Skip to content

Instantly share code, notes, and snippets.

Show Gist options
  • Star 1 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save ecneladis/9ce229173e67e621f927 to your computer and use it in GitHub Desktop.
Save ecneladis/9ce229173e67e621f927 to your computer and use it in GitHub Desktop.

Iowa Liquor Sales dataset via Socrata/data.iowa.gov

(preliminary exploration)

The state of Iowa has released an 800MB+ dataset of more than 3 million rows showing weekly liquor sales, broken down by liquor category, vendor, and product name, e.g. STRAIGHT BOURBON WHISKIES, Jim Beam Brands, Maker's Mark

This dataset contains the spirits purchase information of Iowa Class “E” liquor licensees by product and date of purchase from January 1, 2014 to current. The dataset can be used to analyze total spirits sales in Iowa of individual products at the store level.

You can view the dataset via Socrata

Here are some steps to get the data wrangled. Do your own visualizations/analysis. But it looks like as good of dataset as any to see such things as:

  • Purchase trends during holidays and college football season
  • Most popular brands and types of alcohol
  • Price variance between same-city stores and different-city stores
  • Popularity of Hawkeye Vodka in Iowa City versus Ames

The metadata

The URL for the metadata via Socrata's API, is:

https://data.iowa.gov/metadata/v1/dataset/m3tr-qhgy.json

Or you can see a cached version here. The metadata contains column names and datatypes.

Download it

# bash
curl https://data.iowa.gov/api/views/m3tr-qhgy/rows.csv?accessType=DOWNLOAD \
     -o iowa-liquor.csv

Translate the dates, clean up numbers, pre-import

# via bash
sed -E "s#([0-9]{2})/([0-9]{2})/([0-9]{4})#\3-\1-\2#" < iowa-liquor.csv | 
  tr -d '$' > iowa-liquor-datefixed.csv

SQL

Basic MySQL schema to include all the fields; however, you can probably drop the redundant STORE LOCATION field, at the very least.

# mysql
CREATE TABLE `iowaalcohol` (
  `DATE` date DEFAULT NULL,
  `CONVENIENCE STORE` varchar(255) DEFAULT NULL,
  `STORE` varchar(12) DEFAULT NULL,
  `NAME` varchar(255) DEFAULT NULL,
  `ADDRESS` varchar(255) DEFAULT NULL,
  `CITY` varchar(255) DEFAULT NULL,
  `ZIPCODE` varchar(20) DEFAULT NULL,
  `STORE LOCATION` varchar(255) DEFAULT NULL,
  `COUNTY NUMBER` varchar(4) DEFAULT NULL,
  `COUNTY` varchar(255) DEFAULT NULL,
  `CATEGORY` varchar(20) DEFAULT NULL,
  `CATEGORY NAME` varchar(100) DEFAULT NULL,
  `VENDOR NO` varchar(20) DEFAULT NULL,
  `VENDOR` varchar(255) DEFAULT NULL,
  `ITEM` varchar(20) DEFAULT NULL,
  `DESCRIPTION` varchar(255) DEFAULT NULL,
  `PACK` int(11) DEFAULT NULL,
  `LITER SIZE` int(11) DEFAULT NULL,
  `STATE BTL COST` float(7,2) DEFAULT NULL,
  `BTL PRICE` float(7,2) DEFAULT NULL,
  `BOTTLE QTY` int(11) DEFAULT NULL,
  `TOTAL` float(12,2) DEFAULT NULL,
  KEY `STORE` (`STORE`),
  KEY `DATE` (`DATE`),
  KEY `CATEGORY NAME` (`CATEGORY NAME`),
  KEY `CATEGORY` (`CATEGORY`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

Dirty store names data

The store names aren't normalized; however, STORE seems like it should be a reliable enough foreign key for this other dataset on Socrata: Iowa Liquor Stores.

SELECT `STORE`, `NAME`, `ADDRESS`, `CITY`, `ZIPCODE`, `COUNTY NUMBER`, `COUNTY` 
FROM iowaalcohol
WHERE `STORE` = '2508'
GROUP BY `STORE`, `NAME`
Result
STORE NAME ADDRESS CITY ZIPCODE COUNTY NUMBER COUNTY
2508 Hy-Vee Food Store #1 / Cedar Rapids 1843 JOHNSON AVENUE, N.W. CEDAR RAPIDS 52405 57 Linn
2508 Hy-vee Food Store #1/ceda 1843 JOHNSON AVENUE N.W. CEDAR RAPIDS 52405 57 Linn
2508 Hy-vee Food Store #1/Cedar Rapids 1843 JOHNSON AVENUE N.W. CEDAR RAPIDS 52405 57 Linn

Top spirits by total sales

CATEGORY NAME is seemingly cleaner...here's how to get a quick summation of liquor categories, ordered by total sales, e.g. SUM(TOTAL):

SELECT `CATEGORY NAME`,
ROUND(SUM(`TOTAL` / 1000)) as `total_sales`,
SUM(`BOTTLE QTY`) AS `total_bottles`,
ROUND(SUM(`LITER SIZE` * `BOTTLE QTY` / 1000) / 1000, 2) AS `total_liters`,
TRUNCATE(AVG(`TOTAL` / (`LITER SIZE` * `BOTTLE QTY` / 1000)), 2) AS `avg_cost_per_liter`
FROM iowaalcohol
GROUP BY `CATEGORY NAME`
ORDER BY `total_sales` DESC

Note: I don't really know if I'm interpreting the BOTTLE QTY and the seemingly irrelevant PACK columns correctly.

