How do we know which places are good for … ehm anything

Written by Richard on in Top 10 lists, Triposo.

Our apps are used by hundreds of thousands travelers who trust their mobile phones to tell them what to do and where to go. We love to look at the statistics of how these people travel with our apps since it helps us understand how we can make our product better. What we see is pretty exciting. Here are some stats that paint a picture of the Triposo traveler:

  • They don’t plan ahead much. The most popular time to look up restaurants is at 19:00 which is when they start to get hungry.
  • When they look at places to eat they only look at 4-6 different options.
  • In less than 10% of these cases they take the trouble to look up the Yelp reviews we provide for the place they pick.

This means Triposo travelers live in the now. They check the places to eat when they get hungry and pick one that is close by after looking at just a few. This way of traveling happens to coincide with how we think you have most fun when traveling. Don’t plan too much - just go!

So what does this mean for our app? One of the things it means is that we should make the choice of picking the place to go to as simple as possible. Travelers should not need to wade through 100 reviews to figure out if a restaurant can do a tenderloin. Instead there should just be a “Great for Steak” badge for each place that makes a terrific T-bone.

We hand out badges for many different dishes which our users can select from. It’s quick, easy and it works. But where do we get our badges?

  • Let’s start at the beginning: 9 open content sources form the basis for a travel guide with over 500,000 places of interest.
  • Using our sources we also compile a list of over 1,000 popular dishes from around the world.
  • We then crawl social media and travel-related websites and the wider web and cross-reference each source with our database of places and dishes. Over 5 million opinions are analysed in this process.
  • Rather than base our ranking on these ratings, we use natural language analysis to mine opinions in order to determine how positive the references to a specific dish in a specific restaurant are. Then we have our badges.