Respond to Results
Once your experiment has finished, you can replace your original page with the winning variation, or run an additional experiment.
You might also want to consider setting up redirects from your variation URLs to the page that is your final choice. If users have bookmarked any of your variation pages and you subsequently remove those, you want to be sure those bookmarks lead to your new landing page. (Learn more about redirects and SEO ranking.)
Replace Your Original Page with the Winning Variation
When your experiment ends, users then see only the original page; variation pages are not displayed.
If your experiment demonstrated that one of your variations performed better than your original page, one option is to replace the original with the better-performing variation. To replace one page with another, you need to change the structure of your website--Analytics does not provide any method of exchanging pages.
Run Another Experiment
If your experiment does not produce a page that is a clear winner, you can revise your experiment and launch a new one. For example, if you weren’t seeing pronounced differences in the performance of your pages, you can run another experiment in which you test more dramatic differences between the pages.
If your experiment does produce a clear winner, you might want to run an experiment on other pages of your website, copying the changes from your first experiment. For example, if a change in the color scheme of a landing page helps drive more conversions, try changing the color scheme for the landing pages in other goal paths.
If you run an additional experiment, remember to remove the old experiment code from your original page and replace it with the code for your new experiment.