Top selling liquors in Iowa since January 2014

Note: total_sales and total_liters are in the thousands.

CATEGORY NAME total_sales total_bottles total_liters avg_cost_per_liter
CANADIAN WHISKIES 48053 3577933 3840.07 15.81
80 PROOF VODKA 48046 5960351 5889.17 9.46
SPICED RUM 31601 2082680 2054.59 15.76
IMPORTED VODKA 23880 1166160 1138.00 25.07
TEQUILA 21411 1274034 1049.31 29.32
STRAIGHT BOURBON WHISKIES 20924 1243488 1180.98 20.55
WHISKEY LIQUEUR 19339 1282480 1145.82 17.57
TENNESSEE WHISKIES 17648 804769 648.92 28.09
PUERTO RICO & VIRGIN ISLANDS RUM 12729 1144599 1229.44 11.89
BLENDED WHISKIES 12037 1310974 1262.54 10.82
FLAVORED VODKA 11539 1124827 870.39 13.92
MISC. IMPORTED CORDIALS & LIQUEURS 11417 562464 437.35 28.72
CREAM LIQUEURS 9342 506558 422.24 22.25
IMPORTED VODKA - MISC 9077 548380 402.24 23.72
FLAVORED RUM 8030 610725 532.19 15.16
IMPORTED GRAPE BRANDIES 7742 465402 196.74 42.84
SCOTCH WHISKIES 7309 343235 387.97 26.27
IMPORTED SCHNAPPS 7076 410570 379.27 21.35
AMERICAN COCKTAILS 6314 602536 914.35 7.43
IRISH WHISKIES 5944 246198 209.89 31.13
IMPORTED DRY GINS 5391 237069 228.14 24.63
AMERICAN DRY GINS 5268 741783 580.66 10.12
AMERICAN GRAPE BRANDIES 5137 854924 420.41 13.01
DECANTERS & SPECIALTY PACKAGES 4449 234289 213.42 27.98
SINGLE MALT SCOTCH 4149 99707 76.89 57.69
MISC. AMERICAN CORDIALS & LIQUEURS 3759 297507 209.40 17.71
STRAIGHT RYE WHISKIES 3755 142562 106.65 35.08
COFFEE LIQUEURS 2614 157633 131.57 19.54
DISTILLED SPIRITS SPECIALTY 2601 256087 235.09 20.95
PEACH SCHNAPPS 1755 174310 159.22 10.95
PEPPERMINT SCHNAPPS 1715 259103 235.03 8.34
BLACKBERRY BRANDIES 1254 141897 123.74 10.47
TRIPLE SEC 986 253861 248.09 4.34
AMERICAN AMARETTO 885 138302 128.33 7.40
AMERICAN ALCOHOL 870 65116 49.09 17.78
APPLE SCHNAPPS 805 76803 68.15 12.21
BUTTERSCOTCH SCHNAPPS 638 66228 56.66 11.13
CINNAMON SCHNAPPS 618 55962 46.66 15.31
IMPORTED AMARETTO 591 28728 21.32 27.79
WATERMELON SCHNAPPS 502 45811 43.15 11.88
MISCELLANEOUS SCHNAPPS 502 44459 37.38 14.66
APRICOT BRANDIES 501 56671 48.33 10.70
GRAPE SCHNAPPS 444 40053 38.42 11.83
BARBADOS RUM 396 27347 20.51 20.11
JAMAICA RUM 371 22730 18.78 20.18
SINGLE BARREL BOURBON WHISKIES 356 12079 9.01 44.42
100 PROOF VODKA 320 23346 18.87 17.13
ROOT BEER SCHNAPPS 268 29322 28.15 10.13
PEACH BRANDIES 215 31166 18.29 11.70
FLAVORED GIN 207 20939 15.10 14.02
CHERRY BRANDIES 201 25809 18.09 11.35
RASPBERRY SCHNAPPS 175 20777 17.93 10.26
STRAWBERRY SCHNAPPS 166 22031 16.52 10.34
TROPICAL FRUIT SCHNAPPS 124 17031 15.43 8.13
MISCELLANEOUS BRANDIES 115 5344 3.60 38.19
GREEN CREME DE MENTHE 95 13751 10.32 9.25
90 2903 2.18 40.02
WHITE CREME DE CACAO 77 10958 8.22 9.30
LOW PROOF VODKA 76 6376 8.98 12.66
DARK CREME DE CACAO 73 10383 7.78 9.30
AMERICAN SLOE GINS 70 10028 8.48 8.24
OTHER PROOF VODKA 57 4719 3.54 15.91
BOTTLED IN BOND BOURBON 52 3603 2.93 19.01
ROCK & RYE 48 4566 3.42 14.54
SPEARMINT SCHNAPPS 41 5645 5.65 7.24
WHITE CREME DE MENTHE 24 3440 2.58 9.35
CREME DE ALMOND 14 2007 1.52 9.30
ANISETTE 12 1697 1.27 9.26
HIGH PROOF BEER 4 38 0.03 145.56
{
"name": "Iowa Liquor Sales",
"updatedAt": "2015-04-01T15:40:53.000Z",
"description": "This dataset contains the spirits purchase information of Iowa Class “E” liquor licensees by product and date of purchase from January 1, 2014 to current. The dataset can be used to analyze total spirits sales in Iowa of individual products at the store level.",
"domain": "data.iowa.gov",
"id": "m3tr-qhgy",
"columns": {
"btl_price": {
"fred": "money",
"name": "BTL PRICE",
"description": "The amount the store paid for the bottle",
"physicalDatatype": "money",
"position": 20,
"hideInTable": false
},
"city": {
"fred": "text",
"name": "CITY",
