Why do you need a control group?

Many a times, marketers face this dilemma of what all should be communicated to customer and what all is futile. They would specifically want to know, "Is the communication that I am sending to my customers adding any value to them or my business? Will I be able to fulfill my objectives without sending that communication piece? " Here is where marketing research experts suggest of using Control & Experiment group. 

An experimental group, is a subset of customers we are targeting in an experiment who receive the communication being tested. 

A control group, is a subset of the customers we are targeting with a particular campaign, who will not receive the campaign communication.

While no communication is being sent to control group, MoEngage tracks the conversions for the users falling under this group. Control group serves as a baseline for experiment group to compare performance of communication. When one compares the variation with control, one can check the impact of campaign i.e. whether conversions were better or worse off if specific campaign was not sent to some customers.

The snapshot below offers a comparison between the control group and message where we can see that by sending the notification, the marketer was able to get an uplift of 126% for the primary conversion goal.

So whenever you feel the need of communicating any thing to customers, use a control group to establish the (f)utility of it.

Having known all about the Multivariate experiments, now is the time to know how to create an experiment using MoEngage and how to include control group in your experiment.

Have more questions? Submit a request

Comments