Marketing Experiment Design for Hilton Hotels
Marketing Experiment Design for Hilton Hotels Introduction This report will design a marketing experiment for Hilton Hotels. The objective of the campaign will be to acquire new members into the hotel loyalty program. The campaign will be targeted towards customers who stay at Hilton Hotels but are not members of the Hilton HHonors loyalty program. The campaign will highlight the redemption benefits associated with membership and offer a 20,000 point sign up bonus. The campaign will be conducted via direct mail. A basic experiment will be used. The reason a basic experiment will be used is that it is a simple way to examine whether offering a loyalty point acquisition bonus will incent customers to join the hotel loyalty program.
Experiment Design For this experiment, the independent variable is increased marketing in the form of a direct mail piece sent to a customer. The dependent variable being observed is customer acquisitions in the hotel loyalty program. The experiment will target a random sample of 50,000 customers who live in the California market and have stayed at a Hilton hotel three times in the past 12 months. The California market was chosen due to the fact that Hilton Hotels headquarters was located in Beverly Hills, California from 1969-2009. As such, there is likely an already strong level of brand awareness for Hilton Hotels among these customers. Given the number of randomly chosen customers included in the campaign and the fact that they come from the same state and have stayed at a Hilton hotel the same number of times, the sample size should be statistically significant. A control group of 25,000 customers will receive a mailed brochure communicating the redemption benefits associated with membership and NO sign up bonus. The communication will include a custom URL to track customer sign ups during the experiment. A test group of 25,000 customers will receive an email with the same information regarding redemption benefits AND a 20,000 point signup bonus. A separate and distinct customer URL will be included with this brochure to track bonus point sign ups. The campaign will run for two months in January and February, typically low season for hotels. For this basic experiment, customer acquisitions (sign ups) will be tracked for the test and control groups during the two month promotional period. The number of customers joining the hotel loyalty program during the promotional period will be used to calculate the lift in membership. The experiment satisfies the rules of causality as follows:
Increased marketing is expected to drive increased membership in the hotel loyalty program Without additional marketing, membership grows at the same pace as it did before the promotion (i.e. the rate of growth remains unchanged) Time Sequence – increased marketing activity today drives increased customer signups during the promotional period and, we assume, beyond.
No other external factor – The experiment cannot control external factors; however, it does not appear that there would be any which would have an impact. Because this experiment is focused on growing membership in the hotel’s loyalty program and Hilton is the only entity that controls the program, competitive actions should not have an adverse impact on this experiment.
Anticipated Issues A potential issue associated with this experiment is customer dissatisfaction for those not receiving the direct mail brochure. In today’s digital world, offers are regularly communicated and shared amongst friends (and strangers in various forums on the internet). Customers outside of the target market may contact Hilton and request to be included in the experiment. Customers not in the loyalty program who were not included in the experiment may be resistant to joining until they receive a sign up bonus offer. Loyal customers who find out about the experiment may be unhappy about the generous point offer and want/request an additional incentive to remain loyal. A third issue to consider is the future cost/financial liability associated with expanding this experiment to a larger base of global customers. Despite these potential issues, the experiment will demonstrate the impact of marketing on customer sign ups/acquisitions in the loyalty program. The experiment will also demonstrate what type of membership lift is generated when a sign up bonus is offered. This will provide future guidance on whether or not to offer a sign up bonus.
Experiment Adaptation Another version of the experiment could be done via a targeted email campaign using a full factorial design. The full factorial design would allow the experiment to have multiple independent variables: testing various sign up bonus point offers and highlighting various loyalty program features beyond redemption (in-room amenities, upgrades, late checkouts, etc.). Using email marketing would allow the experiment to be broadly expanded at a significantly lower cost and provide better tracking and analytics. This experiment also would enable us to later evaluate whether these customers who joined the loyalty program increased their stay frequency and/or spend activity with the hotel. This version of the experiment would be more effective and offer better results because it can be evaluated in real time as customers respond (or don’t respond) to the various offers and featured benefits. It also is less costly to implement and can be easily adjusted as necessary. The one thing this experiment would not capture is customers who have not provided an email address to Hilton. These customers would have to be reached in another way.
Experiment Design For this experiment, the independent variable is increased marketing in the form of a direct mail piece sent to a customer. The dependent variable being observed is customer acquisitions in the hotel loyalty program. The experiment will target a random sample of 50,000 customers who live in the California market and have stayed at a Hilton hotel three times in the past 12 months. The California market was chosen due to the fact that Hilton Hotels headquarters was located in Beverly Hills, California from 1969-2009. As such, there is likely an already strong level of brand awareness for Hilton Hotels among these customers. Given the number of randomly chosen customers included in the campaign and the fact that they come from the same state and have stayed at a Hilton hotel the same number of times, the sample size should be statistically significant. A control group of 25,000 customers will receive a mailed brochure communicating the redemption benefits associated with membership and NO sign up bonus. The communication will include a custom URL to track customer sign ups during the experiment. A test group of 25,000 customers will receive an email with the same information regarding redemption benefits AND a 20,000 point signup bonus. A separate and distinct customer URL will be included with this brochure to track bonus point sign ups. The campaign will run for two months in January and February, typically low season for hotels. For this basic experiment, customer acquisitions (sign ups) will be tracked for the test and control groups during the two month promotional period. The number of customers joining the hotel loyalty program during the promotional period will be used to calculate the lift in membership. The experiment satisfies the rules of causality as follows:
Increased marketing is expected to drive increased membership in the hotel loyalty program Without additional marketing, membership grows at the same pace as it did before the promotion (i.e. the rate of growth remains unchanged) Time Sequence – increased marketing activity today drives increased customer signups during the promotional period and, we assume, beyond.
No other external factor – The experiment cannot control external factors; however, it does not appear that there would be any which would have an impact. Because this experiment is focused on growing membership in the hotel’s loyalty program and Hilton is the only entity that controls the program, competitive actions should not have an adverse impact on this experiment.
Anticipated Issues A potential issue associated with this experiment is customer dissatisfaction for those not receiving the direct mail brochure. In today’s digital world, offers are regularly communicated and shared amongst friends (and strangers in various forums on the internet). Customers outside of the target market may contact Hilton and request to be included in the experiment. Customers not in the loyalty program who were not included in the experiment may be resistant to joining until they receive a sign up bonus offer. Loyal customers who find out about the experiment may be unhappy about the generous point offer and want/request an additional incentive to remain loyal. A third issue to consider is the future cost/financial liability associated with expanding this experiment to a larger base of global customers. Despite these potential issues, the experiment will demonstrate the impact of marketing on customer sign ups/acquisitions in the loyalty program. The experiment will also demonstrate what type of membership lift is generated when a sign up bonus is offered. This will provide future guidance on whether or not to offer a sign up bonus.
Experiment Adaptation Another version of the experiment could be done via a targeted email campaign using a full factorial design. The full factorial design would allow the experiment to have multiple independent variables: testing various sign up bonus point offers and highlighting various loyalty program features beyond redemption (in-room amenities, upgrades, late checkouts, etc.). Using email marketing would allow the experiment to be broadly expanded at a significantly lower cost and provide better tracking and analytics. This experiment also would enable us to later evaluate whether these customers who joined the loyalty program increased their stay frequency and/or spend activity with the hotel. This version of the experiment would be more effective and offer better results because it can be evaluated in real time as customers respond (or don’t respond) to the various offers and featured benefits. It also is less costly to implement and can be easily adjusted as necessary. The one thing this experiment would not capture is customers who have not provided an email address to Hilton. These customers would have to be reached in another way.
