A/B Testing
by JunMing Zeng
1. Do’s
1.1. Always test both versions simultaneously.
1.2. Always split traffic between two versions.
1.3. Know how long to run a test before giving up.
1.4. Show repeat visitors the same variations.
1.5. Make your A/B test consistent across the whole website.
1.6. Do many A/B tests.
2. Don'ts
2.1. Don’t surprise regular visitors.
2.2. Don’t conclude too early.
2.3. Don’t let your gut feeling overrule test results
2.4. Don’t reject the results because of your arbitrary judgment.
3. Pros & Cons of A/B Testing
3.1. Pros
3.1.1. 1. Fast - matter of splitting traffic to the two pages
3.1.2. 2. Tests reality - obtaining real results from real users doing real things
3.1.3. 4. Accurate - assuming you have statistically significant data.
3.1.4. 3. Quantifiable - provides actual numbers
3.2. Cons
3.2.1. 1. Can Hurt Web Site Results - make some bad decisions from time to time
3.2.2. 2. Missing the “Why” - you won’t know for sure
3.2.3. 3. Not Predictive -can’t be used to predict future design change impacts
3.2.4. 4. Needs Traffic -need a pretty good amount of traffic