Hi all,
Welcome to the May version of the monthly MDN report. For more information on what we are up to on MDN, you can visit the New on MDN Twitter feed to see some of the content being added to the site by the MDN contributors and paid staff. You can also refer to the MDN Roadmap for quarterly deliverables and initiatives. And, if you want to geek out on what our talented dev team is doing, you can see what development is in process on our kanban board or within github , as well as a feed of what's been released on MDN lately.
We had a busy month in May & are expecting the same in June!
May Highlights:
* N ew homepage and global navigation content launched on May 13! While the overall site design is the same, the home page layout was changed, new links to make Apps & Firefox Os more findable were added, and navigation made more action-oriented.
* Continued chugging along on Apps, Web API & Firefox OS documentation.
* A/B testing infrastructure - we now can do A/B testing using Optimizely . Our first test was a pilot "Page Actions" experiment with +23% conversions to 'Edit Page' from the experiment.
* Added the ability to automatically generate complex lists of content & landing pages, which will help us keep the site up to date as new docs are added. An example is the new Web API reference ; a listing that shows all of the current API documentation available to developers on MDN.
* Google Summer of Code started - Berker Peksağ & Gabriel Ivanica will be working on MDN GSOC projects this summer. Berker, a current MDN contributor, will be working on a Localization Dashboard , which will help our translation community see which localized pages most need updating. Gabriel will be working on MDN CSS Generation Tools, which are visual and interactive tools that a llow users to generate CSS with complex effects on MDN. We're very excited to have these project underway!
* Mozilla Hacks was moved to a continuous delivery model, which will greatly increase our ability to make timely feature improvements.
* Finished backend development for Elastic Search
* Launched MDN-drivers , which will help the team prioritize work on the MDN platform. Drivers is comprised of the MDN community, as well as representatives from the engineering & product teams.
* External documentation proof of concept - a way to see docs on github &
readthedocs.org as if they were native MDN content. This is the first, experimental step of a Q3 project that will allow us to source documentation where it is being written by developers (especially for new or experimental code & engineering/product documentation)
Planned for June:
* Site Redesign -
* Card Sorting link being distributed to developers via the MDN community & Mozilla Reps - the sorting will help drive how we organize the site going forward
* UX will have a proposal for site architecture for review in mid-June, and then spend a couple of weeks getting feedback and making revisions. Once we signoff, wireframes will be handed off to the creative team. The architecture reviews will be open to the community as well as the MDN drivers team.
*
Elastic Search implementation (replacing internal Google search) - in process of designing a new results page, with results filtering to follow in Q3.
* Enhanced user (spammer) banning capabilities
* Embed YouTube videos
* Syntax Highlighting
* Community-organized, public Doc Sprints in Paris & San Francisco on June 1 .
* Help with the repack & update all Jetpack Add-ons documentation for Firefox 22 release on June 25
Metrics
Unique Visitors barely budged from 2,115,515 in April to 2,136,434 in May
Pageviews increased 2.14%, to 6,397,547 in May vs 6,263,307 in April
Bounce rate: 72.60%
Bounce rate represents the percentage of visitors who enter the site and "bounce" (leave the site) rather than continue on & viewing other pages. At first look, MDN has a pretty high bounce rate, on average 72.60% of visitors leave after looking at only one page. The most quoted average across the web is something around 40-50%. However, bounce rate alone doesn't really tell a complete story. The bulk of MDN users tend to search for a MDN page in Google (literally, they type in MDN + a term or element), go to the page, read it, and then leave. According to Neilson, t he average page visit lasts a little less than a minute, with most people landing on a page deciding to leave in 30 seconds or less. On MDN, the average time spent on a page is 02:35 minutes (and in many pages, it is 4-5 minutes). So, for this specific use case, MDN is actually doing quite well. We're working on addressing other use cases where we do less well - like with visitors who come to the home page and want to browse for content, as part of the redesign work we are doing and ongoing improvements and additions to the content of the site.
Apps content (not counting DevHub pages):
Pageviews increased to 49,681in May from 40,583 in April. Most of the increase came after the incremental changes on May 13, so hopefully next month will continue upward.
Bounce rate dropped 3.31%, from 48.87% vs 50.54% (the average bounce rate for MDN is 72.60%)
Average time spent on Apps docs pages: 2:15 minutes for all visitors, but jumps to 4:02 minutes for visitos who go directly to an Apps page from search results.
Firefox OS docs:
Pageviews: increased 14.75%, from 136,748 vs 119,168
Bounce rate: dropped 3.98% (49.82% vs 51.89%, also lower than the average MDN page)
Time spent on Firefox OS documentation pages averages 1:43 minutes for all visitors, including those who browsed to the content. For users who land on the Firefox OS pages, the average visit time jumps to 4:03 minutes.
Web API's
Pageviews: 418,786, with 238,884 visitors
Bounce rate: 73.93%
Average time on an API/DOM page: 2:17 minutes.
Let me know if you have any questions!
ali spivak
MDN Product Manager, Mozilla
408-859-8260
asp...@mozilla.com