6 ms·
http://imgur.com/cMshOJx http://imgur.com/cMshOJx Screenshot for anyone who isn't seeing the new page.
by jkimmel 12y ago
http://imgur.com/cMshOJx http://imgur.com/cMshOJx
Screenshot for anyone who isn't seeing the new page.
- declan 12y agoYep, I'm seeing the same redesigned page. I get the new page with Chrome (OS X) both logged-in and incognito, and also with Safari, but I see the old page in Firefox.
- junto 12y agoThanks for this. I think that access to the new design depends on your geo-location. Even using Amazon.com I still get the old site from Germany.
- yaeger 12y agoNo, I'm also in Germany, got the new site right away. Opened amazon.de in another tab to compare the changes made. Now we'll see how long it takes until amazon.de changes as well.
- frik 12y agoAmazon.de often lags behind one or two versions.
- hobofan 12y agoAmazon uses cookies to track which site design you have seen last. They are constantly redesigning their pages (I've seen up to 4 completely different versions of a page in use at the same time). They have to lock the design through cookies (otherwise you would be seeing a different design on every refresh).
- taoufix 12y agoYou're right. I switched to incognito mode and got the new design.
- gibsonje 12y agoNot sure about this. I'm on the same network as my boyfriend. Yesterday he pointed out that amazon.com looked different. I'm on my laptop and it's not showing the new design.
- reduce 12y agoWorks if I open Amazon in a new Chrome incognito window.
- JOnAgain 12y agoFor those who are curious, Amazon uses a backend service to control what you see. They randomly hash your customer ID (which you can never see) and then assign you to a bucket. It's a sophisticated A/B testing framework that then allows them to measure and compare pretty much everything about the two buckets of people. They can allocate people between A/B pretty dynamically. Most rollouts start with 5% of people and move up from there. Since it's tied to your customer ID, they can retroactively go look at any other variables they didn't think to look at initially just by knowing whether you were an A or a B and the times when you were in which. It's also important to be a "random" hash because you have hundreds or thousands of these experiments going on at the same time, and you want to rule out spurious relationships (e.g. other experiments being a confounding factor). What metric do they use to evaluate the success of an A/B test? Generally straight revenue, though sometimes average order size or category spend will enter into the mix, too.
- pgrote 12y agoThank you for this. I actually miss the boxes down the right side.