"description": "City of store",
"physicalDatatype": "text",
"position": 6,
"hideInTable": false
},
"zipcode": {
"fred": "text",
"name": "ZIPCODE",
"description": "Zipcode of store",
"physicalDatatype": "text",
"position": 7,
"hideInTable": false
},
"name": {
"fred": "text",
"name": "NAME",
"description": "Name of store",
"physicalDatatype": "text",
"position": 4,
"hideInTable": false
},
"state_btl_cost": {
"fred": "money",
"name": "STATE BTL COST",
"description": "The amount that Alcoholic Beverages Division paid for the bottle",
"physicalDatatype": "money",
"position": 19,
"hideInTable": false
},
"store_location": {
"fred": "location",
"name": "STORE LOCATION",
"description": "Location of store",
"physicalDatatype": "location",
"position": 8,
"hideInTable": false
},
"convenience_store": {
"fred": "text",
"name": "CONVENIENCE STORE",
"description": "Contains a \"Y\" if convenience store",
"physicalDatatype": "text",
"position": 2,
"hideInTable": false
},
"description": {
"fred": "text",
"name": "DESCRIPTION",
"description": "Item Description",
"physicalDatatype": "text",
"position": 16,
"hideInTable": false
},
"category_name": {
"fred": "text",
"name": "CATEGORY NAME",
"description": "Category for the liquor type",
"physicalDatatype": "text",
"position": 12,
"hideInTable": false
},
"pack": {
"fred": "number",
"name": "PACK",
"description": "The number of bottles in a case",
"physicalDatatype": "number",
"position": 17,
"hideInTable": false
},
"county_number": {
"fred": "text",
"name": "COUNTY NUMBER",
"description": "Iowa county number where store is located\n",
"physicalDatatype": "text",
"position": 9,
"hideInTable": false
},
"store": {
"fred": "text",
"name": "STORE",
"description": "Store Number",
"physicalDatatype": "text",
"position": 3,
"hideInTable": false
},
"total": {
"fred": "money",
"name": "TOTAL",
"description": "The total amount of the sale (number of bottles multiplied by the bottle price)",
"physicalDatatype": "money",
"position": 22,
"hideInTable": false
},
"county": {
"fred": "text",
"name": "COUNTY",
"description": "County where store is located\n",
"physicalDatatype": "text",
"position": 10,
"hideInTable": false
},
"vendor_no": {
"fred": "text",
"name": "VENDOR NO",
"description": "The vendor number of the company for this brand of liquor",
"physicalDatatype": "text",
"position": 13,
"hideInTable": false
},
"date": {
"fred": "calendar_date",
"name": "DATE",
"description": "Date of order",
"physicalDatatype": "calendar_date",
"position": 1,
"hideInTable": false
},
"address": {
"fred": "text",
"name": "ADDRESS",
"description": "Address of store",
"physicalDatatype": "text",
"position": 5,
"hideInTable": false
},
"category": {
"fred": "text",
"name": "CATEGORY",
"description": "Category code for liquor type",
"physicalDatatype": "text",
"position": 11,
"hideInTable": false
},
"vendor": {
"fred": "text",
"name": "VENDOR",
"description": "The vendor name of the company for this brand of liquor",
"physicalDatatype": "text",
"position": 14,
"hideInTable": false
},
"liter_size": {
"fred": "number",
"name": "LITER SIZE",
"description": "The metric size of a bottle",
"physicalDatatype": "number",
"position": 18,
"hideInTable": false
},
"bottle_qty": {
"fred": "number",
"name": "BOTTLE QTY",
"description": "The number of bottles ordered by the store",
"physicalDatatype": "number",
"position": 21,
"hideInTable": false
},
"item": {
"fred": "text",
"name": "ITEM",
"description": "Item Number",
"physicalDatatype": "text",
"position": 15,
"hideInTable": false
}
},
"ownerId": "phiz-mwit",
"permissions": {
"isPublic": true,
"rights": [
"read"
]
}
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment