The Science of Spending: How Music Psychology Drives Real Purchasing Decisions

Four decades of research shows background music measurably changes what customers buy — a landmark 1.8M-transaction study found the right music lifts sales 9.1%, and the wrong music performs worse than silence. Here's the evidence, and how to use it. VibeBeats is direct-licensed background music for cafés, restaurants, bars, gyms, retail stores, salons, spas, clinics and hotels. No music licensing/PRO fees (*No OneMusic, ASCAP, BMI, PPL PRS or SOCAN licence needed for VibeBeats music.) Other music sources may still require licensing. The AI DJ reshapes each playlist through the trading day.

Most venue owners treat music as decoration. The research treats it as a variable — and when you run it as a variable, it moves money.

The single largest commercial test of this ran across 16 quick-service restaurant locations, 1.8 million transactions and five months. Conducted by HUI Research in 2017 (commissioned by Soundtrack Your Brand, run independently by HUI), it compared brand-fit curated music against random popular hits. Brand-fit music lifted total sales 9.1%. Category by category, the impulse and high-margin lines moved hardest: desserts +15.6%, shakes and smoothies +15%, sides +11%.

The kicker is the control conditions. Random popular hits versus silence produced only +1.2%. And a poorly fitting mix performed 4.3% worse than playing nothing at all.

That is the whole argument in one dataset. The question was never "music or no music". It's "the right music, or everything else".

Forty years of evidence

The field starts in a supermarket.

Milliman, 1982 (Journal of Marketing) played slow- and fast-tempo music in a US supermarket and found slow tempo slowed shopper movement and lifted daily sales roughly 38% versus fast tempo. It's a single-store study from 1982 — frame it as the paper that opened the field rather than a universal law you can bank on. But it opened the field decisively.

Milliman, 1986 (Journal of Consumer Research) took the same idea into a restaurant. Slow tempo lengthened dining time and produced significantly higher bar and beverage spend, while food spend stayed flat. Diners didn't order more food; they stayed and drank. (We're deliberately not quoting exact minutes or dollar figures here — the circulating numbers vary and the qualitative finding is the robust part.)

North, Hargreaves & McKendrick, 1997/1999 produced the study everyone half-remembers. In a supermarket wine aisle, they alternated French and German music on a display of French and German wines. On French music days, French wine outsold German by roughly 5:1. On German music days, German outsold French by roughly 2:1. Then they asked shoppers whether the music had influenced their choice, and around 85% said no. (Ignore the "330%" framing that circulates online; the ratios are the finding.)

Areni & Kim, 1993 put classical music into a wine store and found shoppers bought more expensive bottles — same quantity, higher spend. Premiumisation, not volume. Skip the "3x" figure that gets attached to this study second-hand.

Knoeferle et al., 2012 (Marketing Letters) is the one that should change how you brief a playlist. Testing tempo and musical mode together, they found the tempo effect on sales only held for minor-mode music: slow plus minor was the sales-optimal combination. Tempo advice without emotional mode is incomplete advice.

The emotional triggers underneath

Behavioural results are only useful if you know the mechanism, because the mechanism is what generalises.

Mode drives valence. Major mode maps to positive affect, minor to melancholy and sophistication — confirmed at scale by a 2024 meta-analysis in Physics of Life Reviews. That's why Knoeferle's slow-plus-minor combination works: it slows behaviour while adding perceived depth rather than sadness.

Music bends time perception. Arousal and valence both alter how long an interval feels; work published in Frontiers (2013) found the right music makes waits and queues feel shorter. For any venue with a queue — cafés at 8am, gyms at 6pm, a Friday bar — that's a service-quality lever disguised as a playlist.

Familiarity is a paradox. Yalch & Spangenberg (2000) found shoppers believed they had shopped longer with familiar music, while in fact they shopped longer with unfamiliar music. Which means customer intuition about music is an unreliable guide to customer behaviour — a theme running through all of this.

And it's still producing results. Grossman & Rachamim (Marketing Letters, 2025) found classical music raised the maximum price participants would accept for indulgent products. The premiumisation effect Areni and Kim spotted in 1993 is still replicating in 2025.

What customers themselves report

Survey data isn't experimental data, so attribute it accordingly — these are industry and licensing-body surveys, not peer-reviewed research.

• 66% of UK adults say the music a business plays influences what they buy; 31% are more likely to order pricier menu items with classical playing; 67% admit having left a venue after realising no music was on (PPL PRS / Perspectus Global survey, 2021, n=2,101 — note PPL PRS is a licensing body). • Music is the number one in-store mood improver, positive for 85% of shoppers, while 57% disengage when the music choice is poor (Mood Media global study, 2019, n=10,000+, including international markets). • 41% of customers spent more than planned when the music aligned with the brand (EU Music 360 Living Labs, Finland 2024, published 2025).

