How Not To Run An A/B Test
http://www.evanmiller.org/how-not-to-run-an-ab-test.html

How Not To Run An A/B Test

Interesting argument (with statistical rationalisation) explaining why A/B tests should be run to a fixed number of observations. Advises against pushing a winner too quickly.

A good article which suggests not ending an A/B test too soon just because it has reached 95% confidence.

A simple javascript multivariate testing framework

Cohorts is a simple, purely javascript, multivariate testing framework. It allows you to easily run split tests for visitors on your site, showing them different designs, layouts, or whatever you want. Cohorts also allows you to track interesting events that occur for each of the cohorts. By default, it uses Google Analytics event tracking to store data, but you can customize it to use your own or another.