Airbnb guests next to 30+ bars mention noise 2.8× as often

What we found

• 2.8× as many noise mentions next to nightlife. 2.5% of reviews mentioned noise where no bar was within 300 m, and 7.2% where 30 or more were, rising with every step in between.
• It held in all 16 cities with enough listings in both groups to compare. The gap was widest in Dublin, at 10.3×.
• Most guests still don't mention it. Even next to 30 or more bars, 92.8% of reviews say nothing about noise. A lively street shows up in some reviews, not most of them.
• Noise goes with lower ratings. Listings where a quarter or more of reviews mention noise average 4.70 out of 5, against 4.82 where none do — more below.
• Noisiest stays: Mexico City, at 8.4% of reviews. The quietest was Berlin, at 2.7%.

City by city

The share of reviews mentioning noise with no bars within 300 m, and with 30 or more, sorted by the difference. The small bars show all five groups, from none to 30+.

“Too few listings” means fewer than 1,000 reviews in that group. In some cities few places are mapped as bars — in Italy, for instance, many bars are mapped as cafés — so hardly any listing has 30 nearby.

Where guests mention noise most

Every review in each city, whatever is around the listing. This counts all kinds of noise — traffic, neighbours, thin walls — not only nightlife.

1. 1Mexico City8.4%
2. 2Lisbon5.9%
3. 3Madrid5.8%
4. 4Barcelona5.7%
5. 5Istanbul5.6%
6. 6New Orleans3.9%
7. 7Dublin3.9%
8. 8London3.8%
9. 9New York City3.8%
10. 10Athens3.7%
11. 11Rome3.7%
12. 12Prague3.6%
13. 13Paris3.6%
14. 14Budapest3.4%
15. 15Bangkok3.2%
16. 16Edinburgh3.1%
17. 17Tokyo3.0%
18. 18Vienna3.0%
19. 19Amsterdam2.7%
20. 20Berlin2.7%

Noise and ratings

Listings whose guests mention noise more often are rated lower. Among 98,779 listings with at least 10 English reviews in the year, those where no review mentioned noise averaged 4.82 out of 5 overall. Those where a quarter or more of reviews did averaged 4.70. The drop showed in 20 of 20 cities.

For scale: Airbnb asks for an overall rating of 4.8 or higher for Superhost status.

Bars cut both ways. Guests staying near them rate the location higher and the stay a little lower:

Ratings are each listing's all-time averages on Airbnb; the noise shares cover the last 12 months. This shows a link, not that the noise caused the lower rating.

What guests hear

On quiet streets, guests who mention noise most often name traffic and neighbours. Next to 30 or more bars it is the street outside and the nightlife: 22.0% of noise reviews there name bars, music or partying, against 4.7% where no bar is within 300 m.

Share of noise reviews naming each source. A review can name several. Next to 30 or more bars, 34.8% name none we could pick out (“it was noisy at night”).

When guests mention it

Not in summer. Averaged across the cities, reviews written in August mention noise least (3.6%) and those written in April most (4.6%). August came in below the city's own yearly share in 15 of 20 cities, most of all in Rome, Athens and Madrid.

Airbnb guests have 14 days after checkout to write a review, so the month a review was written is close to the month of the stay.

What hosts near nightlife can do

A lively street is part of why guests book a central stay. The complaints come when it is a surprise, or when it keeps them awake.

• Say it in your listing. “Lively street with bars nearby; bedroom faces the courtyard” lets light sleepers choose, and sets the expectation for everyone else.
• Make sleep easy. Earplugs and a white-noise machine by the bed, and blackout curtains. With Guestate's Shop the Stay, guests can buy the ones they liked.
• Put it in the house manual. Which windows to keep shut at night, how the air conditioning works with them closed, and when the street quietens down. Our house manual template has a place for it.
• Check your own street. Our neighbourhood report counts the bars within 300 m of any address, using the same map data as this study.

How we measured it

• Listings and reviews: Inside Airbnb's latest snapshot of each city, and every review from the 12 months before it. We kept reviews written in English, by a simple test for common English words.
• Nightlife: places mapped in OpenStreetMap as a bar, pub, nightclub or beer garden, read from OpenFreeMap's tiles (© OpenMapTiles). We counted those within 300 m of each listing and put the listing in one of five groups: none, 1–4, 5–14, 15–29 or 30+.
• Mentioning noise: a review counts if it uses one of these words — “noise”, “noises”, “noisy”, “loud”, “loudly”, “louder”, “thin walls” — without a negation just before it (“not noisy”, “no noise”) or after it (“the noise wasn't an issue”). “White noise” and “noise-cancelling” don't count. We left out “party”, which mostly means the guests' own group.
• Checked by hand: of a random 150 reviews the rule counted, 130 really did raise noise as a problem or a warning. Of 75 it set aside as negated, 61 were rightly set aside; most of the rest were warnings phrased with a “not”, such as “if you don't mind noise”. So the rule errs in both directions, and the shares above are close estimates rather than exact counts.
• Sample sizes: a group needs at least 1,000 reviews before we report its share.
• Ratings: each listing's overall rating and location score from Inside Airbnb's detailed listings, for listings with at least 10 English reviews in the period, so their share of noise mentions rests on more than a handful.
• Sources: each review that mentions noise is sorted by the words in its sentences — bars, music and nightlife, traffic and transport, neighbours and the building, construction, appliances and plumbing, the street outside — and can land in more than one. Of a random 100 sorted by hand, 80 were sorted right. Half the misses were neighbours described in words the rule doesn't catch (“people walking above”), so neighbour noise is, if anything, undercounted.
• Seasons: the month each review was written, with every city counting equally, so the cities with the most reviews don't set the pattern on their own.

Limits worth knowing:

• It shows a link, not a cause. Streets with many bars also tend to be central, busy with traffic and lined with older buildings.
• Airbnb places each listing on the map up to 150 m from its real address. Counting bars around a blurred point mixes neighbouring groups, which narrows the gap between them rather than widening it.
• English reviews only, and a word-based rule: a guest who writes “we couldn't sleep” without naming the noise isn't counted.
• OpenStreetMap is mapped by volunteers, more thoroughly in some cities than others.

Use these numbers

Our counts are free to reuse under CC BY 4.0: quote them, chart them, build on them, with a credit and a link back to this page. The tables, city by city, as CSV:

• Noise mentions by bars nearby
• Ratings by noise mentions and by bars nearby
• What noise reviews name
• Noise mentions by month

The underlying listings and reviews are Inside Airbnb's, under CC BY 4.0, and they ask that their data isn't republished — so the files hold our counts only. Bars and pubs from OpenFreeMap © OpenMapTiles, data © OpenStreetMap contributors.