Finding a place to live and avoiding being surrounded by tourists all the time can be quite a challenge in popular destinations like Lisbon.
But how do we define some locations on a map as touristy and others not?
Can we measure the level of “touristy-ness”?
What would be the formula?
One of my favorite websites to browse when traveling to a new city is Hoodmaps. It's a crowdsourced map where people vote for different areas in the city and give them one of 5 labels:
This is what Lisbon looks like:
Of course, it’s a very oversimplified but it gives you a nice overview of which area you might want to book an Airbnb or where to search for hipster cafés.
Tip: If you are looking for more detailed analysis, check out this Lisbon neighborhood comparison.
Based on the voting crowd of Hoodmaps, these are the touristy 🔴 neighbourhoods in Lisbon:
Does this mean that when you cross the border of the red area, there won't be tourists, only hipsters with their flat whites? Not really.
So what influences the density of tourists in specific areas?
Perhaps we can somehow track the mobile devices of tourists, mobile operators definitely have data that would shed more light on this.
In Lisbon, it could be influenced by the locations of tuk-tuk stations and popular “miradouros” (viewpoints) that contribute to creating a “tourist” hub.
🕵️ In researching these questions, I found a few usable datasets. Let's add them to the map and see what we find.
When we add all of the official hotels in Lisbon to the map, we can see that most of them are outside the touristy areas crowdsourced by Hoodmaps.
We can assume also, that people staying in these hotels are spending at least some part of their trip mingling in surrounding neighborhoods to their accommodations.
InsideAirbnb allows anyone to download all the Airbnb listings in a specific location and add them to a map.
What you can see is that there is no strong correlation between hotel locations and Airbnbs.
However, while Airbnb rentals are distributed across the whole city, there is a very high concentration of them located in the central neighborhoods of Misericórdia and Santa Maria Maior.
These neighborhoods, unfortunately, became something of an amusement park for tourists, kicking out the locals to more remote areas of the city.
Tuk-tuks (or rickshas) have become a symbol of mass tourism in Lisbon.
Cruising around the historic neighborhoods they are, to put it simply, destroying the vibe of the city. But the city hall somehow hasn’t banned them and we can see here that the density of tourists around official tuk-tuk stations (red dots) is very high.
Try to google “the best places to see in Lisbon” and you will find thousands of articles with recommendations. I chose this one from the Culture Trip and added the attractions (purple dots) to the map (some of them were outside of Lisbon).
To make this more accurate I'd next time combine places from another 3-4 of the most popular articles to add more data points.
While googling more info about this topic, I discovered that a few scientists are already trying to research the digital footprints of tourists. They analyzed data from public photo websites like Flickr/Panoramio and tried to create heat maps of places with high concentrations of tourist activity.
This can also be applied to tweets or any data tourists share during their stay.
Defining some neighborhoods as touristy and others as not touristy is a very tricky task. With crowdsourcing tools like Hoodmaps, people vote for a calm neighborhood like Ajuda as touristy just because it's close to Belém. But Ajuda is not touristy at all.
Belém is famous for multiple “must-visits places” like Torre de Belém, Pastéis de Belém, and monastery Jerónimos. Tourists are heavily concentrated around these attractions but the rest of the neighborhood is a great and peaceful place to live.
We can definitely consider the central neighborhoods Misericórdia and Santa Maria Maior as very touristy. But the same can be said for plenty of other neighborhoods surrounding them.
All these visualizations show us areas with high concentration of tourists. But can we somehow combine them all together?
The first step would be to assign values to each data point: a hotel = 25, an Airbnb listing = 1, etc.
This way we could calculate a Tourism Density Score for any pin you click on a map. The score would be calculated by the relative distance from data points and their values.
It could look like something like this:
Imagine you can get this score for every listing on Airbnb or any apartment you want to buy or rent.
And while measuring this score over time, you can see the historic data and trends. It could be a great tool for journalists and city planners.
If you expected some conclusion, there isn’t one. I'm just passionate about this topic and wanted to share my exploration so far with you. Once I figure out more, I'll share it on this blog.
If you have some ideas about this topic, feel free to reach me out 🤙
I'm the founder of Surf Office, a one-stop shop for anyone organizing company retreats.
A few side-projects I'm involved in: Hoodpicker, Surfpreneurs Club and Epic Monday.
I write about my experiments that combine hospitality, real estate and tech.
You can reach me out on Twitter and Linkedin.
No Playbook 🤷
Subscribe to my newsletter where I share experiments of building hospitality products.