The Familiarity Trap

There is a study from 2000, by Yalch and Spangenberg, that should have changed how every retailer thinks about background music.

Shoppers exposed to familiar music perceived that they had spent more time in the store than they actually had. Shoppers exposed to unfamiliar music perceived that they had spent less time. The familiar music group left sooner. The unfamiliar music group stayed longer.

Familiar music made people feel like they’d been there forever. They left.

If you are a retailer using licensed catalog music, you have two options and both of them cost you something.

You play music your customers recognize. The hits, the deep cuts from the right era, the songs that match your brand because they already exist in the culture your customer lives in. Mood Media will tell you they have access to over a hundred million tracks. Their curators sort by genre, tempo, mood, brand fit. The problem is the one Yalch and Spangenberg identified. When a customer hears a song they know, their brain starts predicting the next note, the next phrase, the chorus. A 2015 systematic review confirmed this: highly familiar music captures attention and pulls the listener out of the present task. Their mind is finishing the song instead of shopping. The time distortion kicks in. They feel like they’ve been there longer than they have. That perception accelerates their departure.

You played the right song for the brand. The customer liked it. They left faster because of it.

Or you play music nobody recognizes. Deep catalog. Obscure tracks. Music that fits the sonic profile but that no customer has ever heard.

The mere exposure effect, first documented by Zajonc in 1968 and replicated across hundreds of studies, demonstrates that humans have a strong baseline preference for things they have already encountered. A meta-analysis of 208 experiments confirmed it as one of the most reliable findings in social psychology. The mechanism is processing fluency: your brain processes familiar stimuli faster, that speed generates a positive feeling, and you attribute the feeling to the stimulus. You don’t think “I’ve heard this before so my brain is working less hard.” You think “I like this.”

Flip it around. When a customer walks into a store and hears music they have never encountered, their nervous system withholds the comfort signal. The song doesn’t register as dangerous. It registers as nothing. No recognition, no fluency. In a retail environment where every other sensory element has been designed to produce approach behavior, the music is producing a gap. The customer won’t tell you the music was bad. They won’t mention the music at all, because the mere exposure effect operates below conscious awareness. They’ll just feel slightly less settled than they should, and the store will underperform against what the merchandise and the staff and the lighting were set up to deliver.

Familiar music accelerates perceived time and pushes people toward the door. Unfamiliar music withholds the comfort signal that every other element in the store is trying to build. Most retailers don’t know they’re making this choice. They hand the playlist to a provider, the provider builds something that sounds right, and nobody connects the music to the behavioral data because the two systems don’t talk to each other.

The word “familiar” is doing a lot of work in the research, and it’s worth pulling apart. Yalch and Spangenberg mean songs the listener has heard before. Specific recordings, specific melodies. The brain recognizes the song as a discrete object it has previously encoded, and that recognition triggers the prediction cascade. You hear the verse and your auditory cortex starts constructing the chorus before it arrives. That predictive process is what distorts time.

But there is another kind of familiarity. Stylistic familiarity. A song you have never heard that uses the harmonic language, the production texture, the groove conventions, and the tonal palette of music you grew up with. Your brain processes it with fluency because the components are familiar, even though the composition is not. You get the comfort of recognition without the prediction cascade. The processing is easy. The time distortion doesn’t kick in.

Every song in a catalog is either known or unknown to any given listener. There is no in-between. You are always on one side of the trap or the other.

Music generated for a specific customer profile, built from the sonic vocabulary of that profile’s taste formation window, produced with the conventions their nervous system already processes fluently, but containing no melody or lyric or recording any of them have ever encountered. The familiarity is structural. The novelty is compositional. The brain gets the fluency signal. It does not get the prediction cascade. It does not fast-forward.

The research has been sitting there for twenty-five years. Nobody could act on it until the music itself became a variable you could engineer.