A/B Testing
What is ?…and…How to?
What is A/B Testing?
A/B testing is running the simultaneous experiment between two or more variants of a page or a design to assess which performs the best. Essentially A/B testing is about making decisions based on data about how people actually behave when they use the testing page.
A/b testing can be valuable because different audiences behave, well, differently. Something that works for one company may not necessarily work for another.
Almost any content and settings can be tested. web pages and their elements, ads; management strategies and approaches, emails and mailing list items.
A/B testing is a way to test hypotheses and it is A/B testing that allows one to confirm whether the hypotheses will work out well or not.
It doesn’t necessarily solve all the problems of business. It helps in making the right decisions. For example If the product needs a complete revamp, then all the A/B testing in the world can’t save it.
How to start a test ??…
Before starting the A/B testing plan, a thorough research on how the product is currently performing. Data has to be collected like how many users are coming onto site, which page drives the most traffic, etc.
he tools used here can include quantitative website analytics tools such as Google Analytics, Omniture, Mixpanel, etc., which can help you figure out your most visited pages, pages with most time spent or pages with the highest bounce rate.qualitative insights can be derived from session recording tools that collect data on visitor behavior, which helps in identifying gaps in the user journey. In fact, session recording tools combined with form analysis surveys can uncover insights on why users may not be filling your form.
1. Formulate a hypothesis
Before starting the test, deciding on the assumption. They should not be invented. Hypothesis should be built based on web analytics and analysis of visitors.
An important rule is — that in one experiment only one change can be tested.
2. Identify targets
After deciding on a hypothesis, criteria for the specific test need to be selected.
For example bounce rate (how many people visited the page and immediately gone), the time spent on the service resources, the number of applications or registrations, the number of purchases or the average check.
3. Select one test item
This is the one with which hypothesis is tested. For example Calls to action (length, content, location); Buttons (color, size, location, text); Images (size, content, location); Text on page (length, content); etc.
4. Determine the Test sample size
To do the testing in advance the one needs to calculate how many people need to visit the page so that the test results are statistically significant.
The stronger these changes, the fewer people will be required to sample. And you can calculate it for example with the help of online calculators: Optimizely, AB Tasty, Unbounce, AB Test Guide and others.
5. Determine the duration of the experiment
The test should last at least a week, even if you scored the right number of visits in two days. This is necessary because users behave differently on different days of the week.
6. Do A/A testing
Absolutely two identical pages are shown to all users and track the results. If they match, then the traffic is uniform and the results of A/B testing will be reliable. If the results on the two pages are very different from another, then it doesn’t make sense to conduct the test.
7. Do A/B Testing
After you have formulated a hypothesis, chosen goals, figured out the duration and sample, and were convinced of the uniformity of traffic, you can proceed to the main test.
Source :
https://unbounce.com/landing-page-articles/what-is-ab-testing/
https://uxdesign.cc/7-steps-of-a-b-testing-what-how-cf3b209467fd
