Beer2
Download as ppt, pdf1 like342 views
This document discusses using data from Beer Advocate to help breweries determine what types of beers to brew. It provides statistics on the craft beer industry showing strong growth. The document examines sample user comments on beers and outlines next steps such as building dashboards to help answer questions on what beers to brew each month and which flavor profiles might sell well.
Beer2
- 1. Hi, how are you?
- 4. The Augmented Reality Sandbox
- 8. Present on anything. Make it look nice.
- 10. Some of the things I considered talking about.
- 11. A talk that went poorly
- 12. Not Tom Selleck
- 13. Beer
- 14. (I realize this isnt a topic you often talk about with a prospective employer) But I love beer.
- 15. Beer, and beer brewing, is complex.
- 16. and its Big Money Beer Sold in 2013: 196,241,321 barrels (or over six billion gallons) of that 15,302,838 barrels was craft beer (or nearly 500 million gallons) all data according to Brewers Association at http://brewersassociation.org
- 17. $100 Billion Made on Beer Last Year of that $14.3 Billion Dollars on Craft Beer all data according to Brewers Association at http://brewersassociation.org
- 18. There was a 17.2% increase in craft beer sales from 2012-2013. Hundreds of new breweries are opening each year. Big breweries trying to catch up with own craft brands. Craft is a relatively new market. What to make? What to sell? How do we figure it all out?
- 20. (though, in this case, the dataset isnt *that* big)
- 21. 2000-2012 1.5 million entries 400,000+ users
- 23. With craft beer on the rise, so many choices. How to make them?
- 24. Notes on this brew: Brewed with dirty gym socks, and pigfeed corn. May be paired with industrial cleaner to strengthen its effect" MaltLickyWithTheCandy
- 25. Look, I don't know what everyone is bashing...this shit costs like $10 for a 30 pack what do you expect? Golden ambrosia served in a chalice while angels whistle on your dong? aheedratron
- 27. Thesis: We can infer what beers we should brew based on Beer Advocate data.
- 28. A Quick Note: IPAs are terrible.
- 29. Some data analysis from the beer advocate stats.
- 32. When are we drinking?
- 35. Does the data work?
- 36. September
- 37. October
- 38. So, were pretty sure this works. Whats next?
- 43. Next Steps
- 44. Ted Dunning Anomalous data Look for the norm Aggregate ratings with current data, and see if we can find beers that stand out as having potential.
- 45. Next Steps Build additional Dashboards that help answer questions: What beers should I brew this month? What flavor profiles should I be matching for this time period? What beers have the potential to sell well, but arent being pushed by competitors?
- 47. Thanks.