Case Study: Promotion Analytics: Restaurant Industry

the marketing manager for the hottest new restaurant in town, Acme. Acme's top entrees include:

Dish A: Braised Alaskan wild salmon with rice pilaf and grilled asparagus

Dish B: Grass-fed organic beef tenderloin with baked potato and sauted mushrooms in Cabernet sauce

Dish C: Vegetable melee, with red and green peppers, organic beets and artichoke hearts, over a bed of spinach, served with a spicy balsamic vinaigrette

You use two promotion tools to drive business to Acme. The first tool is Facebook (F) to target local diners. The second tool is Groupon (G) to increase trial rate of new customers.

After analyzing historic sales and promotion data, you assemble the data shown in the table below, including Audience/ Ad (the number of people who viewed each ad, as well as Cost/ Ad (the monetary cost to execute each campaign ad). F stands for Facebook campaigns and G stands for Groupon campaigns.

Table 1: Promotion effectiveness metrics, Acme Restaurants

Promotion Vehicle Audience/Ad Cost/Ad

F:Facebook 60 $40

G: Groupon 70 $50

You face several constraints, as summarized in the second table. Due to limited capacity (ability to produce) of your advertising agency, you cannot create more than a certain number of new advertising campaigns per month. You also cannot exceed a certain marketing spend (budget) per month.

Table 2: Budget Constraints

Promotion Vehicle Maximum Ads/Month Maximum Budget/Mo.

F:Facebook 30

G: Groupon 30

Total Promotion Spend $2000

Problem 1

Which of the following choices is the correct form for the objective equation to maximize audience exposure, where F stands for the number of Facebook campaigns and G stands for the number of Groupon campaigns?

Z = 60 * F + 70 * G

Z = 70 * F + 60 * G

Z = 30 * F + 30 * G

Z = 40 * F + 50 * G

Problem 2

Which of the following choices is the correct form for the constraint equation for budget, where F stands for the number of Facebook campaigns and G stands for the number of Groupon campaigns? In the equations, the "<=" symbol is pronounced "less than or equal to" and ">=" is "more than or equal to".

40 * F + 50 * G <= 2000

40 * F + 50 * G >= 2000

60 * F + 70 * G <= 2000

60 * F + 70 * G >= 2000

Problem 3

Which of the following choices is the correct one for values of F (number of Facebook campaigns) and G (number of Groupon campaigns)? To obtain the values for F and G, execute the linear optimization model described in the course.

F = 60; G = 70

F = 16; G = 30

F = 30; G = 30

F = 30; G = 16

Problem 4

Which of the following choices is the correct one for values of total budget to execute the F and G campaigns, where F stands for the number of Facebook campaigns and G stands for the number of Groupon campaigns? To obtain the value for total budget, execute the linear optimization model described in the course.

40

50

40 * 30 + 50 * 16

40 * 30 + 50 * 30

Problem 5

Based on the results of the linear optimization model with the data given in the case, what inference can we make?

Acme is using both F and G resources effectively

Acme is using F resources effectively

Acme is using G resources effectively

Acme is using neither F nor G resources effectively

R Application Case Study: Soccer (Football)

Soccer Data

Name Team Goals Assists Yellows PS

Lionel MessiArgentina 26 16381.9

Neymar Brazil2412680.9

ZLatan Ibrahimovic Sweden 38 13178.6

Gareth Bale Wales1910279.9

Imagine you are a big soccer/football fan, and you enjoy looking at statistics of players and predicting which ones will do well. You have collected data of several top players, including their name, the number of goals (where they shot the ball into the goal), assists (where they helped another player shoot a ball into the goal), yellows (where they received a warning in the form of a yellow card), and PS (which calculates the percentage of saves they accomplished). The table below shows the data, obtained from whoscored.com. The name of the table is "Soccer Data".

Question 1

Please refer to the Soccer Data table above. You want to calculate the average number of goals scored by the soccer players in the table. Place the goals data into a vector called goals. Calculate the average. State the R statements you would use and the expected results.

R Statement: goals <- vector(26, 24, 38, 19); mean (goals); Result: 26.75

R Statement: goals.csv <- vector(26, 24, 38, 19); mean (goals.csv); Result: 26.75

R Statement: goals.csv <- c(26, 24, 38, 19); mean (goals.csv); Result: 26.75

R Statement: goals <- c(26, 24, 38, 19); mean (goals); Result: 26.75

Question 2

Please refer to the Soccer Data table above. In order to do additional calculations, you plan to enter the SoccerData dataset into R. You will call the R dataset "Soccer Data". Assume that the original dataset is in a working directory dedicated to R and that you will be including the header row. State the statements you would use.

SoccerData <- read.csv ("SoccerData.csv", header=T)

SoccerData.csv <- read.csv ("SoccerData.csv", header=T)

SoccerData <- read.csv ("SoccerData", header=T)

SoccerData <- read.csv ("SoccerData.csv", header=F)

Question 3

Please refer to the Soccer Data table above. You wish to study the relationship between Goals (the dependent variable) and the independent variables of Assists, Yellows, and PS. Execute a regression analysis and place the results in a file called "Soccer Regression". State the statements you would use.

SoccerRegression <- regression (Goals ~ Assists + Yellows + PS, SoccerData)

SoccerRegression.csv <- lm (Goals ~ Assists + Yellows + PS, SoccerData)

SoccerRegression <- regression (Goals ~ Assists + Yellows + PS, SoccerData)

SoccerRegression <- lm (Goals ~ Assists + Yellows + PS, SoccerData)

Question 4

Please refer to the Soccer Data table above. After executing the regression analysis, you study the output to determine the validity of individual variables, such as Assists, toward the model. Assume each variable has a corresponding t statistic. Which of the values below best represents the probability of encountering an equal t value in random data? The answers below apply standard statistics terminology; different functions within R could use non-standard terminology or even different letters; what we are asking for here is the standard statistics terminology.

P-Value

F-Value

S-Value

Standard Error