Question

Business Statistics

Data Analysis for Managers making decisions

You manage a line of consumer products and want to change your marketing campaign. You decide that you want to test a new campaign against the existing one to see if the new one is any better. How would you design the test? What sort of data would you collect? Would you be using a dependent or independent variable? Would you use a simple regression analysis or a multiple regression analysis? Justify your answer

Answer 1

The problem discussed above is the problem related to Market Research domain.

Problem:- We have to make a decision on new marketing compaign against the existing one and analyze, which one is better.

Note:- If we are applying the new marketing compaign then we must measure its impact on business in monetary terms(obviously).Profit/Gross business will be the study variable in this case.

Two types of data can be collected for this kind of study:-

1. Collect two samples of size (say)10 observations on daily sales(in Rs.) with the old compaign and also for new compaign.

So, in this case one can design a test of hypothesis based on T-Distribution.

Under this we have two choices a) Two - Sample t test for comparison of means; b) Paired t-test independent observation

a) Here,we will the 10 observations to test whether the mean of both sample is same or not assuming same variance.

Conclusion in business language:- If the test rejecting the hypothesis that both samples have same mean it implies that old marketing compaign and new marketing compaign has different impact on gross daily sales(in Rs.) i.e both are not same at least.

Hence,we proceed to the case b)

b) This test is based upon the difference(d) between values of variables x and y

where,

x: Daily sales before the new marketing Compaign

y: Daily sales after the new marketing Compaign

d=x-y

Here,the original scenario will come into picture that whether the new marketing compaign is actually significant or not. After applying this test if we get some statistically significant values then we can strongly say that New marketing compaign will be more helpfull in increasing the daily sale.

2. Obviously, this kind of business problem can also be studied using the linear regression technique.

For using this statistical technique we need data in the form of dependent variable(Gross business/Profit) and independent variable/s.

Multiple Linear Regression will the best regression we should use in such scenarion

Now question is to determine the independent variables,those variables which are affecting the Gross businees /Profit,there may be several variables which are contributing to make new marketing campaign successful for business.So, one has to identify them and build a model based on those variables to estimate the profit in business.

i.e we can predict business / profit using the new compaign's results and old compaign's and check whether which one is giving good estimated profit.

In this way new compaign's impact on business can be studied using linear regression technique and decision can be taken accordingly.