Read those against the North wine-aisle result — 85% of influenced shoppers denied being influenced — and the pattern is clear: customers notice bad music and underestimate good music.

The AI music angle

Here's the newest data, and most competitor content hasn't caught up with it.

In blind testing, 97% of listeners could not distinguish fully AI-generated tracks from human-made music (Deezer × Ipsos, October 2025, n=9,000 across eight countries). In 2026 experiments reported by ProMarket (Friedrichsen et al.), listeners rated AI-generated songs higher than human tracks — until they were told the tracks were AI.

Now the honest part: no published field study yet measures the effect of AI background music on in-venue sales specifically. Anyone telling you otherwise is selling you something. What the literature does establish is that the mechanisms — tempo, mode, brand fit, volume, timing — are format-agnostic. They're properties of the music, not of who or what made it. Those properties are exactly what VibeBeats engineers for, and we'd rather say that plainly than invent a study.

The practical playbook: five evidence-backed rules

Fit beats popularity. HUI 2017: brand-fit +9.1%, random hits +1.2%, poor fit −4.3% against silence. Popular is not the same as appropriate. Set tempo and mode together. Knoeferle 2012 — slow tempo only paid off in minor mode. Briefing "slower music" alone can leave the effect on the table. Treat volume as a lever, not an afterthought. Guéguen's 2008 field work in French bars found louder music sped up drinking. We won't quote the exact drink counts, but the direction is well established, and it cuts both ways: too loud costs you dwell time and conversation. Day-part everything. Time perception, arousal and trading rhythm all shift through the day. One playlist for open-to-close is one setting for four different businesses. Measure your category mix, like HUI did. Total sales moved 9.1%; desserts moved 15.6%. If you only watch the top line you'll miss where music actually works — the impulse, high-margin add-on.

For the perception side of this — how sound changes what food tastes like and what a room looks like — read our companion piece on the colour of sound and crossmodal psychology. For tempo specifically, see the BPM effect on motivation and mood. For the retail application, how music increases retail conversion and our guide to fine-dining restaurant music go venue by venue, and what curated music means explains the difference between a playlist and a programme.

The cost side, honestly stated

Every study above assumes you're paying to play. In international markets, recorded background music from a commercial supplier attracts a local PRO licensing licence on top of your platform subscription, re-indexed each September.

VibeBeats' catalogue is directly licensed from artists and rights holders, so there is no local PRO licensing or local composition PRO fee for playing our recorded background music — it's included. The standard caveats hold: live performance, DJs playing commercial catalogues and karaoke still need collecting-society licensing; a single society track from a radio, TV or staff Spotify account re-triggers your venue's obligation; and consumer streaming services prohibit commercial use in their own terms, which no music licence fixes.

Run the experiment on your own room: start a free trial or book a live demo at app.vibebeats.ai/demo.

Sources

• HUI Research / Soundtrack Your Brand (2017), 16 QSR locations, 1.8M transactions: https://www.qsrmagazine.com/sponsored_content/study-shows-brand-fit-music-lifts-sales-9-percent/ • Milliman, R. (1982). Journal of Marketing; Milliman, R. (1986). Journal of Consumer Research. • North, A., Hargreaves, D. & McKendrick, J. (1997/1999). In-store music and product choice. Nature / Journal of Applied Psychology. • Areni, C. & Kim, D. (1993). The influence of background music on shopping behaviour. Advances in Consumer Research. • Knoeferle, K. et al. (2012). It is all in the mix: tempo and mode. Marketing Letters. • Major/minor mode meta-analysis (2024), Physics of Life Reviews: https://www.sciencedirect.com/science/article/pii/S1571064524001672 • Yalch, R. & Spangenberg, E. (2000). The effects of music in a retail setting. Journal of Business Research. • Grossman & Rachamim (2025), Marketing Letters: https://link.springer.com/article/10.1007/s11002-025-09803-4 • PPL PRS / Perspectus Global survey (2021, n=2,101): https://pplprs.co.uk/benefitsmusic/how-can-music-influence-consumer-behaviour/ • Mood Media global study (2019, n=10,000+). • EU Music 360 Living Labs (2024/2025): https://music-360.eu/2025/02/03/music-in-retail/ • Deezer × Ipsos AI music survey (October 2025, n=9,000): https://newsroom-deezer.com/2025/11/deezer-ipsos-survey-ai-music/ • ProMarket / Friedrichsen et al. (2026): https://www.promarket.org/2026/05/04/consumers-prefer-ai-music-until-theyre-told-its-ai/ • Guéguen, N. et al. (2008). Sound level of environmental music and drinking behaviour. Alcoholism: Clinical and Experimental Research.

Start your 14-day free trial — no credit card required.