10 Important Guideline for Running Your A/B Test
A/B testing is important to your company to make changes. A/B testing allows you to test one variable at a time. This gives you the results and allows you to pick the better of the two variables. A/B is used to make changes to your website, blogs, products, and emails. Deciding on what variable you want to test, leads to many questions. What should you test? How long should you run the test? Before running your A/B test follow these guidelines to help measure your results and what variations perform better.
1. Only conduct one A/B test at a time.
Never run more than one A/B test at a time. Multiple test running simultaneously results in confusing results. Testing an email campaign at the same time you test your landing page gives conflicting results. How can you tell which results caused the increase in your site visits?
2. Only test one variable at a time.
To help evaluate the change in your home page, call-to-action or email campaign, select one element to test. Isolate one variable on your home page to run your A/B test on. Select either a web page, email, or call to action as your variable.
3. Always test minor changes when using A/B testing.
Don’t forget the small changes, they are just as import as the big ones. By testing something as simple as your subscribe button color can make all the difference. Remember small changes can result in big improvements. Small changes are easier to measure than big changes.
4. Use A/B testing to test the entire element.
With A/B testing you can test a button color change, background color, landing page, call to action, and email as your variable. Why not consider redesigning your landing page giving it a whole new look? Create a new landing page and test it against the one you use now. Consider starting with this test before you start making smaller changes to your site.
5. Use A/B testing to measure, as far down the funnel, as possible.
After testing your landing pages and making changes has this improved your sales? A/B testing has a significant effect on your bottom line. When creating your A/B test, consider how it affects your, click through rates, leads, traffic to lead conversion, and demo request.
6. Use control and treatment when A/B testing.
With A/B testing always keep a copy of your original page before making changes. Create a new alternative version of your page as a control. After, build your new variations or treatment page for testing against your control page.
7. Deciding what you want to test.
Look for different elements in your marketing resources, for design, page layout, and wording. Certain areas for testing you don’t recognize immediately. Consider testing different targeted audiences, timing, alignments between your email and landing pages.
8. Randomly spit your sample groups you test.
To receive conclusive results, use your A/B test on two or more audiences that are equal. When testing your email listing, list your sources, list type, and length of time one name has been on your email list. When testing two email lists, keep all other aspects of your design identical. This gives you clean results based on your lists and nothing else.
9. Always run your A/B test at the same time.
When conducting your test, run your tests together. Never run one test and a month later run your second test. Running your tests at different times gives inclusive results.
10. Decide on your necessary significance before testing.
Before running your test decides what significant results, you’re looking for. Set a significance goal for your test before starting. Normally a result of 97-99% statistical significance is a good percentage to aim for.
About Sarah Johnston
Sarah is an IT consultant with a degree in marketing. She enjoys helping others to learn the ins and outs of Internet, Website building and blogging.