Designing a Marketing Experiment

INTRODUCTION

This report aims at designing a marketing experiment for a hiking products by Decathlon S.A.
company which is a leading sporting goods retailer. The campaign is a billboard advertising campaign
focusing on the improved quality of the new range of Hiking and Trekking products. The lack of
physical activities and exercise due to daily stress and lack of time is a growing concern for the
present generation as it is heading towards obesity and illness. Hiking and Trekking provides a good
alternative to mundane gym exercises and is proven to be a good adventure. The experiment will be a
before-after design experiment. A full factorial experiment would be difficult in quantifying the
impact when several parameters are changed simultaneously.

EXPERIMENT DESIGN

For this experiment, we consider the independent variable to be increased advertisement in form of
billboard advertising campaign while considering the dependent variable to be to be sales.

The experiment will be run in Bengaluru, Karnataka and the observed control market will be Kolkata,
West Bengal. The test and control markets are similar in terms of population demographics and other
city attributes and are hence chosen for this experiment. Bengaluru and Kolkata are almost similar
sized cities with population of around 12,000,000.

Since the entire cities will be exposed to the experiment, the sample size would be large enough to be
statistically significant.

Given we have a before-after experiment, sales will be recorded for both the control and test group
for a 3 month average from September till November. The experiment will be run for the next 3
months from December till February. The change in sales during the experiment will then be used to
calculate the lift in sales.

The experiment complies to the first three rules of causality but not over the fourth rule that deals
with the presence of external factors. However there appears to be no external factors affecting the
experiment apart from the competitive response which will be accounted for.

ANTICIPATED ISSUES

Both the before and during experiment results will be collected during the Autumn and Winter
months from September till February. If the field implementation is executed anywhere in between
March till August, the summer months might see a dampening in the impact due to the seasonality
involved.

Any other external factors during the implementation such as a new competitor entering the market
or decreasing the price could adversely affect the results and the sales. However, the experiment will
still demonstrate the impact of billboard advertising on sales. The seasonality and external factors will
provide a fair guideline on whether to proceed with the nationwide marketing campaign. The lift in
sales would also provide a definitive idea n the appropriate amount of spending on the campaign.
EXPERIMENT 2.0

Another alternate ti the experiment can be through paid web advertising which provides much better
control on the target audience, test markets and analytic. Thus, it can be a full factorial design with
the independent variables as both price and advertising theme.

This would also yield more informative results as -

 It demonstrates the impact f more than one independent variable.
 The web analytic gives far more insights on the buying preferences of the audience.

This version of experiment is also generally cheaper and quicker to implement that the offline
campaigns such as TV or billboard advertising.

The only concern with the web experiment is that it does not include the chunk of audience that is
not so active on web. In the context of this particular product, that chunk of inactive users could be
huge.