I am currently working in a new visualization for the Music Viz Contest, to try to win the Big Price: a trip to Las Vegas to compete in the Iron Viz competition at Tableau Conference 2015. I know the probabilities of winning are not very high… but after my first Viz of the Day of last week I feel full of confidence!
I don’t want to tell the idea for my Viz before finish it, but I found a difficulty that almost maked me change completely my initial idea, and I would like to share how I solve it, maybe somebody else has had this same problem.
I am working with 2,000 locations in a big number of countries from Latin America, North America, Europe, Africa, Asia and Oceania, and for each location I have a venue name, city name and country name. But when I tried to map each location in Tableau I get a lot of ambiguous and unknown results and couldn’t map the 10% of them. It seems that with a very big dispersion of locations, if you don’t have additional info like state, region or similar, Tableau has difficulties to map all the points even adding the country and city dimensions to the detail shelf.
My first idea was to look for all the regions / states of all the database hoping that with that additional information will be enough (City, State and Country), so I found this information for the 2,000 locations and although the number of ambiguous and unknowns decreased to 80, it was still bigger than what I needed. The only final solution that I could think of was to add the latitude and longitude of at least those 80 ambiguous or unknown locations, and because of my little knowledge of Google’s API for geocoding, I had many doubts that I have time to finish my viz in time if I had to search for each location in Google Maps, copy the latitude and longitude and paste them in my database.
Instead, after some time searching through the web I finally found a very useful page that I’m sure it will be very useful for a lot of Tableau users: EasyMapMaker. With this web you just need an Excel file with your addresses, the name of the city and the country and it will add the latitude and longitude information. for example, let’s try to geolocalize these locations (4 football stadiums, the address of my previous flat in Madrid and the Data School address, with no region or postcode). For general venues with no address is better to also include in the venue name the city and the country for gaining more accuracy, but as you can see, in the stadiums there’s no address, just the name of the stadium and in the addresses no region, state or postal code info:
Venue | City | Country |
Estadio Santiago Bernabeu Madrid Spain | Madrid | Spain |
Nou Camp Barcelona Spain | Barcelona | Spain |
Emirates Stadium London United Kingdom | London | United Kingdom |
Anfield Stadium Liverpool United Kingdom | Liverpool | United Kingdom |
Calle Mancebos, 6 Madrid Spain | Madrid | Spain |
33 Cannon Street London United Kingdom | London | United Kingdom |
Venue | City | Country | emm_lat | emm_lng |
Estadio Santiago Bernabeu Madrid Spain | Madrid | Spain | 40.453054 | -3.688344 |
Nou Camp Barcelona Spain | Barcelona | Spain | 41.380896 | 2.12282 |
Emirates Stadium London United Kingdom | London | United Kingdom | 51.557834 | -0.098675 |
Anfield Stadium Liverpool United Kingdom | Liverpool | United Kingdom | 53.430829 | -2.96083 |
Calle Mancebos, 6 Madrid Spain | Madrid | Spain | 40.411986 | -3.712138 |
33 Cannon Street London United Kingdom | London | United Kingdom | 51.512611 | -0.09484 |