Hi, we are wanting to test a new version of one particular page on our website. What's the best way to run a test for a week without harming rankings. Do we use a 302 redirect or direct 100% of traffic using a tool like Optimizely?

[Mike_The_Coffee_Man]

Question about the best way to do a temporary redirect

[WebElaine]

I personally play the Google game - use Google's own products where possible. Google happens to have Google Optimize, which will let you run all sorts of A/B tests pretty easily. Benefits: Google itself knows you're running an A/B test so your rankings shouldn't fluctuate; you get access to any Google Analytics goals and events, in case you want to set up your own custom metrics.

I've also successfully used tools like Optimizely and VWO. Most do a pretty good job of ensuring that spiders just see the original while human visitors are split. It depends on how specifically you set the test up. If you send all the traffic to a single URL and JS does the magic, you're safe. It's when you code an entirely separate codebase and then split the traffic between two URLs that it can start to splinter traffic and adjust rankings. You can still do that safely; you just need to make sure the "B" (and C, D, ... N) version(s) have a canonical set to the original page.