is invisible to search, and a client-rendered player would have kept it that way. This is the one case study on this site where the SEO reasoning is the deliverable rather than a property of it.
The ongoing half is content operations — schedule, integrations and technical documentation kept current, working bilingually with internal teams and external suppliers across two time zones.
Where it stands
Ongoing. The platform is live and maintained, and the strategic question it was built to answer is not the kind that a launch closes. Converting a rented audience into an owned one is measured in months of the station having somewhere of its own to send people, and then in what those people do when they get there.
What has changed is that the question can now be measured at all. Before this there was no owned surface to attribute anything to.
Concretely, that means three things worth watching: whether the site starts appearing for the station's own name and for the presenters by name, whether the live player is reached directly rather than through a social post, and whether an advertiser can find the commercial page without being sent a link. None of those were measurable before, which is a smaller claim than a launch usually makes and a more useful one.
What it runs on
Where to next
Want the same for yours?
Tell me what you are trying to build, or what is not working now. Working out the shape of the problem is part of the job.