Does background music make people drink more? What the research actually found, and what it doesn't

Some of it, sometimes. The study everyone quotes found young men in two French bars ordered 3.4 drinks when the music was loud and 2.6 when it was normal, and finished each glass about three minutes faster. Slow music keeps people in their seats longer. Music that fits the product changes what people buy. But a 2024 restaurant experiment found slow music kept tables 40% longer without raising the bill, and a 2022 review of 56 studies concluded it's the choice of music that matters more than the volume or the speed. Here's each study properly, and how to test it on your own till.

Every pub has a volume argument. The bar staff want it up, the dining table in the corner wants it down, and the regular at the end wants whatever was on in 1987. Somewhere along the way someone always says "loud music makes people drink more, it's been proven". It has, sort of. In one small study. So I went back to the actual papers.

The studies, one line each

Study	Where	What changed	What happened
Milliman, 1986	A US restaurant	Slow vs fast tempo	Slow music: longer meals and higher bar purchases
Areni and Kim, 1993	A US wine shop	Classical vs Top 40	Classical: people spent more, by picking dearer bottles, not more of them
North, Hargreaves and McKendrick, 1997	An English supermarket	French accordion vs German Bierkeller music	French music: 40 French bottles to 8 German. German music: 22 German to 12 French
Guéguen and colleagues, 2008	Two bars in western France	72 dB vs 88 dB	Loud: 3.4 drinks vs 2.6, about 11.5 minutes a glass vs 14.5
Malcman and colleagues, 2024	An Italian restaurant in Tel Aviv	Slow vs fast vs the usual playlist	Slow: tables stayed 40% longer than fast. Bills: no significant difference
Trompeta and colleagues, 2022	Review of 56 hospitality and tourism studies	All of the above	How the music is chosen matters more than whether it's on, and preference beats volume and speed

Loud music and faster drinking: Guéguen, 2008

This is the one behind the "turn it up" advice. Nicolas Guéguen's team watched 40 men aged 18 to 25 in two bars in a medium-sized town in western France, across three Saturday nights. They only counted people who ordered a 25cl glass of draught beer, a bit under half a pint. The music was Top 40, at either 72 decibels, which they called normal, or 88, which they called high.

At the loud level the men ordered an average of 3.4 drinks against 2.6, and took about 11.45 minutes a glass against 14.51. The researchers offered two reasons: loud music raises arousal, and it makes conversation harder, so people talk less and drink more. It was published in Alcoholism: Clinical and Experimental Research in October 2008.

Now the honest bit. Forty people, all young men, two bars, three nights, one drink. That's a small study, and it measured how fast people drank, not whether the pub took more money across a week. Nobody's shown the same thing in a British local on a Wednesday with a mixed crowd. I'd treat it as a real effect on a lively young drinking crowd, and not much more than that.

There's a second problem the paper doesn't mention, and it's yours. 88 dB is a lot. Under the Control of Noise at Work Regulations 2005, an employer has to act once a worker's daily exposure reaches 80 dB, the upper action level is 85 dB, and the exposure limit is 87 dB. Those are averaged across the working day, but if your bar sits at 88 dB for a full shift, your staff are over the legal limit before you've counted a single extra pint. Louder for the trade is not a free lunch.

Slow music keeps people in their seats: Milliman, 1986

Ronald Milliman's restaurant study, in the Journal of Consumer Research, is the other classic. He switched between slow and fast music on Friday and Saturday nights. With slow music, diners took significantly longer over their meals, and the charges at the bar were higher. Food spending didn't change much. The full paper is behind a paywall and I haven't been able to check its own figures, so I'm not quoting the numbers you'll see on supplier blogs.

The 2024 Tel Aviv study ran the same idea properly, with 282 tables at an Italian restaurant and days randomly assigned to slow, fast or the normal playlist. Slow-tempo tables stayed an average of 80.3 minutes against 57.3 for fast, 40.2% longer. The bill didn't move in any way that counted, about 207 shekels for slow, 197 for fast and 209 on the normal playlist, and the researchers say tempo had no significant effect on it. So slow music held people longer, and that was it.

That's the trade-off a pub actually faces. Slow music on a busy Saturday when you need the tables back is costing you covers. Slow music on a quiet Tuesday when the room's half empty and you want people to have one more costs you nothing, and might earn a round.

Music that fits changes what people choose

The wine studies are the ones I find most useful, because they're about choice, not volume.

