9 Common A/B Testing Mistakes

Episode 258 | 9 Common A/B Testing Mistakes
Date: 2015-10-13
Link: Startups for the Rest of Us
Startups for the Rest of UsIf you’re going to start A/B testing really figure out where you can make the biggest impact on your sales funnel. Don’t just randomly test things on your website.
Have a methodology that you follow.
Your test needs to run long enough to be statistically significant.
Having one hundred conversions coming through (is a good length).
Split tests are not foolproof.
Most test you run are not going to show significant differences.
You don’t want to ignore small gains if they are statistically significant.
Run one test at a time until you feel comfortable with it.
It is hard to learn anything unless you have a goal in mind.
Be careful you aren’t accepting everything as fact.