There’s something special about talk radio. Especially late night talk radio.
Not podcasts. Not polished video essays. Proper radio. Crackly phone lines, awkward pauses, callers who may or may not be entirely trustworthy, and presenters who somehow manage to sound both reassuring and slightly unhinged at the same time, perhaps taking the listeners further down the rabbit hole.
I’ve loved it for years.
Growing up in the North East, Alan Robson and Night Owls on Metro Radio was a genuine institution. It was a rite of passage for friend groups to ring in - fun fact legendary repeat caller - Johnny from Crawcrook was in the year above me at school. Everyone knew it. Even if you didn’t listen regularly, you knew of it. You’d hear people talking about a caller the next day at work or school like it was some major cultural event.
It was ghost stories, strange encounters, local drama, heartbreak, conspiracy theories, people oversharing wildly at midnight, and occasionally towards the end of its run, somebody from Gateshead talking about how the lampposts and crows were killing people with 5G.
And somehow it all worked.
You’d end up completely engrossed in somebody else’s perspective for ten minutes at a time (maximum of one call per week per person). Sometimes they sounded believable. Sometimes they absolutely did not. But that was never really the point.
The point was hearing people trying to explain their experience of the world.
I actually came to Art Bell much later. I found clips of Coast to Coast AM on YouTube years afterwards and got completely hooked listening while coding surveys at work. Which, in hindsight, probably explains a lot about me as a researcher. Fun fact - I was listening to it so constantly I was actually pulled up on my bandwidth use, which, you know - was an awkward conversation.
There’s something oddly hypnotic about listening to somebody talk in detail about a UFO sighting or a never ending hole in a yard somewhere in the Pacific Northwest while you’re building survey logic.
But again, it wasn’t really about whether the stories were true.
It was about how people tell stories when they care about something deeply. The weird little specific details they include. The bits they skip over. The way emotion creeps in unexpectedly. The pauses. The contradictions. The confidence people have in things they absolutely shouldn’t be confident about.
That’s user research.
Honestly, I think that’s a huge part of why I ended up loving qualitative research so much.
The best interviews don’t feel transactional. They don’t feel like data collection. They feel like you’ve been temporarily invited into somebody else’s reality.
A good interview is rarely just “Question. Answer. Insight.”
The worst interviews I’ve been part of are when somebody is more or less just reading a survey aloud.
It’s tangents. It’s half-finished thoughts. It’s somebody suddenly remembering a frustrating experience halfway through another story. It’s hearing the emotion underneath the practical problem.
People will tell you what they do in a survey.
They’ll tell you why they do it in conversation.
And sometimes they’ll tell you something completely different from what they originally thought they believed.
Human messiness matters.
Which is why I remain deeply sceptical whenever somebody confidently announces that synthetic users are about to replace qualitative research.
AI can absolutely help researchers. It already does. High level summaries, clustering, helping with synthesis, bouncing off awkward turns of phrasing so people understand what I’m trying to say, sense-checking, drafting, unnecessary em-dashes, brilliant. Genuinely useful.
But there’s a world of difference between generating probable responses and sitting with an actual human being while they explain something frustrating, funny, irrational or unexpectedly emotional.
Synthetic users can give you averages of data that already exists.
Humans are storytellers.
And stories are where the important stuff is hiding.
Nobody stayed up listening to late-night paranormal radio because they were conducting a peer-reviewed search for objective truth - although I still do listen to paranormal talk radio whilst writing user research analysis.
They listened because human beings are fascinating.
Give them the space to talk, and then listen to them.