Why Spotted by Locals Doesn’t Use Algorithms

Founders Sanne & Bart‘s frustration with travel websites and apps that work with algorithms was one of the main reasons for starting Spotted by Locals in 2008. Algorithms have become a whole lot “smarter” since and they can do some awesome things, but we think they don’t work for Spotted by Locals. Not now, and probably never!

Why? We’ll try to explain in this article.

But first: what’s an algorithm?

An algorithm is a series of instructions a computer follows to perform a task or solve a problem. The computer gets an input, performs the instructions, and provides an output. An example is any search engine, like Google: typing “buy socks online” in the search bar is the input. The algorithm is what makes Google search its database for items relevant to your query. The Google page with a list of online shops where you can buy your socks is the output.

Algorithms and travel websites

Before the Internet era, traveler mostly relied on word of mouth, friends and family recommendations, or paper guidebooks. Especially in cities, paper guidebooks were often out of date, and “hidden gems” became famous and thus touristy quickly.

A year after TripAdvisor was founded (2000 – 4 years before Facebook and Yelp), it changed from an aggregator of guidebook reviews to a website where travelers could post their own reviews. It’s hard to imagine now, but it was really revolutionary! Most people thought they would once and for all resolve the mentioned issues users of traditional guidebooks had.

Content on Tripadvisor grew exponentially after that decision. Right now there are 859 million reviews. On all travel websites with user generated tips and reviews like Yelp, Google Local and Foursquare, algorithms decide what to show you when you look for recommendations.

The way algorithms rank spots

Every website has its own algorithm to decide which spots to show. On Tripadvisor for example, the criteria are quality, recency, and quantity. The algorithm analyzes reviews on all activities or places that match your query. Then, it ranks them according to the number of good reviews that are recent enough. For example, a place with 100 5-stars reviews but most of them are from 2016, will come after a place with 70 4-stars reviews but all from 2020. Also, a place with 500 reviews and an average of 4.3 stars, will place better than a place with an average of 5-stars but only 50 reviews.

Sites like Yelp, for example, take even more things into consideration when giving a score to a business. They analyze the user as well, taking into account how often it uses the app and its credibility as well. The credibility factor means checking that a user is a real one and not a bot. Another thing modern algorithms do is to analyze the contents of the review, making sure it’s relevant for users and businesses. The goal here is to filter out reviews that are too generic and that don’t tell much about a place.

Our issue: algorithms lead to predictable recommendations

This way of ranking and showing content to users has been perfected over years and is very efficient and valuable. It works fine for most people! The content you see as a user has been chosen for being the “best rated” by a number of people worldwide. This means you will see the average things everybody enjoys, the most liked places. The places you’ll see are of good quality, but they often cater their offer to tourists.

The more interesting and edgy spots – or at least what we consider more interesting at Spotted by Locals – are usually somewhere, but buried deep and are rarely chosen by the algorithm to appear.

The result is that people travel to many different cities, but have the same standardized and predictable experience in each one of them. That makes travel into a kind of mundane activity that follows rules and schedules that apply to everyone.

Leaving Your Comfort Zone

Cox18 squat Milan (by Susanna Baggio)

We at Spotted by Locals see things in a different light. To us travel is about leaving your comfort zone, learning about different cultures, and growing as a person. Not about seeing “the most beautiful thing” or collecting the tourist highlights.

We want to encourage people to really explore, dig deeper into a destination and do the things that algorithms would probably think you would not want to do. On our website and in our apps you will find content in purely chronological order. It’s up to our users to find and choose the perfect gems to add to their journey.

One example out of the 12.000 up-to-date spots our locals in 80+ cities currently recommend: Cox18 Milan. A squat with amazing punk gigs way out of the tourist center. Definitely not for everybody, but if you’re into this kind of spot you could have one of the most memorable Milan nights ever thanks to our Spotter Susanna.  Not on Tripadvisor, no reviews on Yelp, mostly very old reviews on Foursquare.