Comments on: Recap: A More Dynamic Experience

By: Improving Homepage Experience

Improving Homepage Experience — Mon, 08 Feb 2010 06:32:06 +0000

[...] Results: As a result of this data-driven design, bounce rates to these key pages improved between 4% and 18%, and download conversion (the number of people that landed on the page that downloaded Firefox) improved. Creating dynamic content based on browser type is now implemented permanently on the Mozilla.com website. Read more here. [...]

By: about:mozilla - Mozilla服务周、诗性+实用主义、AMO、Personas等更多消息 < MJiA

about:mozilla - Mozilla服务周、诗性+实用主义、AMO、Personas等更多消息 < MJiA — Wed, 06 May 2009 03:26:15 +0000

[...] 市场和度量团队最近在努力提升Firefox下载页面的用户使用体验并希望能够提升访问/下载比率。在实验和测试了不同的设计之后，他们在en-US语言版本的页面中对浏览器进行检测，这样会根据用户是否使用Firefox、使用什么版本的Firefox等因素来提供不同的内容。“现在看来提供给访问者不同的内容是可行的！”Ken Kovash 和 David Rolnitzky最近都在博客中讨论了这个想法。 [...]

By: Improvements Have Arrived at Firefox.com < Blog of Metrics

Improvements Have Arrived at Firefox.com < Blog of Metrics — Fri, 24 Apr 2009 17:22:59 +0000

[...] different set of content to existing Firefox users seems intuitive, right? Based on David’s findings, visitors to the test pages agreed. The download conversion rate of IE visitors increased [...]

By: DR

DR — Fri, 24 Apr 2009 15:45:07 +0000

@ Axel – good question. Since this was a first test done in en-US, we don’t have data for other languages. While some of the copy may not be a perfect fit for translation, there are some good lessons that can be applied from a structural and contextual perspective (e.g., “switch” content, adding more info about customization, making “tips and tricks” content more prominent, etc.).

By: David Naylor

David Naylor — Fri, 24 Apr 2009 07:20:50 +0000

A thought about the IE page:

The link “Make the switch today!” is a bit ambiguous.

Something like “See more reasons to switch here!” is a better description and helps the user decide what to do next.

By: Axel Hecht

Axel Hecht — Fri, 24 Apr 2009 05:29:23 +0000

From an international perspective, I could see that phrases like “So 2006″ are great for the US and Canada, but for non-native-language folks ending up there and not on the tailored landing pages, it might be an idiom too far out of what folks learn (or not) in school, and thus confusing.

Which is a looooong sentence and possibly the question, do we have the data to see how those experiments worked across languages?

By: DR

DR — Fri, 24 Apr 2009 04:47:36 +0000

Great point Jesse. We’re looking at multi-variate capabilities that will allow us to do tests exactly in the way you describe but in a scalable way. In the short-term, we’ll continue to do this type of A-B testing.

By: Jesse Ruderman

Jesse Ruderman — Fri, 24 Apr 2009 01:31:29 +0000

There are a lot of differences between the two pages: color/background, size of the download box, presence of awards, presence of “switching tips”, and positive vs IE-antagonistic text. There could even be speed differences. Wouldn’t it be better to test all those changes independently, so you know which ones actually help?