In 1997 Adrian North and colleagues at Leicester put four French and four German wines next to each other in an English supermarket, matched on price and sweetness, and played French accordion or German Bierkeller music on alternate days for two weeks. On French days the shop sold 40 bottles of French wine and 8 German. On German days, 22 German and 12 French. When they asked shoppers afterwards, only 6 of the 44 who answered said the music had influenced their choice. People didn't notice it working on them.

Areni and Kim, in a US wine shop in 1993, played classical or Top 40. Classical didn't make people buy more bottles. It made them pick dearer ones. In pub terms, that's the difference between more glasses of wine and better glasses of wine, and the second one is where your margin is.

The review that pulls it together

In 2022 Maria-Angeliki Trompeta and colleagues pooled 56 studies with 209 effects from hospitality and tourism settings in the Journal of Business Research. Their conclusion was that it's not so much the presence of music as the design of it that influences customers, and that whether people like the music matters far more than the physical side, the volume and the tempo. The Tel Aviv authors also point to it finding no significant effect of tempo on overall spend.

Put all six together and the honest reading is this. Volume can speed up drinking in a young, lively crowd. Slow music holds people longer, but holding them doesn't automatically mean they spend more. Matching the music to what you're trying to sell can shift choice. And playing what your actual customers like beats any trick with the dial.

What I'd actually do with it

Time and crowd	Try	Why
Quiet midweek evening, a few regulars	Low volume, slower tempo, stuff they like	You want them to stay for one more. Nobody needs the table
Food service	Quieter, conversation level	Holding tables is good until you need them back. Diners mostly come to talk
Busy Saturday, younger crowd	Up, but within limits	The one setting the Guéguen study fits. Keep an eye on staff exposure
Pushing wine by the glass	Something that suits the wine, not the jukebox	The wine studies are about choosing dearer, not drinking more
Last half hour	Down	Helps the room wind down and makes last orders an easier conversation

One thing to keep in mind if you go loud. Faster drinking means people get drunk faster, and selling alcohol to someone who's drunk is an offence under section 141 of the Licensing Act 2003. The extra round isn't worth a review hearing.

And whatever you play, background music in a pub needs the right licence, which for most pubs is TheMusicLicence from PPL PRS. What it costs in 2026 is in the pub music licence guide. Buying a track or a streaming subscription for yourself doesn't cover playing it to customers.

Test it on your own till, not on a study

None of these studies were done in your pub, so don't take mine or anyone's word for it. Run it like any other change. Pick one quiet night of the week. For two weeks keep the music as it is. For the next two, change one thing only, the volume or the tempo, and leave everything else alone, same staff, same offers, same football.

Then compare the till by hour and the pints actually poured from your stock movement. The till tells you whether people stayed and when they left. The stock count tells you what really went over the bar, which matters because a change in the room is exactly the kind of thing that makes comps and half-pours drift and nobody notice. A quick count of your three busiest lines at open and close does it. If the number doesn't move, the music wasn't the problem, and you can stop arguing about the volume. If you're trying to work out whether a quiet night is worth opening at all, that sum is in should a pub close on quiet nights.

My take

Music is the cheapest thing in the building to change and the easiest to overthink. The research is real but small, and the effects are about pace and choice, not some hidden lever that makes people spend more. Play what your customers like, keep it down when people are eating or talking, let it lift when the room's busy and young, and keep your staff's ears inside the law. Then let your stock count tell you if any of it made a difference.

Common questions

Does loud music make people drink more?

In one small study, yes. Guéguen and colleagues (2008) found young men in two French bars ordered 3.4 drinks at 88 dB against 2.6 at 72 dB, and drank each glass about three minutes faster. It was 40 men on three nights, so treat it as an effect on a lively young crowd, not a rule.

Does slow music increase bar spend?

It keeps people longer. Milliman (1986) found slower music meant longer meals and higher bar purchases. A 2024 restaurant experiment found slow music kept tables 40% longer than fast, but with no significant difference in the bill.

How loud is too loud for bar staff?

Under the Control of Noise at Work Regulations 2005, an employer must act at a daily exposure of 80 dB, the upper action value is 85 dB, and the exposure limit is 87 dB, averaged across the working day. A bar held at 88 dB for a whole shift is over the limit.

What music sells more wine?

Music that fits the wine. In a 1997 English supermarket study, French music days sold 40 French bottles to 8 German, and German music days 22 German to 12 French. In a US wine shop, classical music led customers to choose more expensive bottles, not more of them.

Do I need a licence to play background music in a pub?

Yes. Playing recorded music to customers needs a licence, which for most pubs is TheMusicLicence from PPL PRS. A personal streaming subscription or a bought track doesn't cover public play.