Tourist traps and fake reviews: how to spot them in the ratings

What a trap looks like in the numbers

A tourist trap rarely has a terrible rating. It sits where crowds pass — beside a landmark, on the main square, outside the station — so it collects a lot of reviews, and a steady flow of one-time visitors keeps its average from collapsing. The pattern to look for is a large review count combined with a rating clearly below what is normal for that city and that kind of place.

"Normal" varies more than most people expect. Among the 9,778 places on our lists in 92 cities (snapshot of 13 October 2026), the median rating is 4.5 and 59% are rated 4.5 or higher. In Japan the median is 4.4 and only 38% reach 4.5; in Thailand the median is 4.7 and 87% do. The type of place matters as well: restaurants on our lists have a median of 4.6, shopping places 4.3. A 4.2 restaurant with 6,000 reviews next to a famous square in Italy is well below the local norm; the same 4.2 is far less unusual in Japan.

These figures describe the places that made our lists, which already lean towards well-reviewed places. Use them as a yardstick for comparison, not as the average of every business on Google Maps.

What research has found

The clearest evidence that reviews push tourist traps out of business comes from Rome. Dante Donati, a professor at Columbia Business School, used the end of mobile roaming charges in the European Union in 2017 as a natural experiment: from one day to the next, visitors from other EU countries could check reviews on their phones at no extra cost. Combining Tripadvisor ratings with Italian social-security records, the study found that afterwards revenues and employment at higher-rated restaurants rose by about 4–10%, and low-rated restaurants became roughly twice as likely to close. Low-rated restaurants that survived hired more experienced, better-paid staff, and their ratings eventually improved. The paper was published in Marketing Science in 2026.

A second finding concerns who writes the reviews. A study in Information Systems Research by DaPeng Xu and colleagues compared the same reviewers on a large Chinese review platform when they reviewed restaurants at home and when they travelled: as tourists, people were at least 13.4% more likely to give a higher rating than as locals. The authors link it to the mood of travel — people away from home are more forgiving and pay more attention to atmosphere than to price. The data come from China, so the exact size of the effect belongs to that setting; the direction is the useful part.

Put together: reviews do discipline bad restaurants, but only where visitors actually read them, and a place reviewed mostly by visitors can look better than locals would rate it. Reading the reviews written in the local language, which are more likely to come from residents, is a sensible correction — though that is our inference, not something the study tested.

Fake reviews: the scale of the problem

Google publishes a yearly count of the policy-violating reviews it blocks or removes on Maps, and the total has grown in each of the last three reports. It says the vast majority of those caught in 2024 were removed before anyone saw them — but the totals show how much effort goes into faking ratings, and that some fakes will get through.

Google's systems look for patterns such as one reviewer posting the same review on several businesses, or a business suddenly receiving a flood of 1- or 5-star reviews. When a spike of spam reviews hits a profile, Google says it may pause new reviews and show a banner explaining why, and in 2025 it began warning on Maps when it has recently removed suspicious 5-star reviews from a profile, starting in the US, the UK and India. If you see either notice, treat the rating with extra caution.

Faking reviews is also increasingly illegal. In the United States, the Federal Trade Commission's rule on consumer reviews and testimonials took effect on 21 October 2024. It bans creating, selling or buying fake reviews — including ones attributed to people who don't exist or to people who never used the business — paying for reviews that express a particular sentiment, positive or negative, and suppressing negative reviews through threats or intimidation, and it lets the FTC seek civil penalties against knowing violators. It is a US rule; other countries have their own.

Warning signs in a place's reviews

No single sign proves that reviews are fake; look for several at once. The table separates the signals backed by a source from the ones that are only common sense.

• Sort by newest: ratings lag behind a change of owner, chef or prices.
• Read the 2- and 3-star reviews — they often weigh what was good against what was not.
• Switch to reviews in the local language, or translate a few.
• Open the profiles behind two or three of the most glowing reviews.
• Compare the rating with the local norm for that type of place, not with 5.0.

On the street: rules of thumb, not research

Travellers share plenty of signs of a tourist trap. We know of no study that has tested them, so treat them as reasons to check the reviews before sitting down, not as verdicts:

• Staff outside pulling passers-by in, especially right next to a major sight.
• A menu without prices, or priced "by weight" or "market price" with no way to check before ordering — Australia's Smartraveller advises always asking for the menu and checking prices first.
• Menus in many languages but none in the local one.
• Dishes from several unrelated cuisines on one menu.
• Bread, starters or drinks arriving that you didn't order — ask whether they are charged before accepting, as Smartraveller also recommends.

One rule of thumb with an exception: photo menus

Photo menus are often cited as a sure sign of a trap. That holds poorly outside Europe. In Japan, picture menus and plastic food models in the window are ordinary even in places that serve almost only locals, and photo menus are common in much of Southeast Asia. There, a picture of the food tells you nothing about the kitchen; the reviews and the local norm in the table above tell you more.

Scams governments warn travellers about

Some traps are not restaurants at all. Australia's Smartraveller and the US State Department keep lists of scams that target visitors, and several happen exactly where people look up ratings:

• Overcharging in bars and restaurants: you are invited to order without seeing prices, or served food you didn't order and charged for it when you leave (Smartraveller).
• The friendly invitation: a stranger invites you for a massage, to a tea house or to practise English in a café or bar, then you get an inflated bill and are not allowed to leave until you pay (Smartraveller).
• Taxi detours: unlicensed drivers at airports and attractions offering inflated flat fares, or friendly drivers offering cheap tours that end at shops paying them a commission. Use licensed taxis and agree the fare, or the meter, before you get in (Smartraveller).
• Street games: shell and card games in tourist areas where accomplices win at first to draw players in (US State Department).
• Card skimming: devices on ATMs capture card data and PINs; tourist areas are popular targets, so prefer well-lit indoor ATMs and cover the keypad (FBI).

What our rankings can and can't catch

Our lists are built from Google's rating and review count. By default they are sorted by number of reviews, which shows what is popular — the order in which a busy tourist trap can sit near the top. Switch to "Best" for our score. Its core blends each place's rating with a global average of 4.0 counted as if it were 300 reviews, so a place needs a few hundred reviews before its own rating dominates: a 5.0 from 12 reviews comes out at 4.04, a 4.1 from 6,000 at 4.10. On top of that come a bonus for reach, which grows with the number of reviews (about 0.5 points at 300 reviews, 0.75 at 6,000), and a small one for recent momentum.

What that catches: a handful of fake 5-star reviews barely moves a place with hundreds of genuine ones, and in "Best" a new place with a perfect score from a few accounts can't jump to the top (in "Highest rated", which uses Google's rating as it is, it can). Reviews that Google removes also drop out of the rating we read at the next refresh.

What it can't catch: we don't read review text, so copy-paste wording, reviewer profiles and the language mix are up to you — that is the checklist above. We can't tell tourists' ratings from locals'. The reach bonus favours places with many reviews, so a busy place with an ordinary rating can still outrank a smaller, better-rated one: before any momentum bonus, a 4.2 with 6,000 reviews scores above a 4.7 with 300. A large, sustained fake-review campaign that Google misses raises our score just as real reviews would. Our "Hidden gems" list deliberately favours high ratings with 40 to 2,500 reviews, counts low enough that a batch of fake reviews can move the average noticeably, so read those reviews with extra care. And the momentum bonus, worth at most 0.15 points, rewards places gaining reviews quickly — usually a good sign, but a burst of fake reviews would earn it too.

Use the rankings to build a shortlist, then spend two minutes on the reviews of the places you actually plan to visit.