location-data data freestyle: who in NYC gets up early, who parties late, good spots, and more.

location data is awesome…  I have been obsessed with it for a while –  I got my first GPS for my bar-mitzvah in 1996 (it was the only thing I asked for other than night vision), in V1 of the internet I got to hang out with early innovators like John Ellenby / GeoVector, and then guys like Mao/Sense networks in 2005 …  the first real post on this blog was about location and I have kept on posting graphs off my Garmin watch.  When foursquare came out last year the first thing I did was start logging all the checkins I could get my hands on and trying to use the data for stuff like this, and then with Bill Piel and John Steinberg socialgreat started logging millions of them as *i believe* the first released foursquare app.  Everyone knows that conceptually location is a huge deal because it is an enormously relevant and relatively un-captured dataset… 

the question is, after a decade of trial, is there finally usable sample/insight in the noise?  Are we finally getting to the point where ‘location’ is both as ubiquitous and as usable as timestamps? 

last weekend my girlfriend and I were trying to figure out what to do on a sunday afternoon, and out of that I was pushed back to looking at my personal foursquare data-set for insights.  this is some of the stuff I found using the 27K check-ins logged by a few hundred NYC forusquare ‘friends’ from 2/8/2010 to 5/15/2010 (I use CSVemail.com + email – the most basic API – to continuously log everything in an easy to manipulate format) — I was going to use the 5M checkins logged in socialgreat, but I didn’t feel like opening anything more serious than excel, and heck – 27K data-points from early adopting new yorkers seems like a good start to me… 

checkins per day in the set

Checkins per hour in the set of my ‘friends’

Checkins per day per person distribution

Stuff I learned from my cut of 27K checkins at 6.7K locations:
1.  the top 5% of my friends drive 25% of all checkins (10% -> 38%)
2.  the top 1% of locations drive 20% of all checkins (5% -> 43%)

People who get up early
(check in highest % of the time between 5 and 9am)
1.  Andy Weissman
2.  Darren Herman
3.  Blake Robinson
4.  Fred Wilson
5.  Jon Steinberg
6.  Roger Ehrenberg
7.  Jim Moran

People who party late
(check in highest % of the time 10pm – 5am)
1.  Andrew Stillman
2.  James Nord
3.  Grellan Harty
4.  Ken Zamkow
5.  Drew Grant
6.  Tina Hui
7.  Josh Newman

Popular early morning spots
1.  Naples 45
2.  Cafe Bacio
3.  Mendez Boxing
4.  Hospital for Joint Diseases
5.  Carrot Creative
6.  Tony Dapolito Recreation Center
7.  Irving Farm
8.  9th Stree Espresso

Popular late night spots
1.  The Wiskey Ward
2.  Planet Rose
3.  The Commodore
4.  Bleecker Street Pizza
5.  Home Sweet Home
6.  Maracuja
7.  Tappan Zee Bridge
8.  Amsterdam Billards and Bar

Overall if I gut check this, some of it feels right — like the ‘early riser’ list — those are definitely the ‘up and at them folk I would think of in the NYC tech scene that use foursquare… others I am just not hip enough to know about…

but the point (other than that data is fun) — looks like there is some useable data in there…  soon enough, location will be every bit as tied to status/sentiment/etc as time is now, and the more dimensions the better in our information. 

I will be playing with more location data and publishing findings at http://letter.ly/modestproposals in the coming weeks


No Comments

Post a Comment

Your email is never shared. Required fields are marked *