CNBC.com published the following data from NPD Group (July 2011) in an effort to reveal “America’s Most Caffeinated Cities,” with ranking based on each city/metropolitan region’s total # of coffee shops: 12. San Francisco/Oakland/San Jose – 1,379 coffee shops 11. Honolulu – 229 coffee shops 10. Austin, Texas – 199 coffee shops 9. San Diego – 455 coffee shops 8. Sacramento/Stockton/Modesto, Calif. – 556 coffee shops 7. Eugene, Ore. – 138 coffee shops 6. Denver – 743 coffee shops 5. Spokane, Wash. – 251 coffee shops 4. Boise, Idaho – 143 coffee shops 3. Portland, Ore. – 876 coffee shops 2. Anchorage, Alaska – 172 coffee shops 1. Seattle/Tacoma – 1,640 coffee shops Problems with the CNBC report: First, a city’s total number of coffee shops alone isn’t meaningful without the context of knowing how many people are in that city. Second, the data above is hard to interpret. It’s not visual or interesting. Third, something in the data from CNBC’s story feels fishy. It’s hard to believe Eugene, OR (pop 156,185) has more coffee shops (138) than unmentioned Chicago (pop 2.6 million) and New York City (8.1 million), as well as lots of others. We didn’t attempt to solve or fact check this, but it seems noteworthy. SOLUTIONS: First, figure out a metric that is meaningful and easy to understand. In this case, we chose the # of coffee shops per 10,000 people in each city. U.S. Census Bureau provided the population figures. Thanks to the abacus in our office, the math was simple. Second, how to make it visual and easy to understand? Since we’re comparing cites, mapping was the obvious choice. Using an image of a coffee shop to represent each one per 10,000 people enables viewers to quickly glance at the graphic and easily see which cities are the most caffeinated.