Marketing experimentation
Marketing experimentation is a research method which can be defined as "the act of conducting such an investigation or test". It is testing a market that is segmented to discover new opportunities for organisations. By controlling conditions in an experiment, organisations will record and make decisions based on consumer behaviour. Marketing experimentation is commonly used to find the best method for maximizing revenues through the acquisition of new customers. For example; two groups of customers are exposed to different advertising. How did consumers react to advertising compared to the other group?. Did the advertising increase sales for each group?.
Characteristics
There are three characteristics which are the make-up of a market experimentation:Experimental subjects - Humans are usually participants in experiments. Subjects are divided into two or more groups and can be referred to as focus groups. Subjects can be made up of a particular age group (demographical), from a particular area (geographical), or culture.Conditions - Known as the independent variable where, conditions are tightly controlled and manipulated by the tester. In a marketing experiment, you may adjust a value within the 4 P's of marketing, or marketing mix. These consist of product, price, place, and promotion. For example, you may run an experiment in which you compare two prices for the same product, to see whether one price-point results in higher overall revenues compared to the other.Effects - Are the results of the test known as the dependent variable. Results are measured and cannot be changed. If the tester wants to see different results they would have to change the conditions of the independent variable to measure the effects.To gain an accurate result from experiments, the experimenter must consider outside factors that could affect the dependent variable. Continuing from the advertising example above; did sales increase because of a festive seasons at that particular time.
Applied application
In marketing email marketing is popular it is more targeted and there are many testable features such as font, colour and pictures. By tailoring emails it gives the experimenter control over conditions. For example; promoting discounts to experimental subjects via email.It is here the experimenter has control and sets conditions by sending one tailored email to one age group. The experimenter will send another tailored email to an older age group offering the same promotion with a different presentation. The experiment so far is exposing group A and B to different advertising messages.
There are many measurable outcomes, a few are as follows:
- How much time is spent on the site from the URL link in an email
- Better measurement in purchase process than other mediums - Subject A and B using promotion codes from the email
- Users unsubscribing
A/B Testing and Multivariate Testing
Two ways to perform a marketing experiment are A/B testing (also known as split testing), and multivariate testing. In an A/B test, you compare two versions of marketing collateral against each other and determine which one performs better. For example, you might compare the open rates for two email subject lines. Usually, the test will be performed on a subset of your entire population, and the results are then used to make a decision on how to market to the rest of the population. In the email example, you might test two subject lines on 20% of your contact database. If one subject line results in significantly more people opening the email than the other, then that subject line is the winner of the test. The next logical step is to use that subject line to send the remaining email messages to the rest of the contacts in your database.In a multivariate test, you would run an experiment that produces differences in more than one variable on the desired outcome. For example, you might run a multivariate test on a website landing page, in order to determine which version produces the highest percentage of people landing on the page who complete an email registration form. Some of the variables that you include could be two versions of headlines, three versions of page content, and four versions of the call-to-action that compels users to complete the form. In this case, the total number of combinations would be 2x3x4 = 24 total combinations of headlines, content, and calls-to-action. The winner of the test is the best combination of all of the variables. Because of the nature of a multivariate test, it requires much more data than a simpler A/B test for statistical significance.