Google juice
I find it such a pity that after so many years of work on MetaCPAN by so many contributors it still accounts to 27.3% only of the total CPAN traffic.
Maybe if more people were linking to pages of MetaCPAN from their own blogs in the relevant context.
For example if you write about databases and Perl you could link to the Perl DBI for Database Access. When you talk about AJAX and JSON, you could mention JSON and Perl. If you are talking about Dates or Timestamps, you could mention DateTime in Perl.
Or any other module that is relevant to your blog post.
I blog about Perl.
Agggghhh. Don't leave us hanging! What does constitute the bulk of the traffic?
I really like the search.cpan.org interface and I actually skip over the metacpan google links. Mostly because the font size is bigger and more readable and the content isn't squished into the middle of a 3-column layout. But also it's less cluttered and doesn't bug me to log in. What is the history behind metacpan? Why do all the contributors choose to work there instead of improving search.cpan.org?
I wonder if the premise of your post is incorrect. It seems like you are saying that the traffic is low(er) because people don't know about it, and that if we all improve the google rankings more people would use it. But maybe the other 70% of traffic just prefers search.cpan.org or something else?
search.cpan.org is closed source. It has not been updated for about 10 years. (AFAIK There were some minor tweaks.) MetaCPAN is Open Source.
Most of the traffic to search.cpan.org comes from search Engines.
See some CPAN traffic numbers here