A/B Testing is a process used in testing drugs to identify if a drug is effective or not.
A similar process is used to identify if a particular feature is effective or not. As you can see in the diagram, randomly selected half of the users are shown variation A and other half is shown variation B. We can clearly see that variation A is very effective because it is giving double conversions. This method is effective only if we have a significant amount of users. Also, the ratio of the users need not be 50-50.
So, A/B testing utilizes the data in product development. At Amazon, we have many such experiments running all the time. Every feature is launched via A/B testing. It is first shown to say 1 percent of users and if it is performing good, we increase the percentages.
To manage so many variations on such a high number of users, we generally need Big Data platforms.