How many places do you visit anyway?

The study, titled “Evidence for a conserved quantity in human mobility’ is published in Nature Human Behaviour is based on analyses of 40,000 people’s mobile traces collected in four different datasets.

It is also the first of its kind to investigate people’s mobility over time and study how their behavior changes.

Behind the project are Dr. Laura Alessandretti and Dr. Andrea Baronchelli, researchers in the Department of Mathematics at City, University of London, together with Professor Sune Lehmann from DTU Technical University of Denmark and the research team from Sony Mobile Communications.

“We first analysed the traces of about 1000 university students. The dataset showed that the students returned to a limited number of places, even though the places changed over time. I expected to see a difference in the behavior of students and a wide section of the population. But that was not the case. The result was the same when we scaled up the project to 40,000 people of different habits and gender from all over the world. It was not expected in advance. It came as a surprise,” says Dr. Alessandretti.

Old places disappear

The study showed that people are constantly exploring new places. They move to a new home, find a new favorite restaurant, find a new bar, or start going to another gym, etc. However, the number of regularly visited places is constantly 25 in a given period. If a new place is added to the list, one of the places disappears.

The pattern is the same when the researchers divide the locations into categories based on how often and how long time they spend at the location.

“People are constantly balancing their curiosity and laziness…

That is by John Stevenson, via K., here is the original research by Laura Andressetti  On this question I retain an open mind.


Leave it to an academic to 'discover' that people go to work, visit friends and come home.

Sorry, five academics.

That's not conserved, obviously- a hundred years ago it took one academic.

magnificent publish, very informative. I wonder why the other experts of this sector
don't understand this. You must proceed your writing.
I'm confident, you have a huge readers' base already!

I think their result is more constrained by time limits than they realize.

If you did a time study, you could answer "In a given day, how many hours do you use, anyway?" If you use an hour a new way, then it appears that also you stop using an hour an old way?

In the same way, if you have time for X amount of visiting in your life, adding a new place to visit regularly would seem to naturally require you to subtract another place. In other words, without efficiency increases, your personal time is zero sum.

I don't have to go anywhere, I live in America--the best place. People file lawsuits like Trump v. Hawaii to come to me.

I wonder what it was like 30 years ago, before the internet?

Pretty much like today, without it.

I wish there was a like button for this post. That must have been how they felt 30 years ago.

They had like buttons 30 years ago, but sadly they had no use for them.

They had, like, buttons 30 years ago. But they were all, like, physical. I, like, even heard a story about this guy Joe, who, like, worked in a button factory.

I just want you to know how happy this comment makes me.

I am trying to think of what the 25 places are for a typical person -- or even for me personally.

After the obvious top 10 or 15 of home, work/school, stores, restaurants, cinemas or theaters, recreation or exercise areas, friends' or family's homes, you get to some places that just aren't that frequently visited. Especially if you're not a user of public transportation (and if they're counting your most-used subway stations or bus stops among your top 25).

I'd guess that the typical number of truly regularly visited places is even smaller than 25. After that, they're not so much regularly visited places so much as just part of a big list of places that you happen to be at more frequently than others. As Thomas Sewell notes in his comment, there's not enough hours in the day to regularly be at a ton of places.

It's the definition of location.

I was on a client's office on May 4th and June 22th. Since I was there twice for more than 10 min and only 7 weeks between the events (<20 weeks) it counts as a location.

Also consider the airport, even if most people don't consider as "public transportation".

Location choice as a multi-armed bandit problem

From the article:

1) Sampling was at least 12 months, max 19 months.
2) 'locations' are defined as places where participants in the study
stopped for more than 10 min. Thus, the locations definition includes metro stations or any other place where you spend more than 10 min.
3) ...visited at least twice and where they spent on average more than 10 min wk–1 during a time-window of 20 consecutive weeks preceding time t.

You can see that it's a pretty low bar for 'locations', spend there 10 min two times over 20 weeks and it counts.

It's tempting to joke about academics discovering the obvious. But this is good insight for data companies. Maybe in a few years we'll read in a McDonalds quarterly report about "out of the 25 most visited places, for 30% of our clients we are in position 5 and above" or something like that.

I don't think it's obvious that people have about 25 places that they regularly visit. It is certainly plausible though.

Or it could be that people just follow the crowd. Or as Rene Girard stated: "People don't go there anymore, it's too crowded".

So, Tyler, are the complacent over-lazy, under-curious or both?

Comments for this post are closed